Sometimes you want to write an article about messing with a spreadsheet, today is one of those days.
DW-NOMINATE is an ideological classification system for the United States Congress that works by mapping voting records onto a sort of spatial map, and grouping those who vote more like each other closer together. The map is two dimensional, and the first dimension produces a relatively clear left-right continuum, the second dimension is weird. In the past it was essentially the racism-o-meter. The vast majority of Democrats were left of center on economic issues, but the southern Democrats of yore were staunch segregationists, thus this second dimension could work to separate a Hubert Humphrey from a J. William Fulbright, both similarly liberal on economic redistribution and Fulbright even more left wing on foreign policy, but one a civil rights activist and the other a staunch segregationist. Now that second dimension serves as like this vague sense of anti-establishment views. For instance, The Squad don’t score all that left wing on the first dimension because of how many Democratic bills they shoot down for being insufficiently left wing, but it pushes them way down on the second dimension.
Anyway with that boilerplate out of the way, I wanted to look through the 115th Congress (Trump’s first), and look at some super low probability Senate (simply because I’m more likely to recognize a name) votes during that time:
Kamala Harris voted to confirm Trump’s CIA General Counsel nominee.
John Kennedy, again, voted in favor of some Jeff Merkley amendment on the TCJA back in 2017, I assume it made it less regressive, but I’d genuinely have to watch a lot of C-SPAN to figure out precisely why this was.
John Kennedy again this time voting down a Trump judicial nominee for the DC Circuit Court.
Brian Schatz, a liberal Democrat, voted for Trump assistant Interior Secretary nominee.
Rand Paul voted against the FY2019 NDAA because of course he did.
I might go deeper into this later, but I just wanted to check this out a bit, not much else to write about today.
Monday: Congress is out of session so I really don’t have much to write about over the next few days, but next week will feature deep dives into Biden proposals and cabinet nominees, starting with a general overview of what the first 100 days of the Biden administration could look like on Monday.
You might not have heard about this, but there were some federal elections held in the United States a week and a half ago. 572 of them in fact, 51 of them for President, 51 for Vice President, 35 for the Senate, and 435 for the House of Representatives. Now, all of those races except a few in the House have been resolved, regardless of what President Trump has to say. While I don’t think it’s that worthwhile to make too many super confident takes this early on, I did make a prediction of the Presidential map, and I think it’s worth interrogating how reality differed from that map, because I think it tells a fairly noticeable story.
The Trends That Do Persist
One thing that I try to keep in mind when it comes to political analysis is that pundits tend to overestimate the amount to which new things that happen are the products of long-term trends over things that will just revert to the mean. The support of Donald Trump among Midwestern non-college-educated whites is one of those things. I figured it wouldn’t go back to Obama 2012 levels or anything, but I was seeing poll results that told me that it would get far closer to that than what actually seems to have ended up happening based on county-level results (if you try to make these inferences based purely on exit polls the 538 Politics newsroom will come to your house and shoot you). Here was my ultimate predicted electoral map:
Might as well put out my final election prediction so I can be made fun of for being wrong in 24 hours pic.twitter.com/BqeZLzP7jw
I got Florida, Georgia, and North Carolina wrong, plus ME-02’s isolated electoral vote. You will note how similar my predicted map was to 2012, I really believed that Biden was largely recreating the Obama coalition. Instead it appears that he just leaned in really hard to the college-educated whites niche. That is enough to make up a majority of the nation, but it’s incredibly electorally inefficient. The median state for Senate purposes here is probably Georgia or North Carolina, which are maybe R + 5 or + 6 compared to the median American voter, and the staggered nature of American elections means that Democrats need to average that over a six year group of three election cycles to take the Senate, it’s rough.
All that being said, Joe Biden will become the President on January 20th, and that is a tremendous relief after the last four years, pessimism about Congressional Democrats aside. I will likely be writing about actual things that, like, the United States government will actually do instead of just having the President whining on Twitter all day, and that will be fun.
Also, stay tuned for 2022 projection stuff, I think it’s a reasonable goal of mine to be able to publish a probabilistic forecast for the Senate by the time that election rolls around, and some of the work I’ve done on slowly going over Senate results and polling error will build to that.
Brief note: Most of this article was written in February and early March, in the sweet before-times when the world wasn’t on fire and I assumed there would actually be a regularly scheduled MLB season. If I don’t acknowledge that in the article that is the reason. Take this as a sort of nice insert from the parallel universe in which the last two months or so never happened.
Introduction
I am not a Washington Nationals fan, I’m not even a fan of a National League team, but fandom doesn’t matter when wanting to take a retrospective look at some greatness. I have been a regular watcher of Major League Baseball for five years now, since 2015. And in that time one pitcher has stood above his peers with a peak for the history books, Max Scherzer. All of his seasons in this time with the Washington Nationals, two NL Cy Young Awards, a lot of down-ballot NL Most Valuable Player votes, $125M paid in salary (kind of, his contract has a lot of deferrals), a 2.74 Earned Run Average, an MLB leading 34.9 Baseball-Reference Wins Above Replacement, and an MLB leading 32.6 Fangraphs WAR. He has been the most valuable pitcher in baseball over that time by most reasonable measures. That’s fun to talk about on its face, but I want to go deeper here, let’s look at Scherzer’s general style and pitch mix, and then go over one of his years during this span. This is a bit of a crazy project, but I’ve ended up in a situation where I have a lot more time on my hands now (this wasn’t initially written about COVID, but it now I guess also works because of COVID), so I can handle this.
Scherzer: The Man
It’s worth starting off with detailing the person this article discusses before going into him purely as a pitcher, personal aesthetics is a large part of sports, and Scherzer’s are great.
This is Mad Max, a heterochromatic scraggly bearded man in his 30s who foams at the mouth and screams while he uncorks fireballs. Scherzer isn’t particularly tall compared to other pitchers of his caliber, he’s 6’3″ while Jacob deGrom, Clayton Kershaw, and Justin Verlander are all taller. But the amount he leans into the Mad Max persona has to create a level of intimidation for batters that puts anyone at least a bit off-kilter even without the pure size. He is by all means a nice person, but once his pregame routine starts up, he’s in the zone, and he will tear the arm off of anybody who wants to tear him from the zone. Quoting this great article from Joe Posnanski:
Another awe-inspiring thing is the rage and focus he brings to every start, every inning, every batter. There are countless stories about the mask of rage that comes over him when he’s on the mound. When one young player offered him a high-five before a game, Scherzer walked right by him. “I don’t do that,” he later explained. “I’m in the zone.”
I’m not usually one to get caught up in the emotion of watching players, MLB is largely an exercise in a sort of aesthetically pleasing mathematics to me, but Scherzer is an exception. This is a man who once pitched with a broken nose. His intensity is enough for me to even let out a “hell yeah” or two when he’s giving it his all, which he rarely isn’t.
Scherzer: The Pitcher
Let’s get down to the actual baseball, Scherzer is a right-handed starting pitcher. Early in his career Scherzer experimented with a variety of pitch arsenals. But upon his arrival in DC at the start of 2015, he had finally decided on a five-pitch loadout:
A four-seam fastball thrown about half the time over this period. Averaging about 95 MPH. With a great but not quite elite 2500 RPM spin rate. It’s got a bit of sinking action, but that’s more a function of a lower release point than most of his other pitches than anything. It’s just not a pitch that moves a lot, and because of the velocity and backspin, that’s enough to surprise batters with the rising fastball effect. It generally ends up close to the middle, a little above the middle perhaps, of the zone when he gets one where he wants it. The stuff is good enough to make that work. It’s his bread and butter for a reason.
Then Scherzer’s main off-speed pitch. A 86 MPH slider thrown almost exclusively to right-handed hitters at a 2350 RPM spin rate. It’s his most common put-away pitch, going equal parts down and away from right-handed batters, and they’ve generally done the worst against the slider of any of his pitches. One might think that because he mostly uses it as a put-away pitch against righties that it would be predictable, but because Scherzer has that low fastball release point, he mixes them up well enough so that a batter won’t clearly recognize which is which until about 150 milliseconds before impact. This is his best pitch, with a whiff rate 29% over league average for sliders over this time-frame.
A changeup thrown mostly but not entirely to lefties. 85 MPH, 1500 RPM. Mixed really well with the slider in terms of release point and path, but suddenly takes a sharp turn outside (from a lefty’s perspective) and hits the ground immediately. Because of how often he throws those pitches to each handedness, it can be a bit predictable which of the two he’s going to throw, but the fastball remains a constant, and I imagine the reason he still throws the changeup to righties, and very occasionally the slider to lefties just to keep them on their toes.
A curveball used as a secondary off-speed pitch almost exclusively to lefties, it’s his only pitch that has an obvious tell, he has a significantly higher release point for it, and I don’t think it’s a coincidence that this pitch has his worst outcomes. Its hit for home runs more often than any other pitch he throws and has the lowest whiff rate. It’s probably best used as a sort of distraction from the slider and change more than what it does in itself. 79 MPH, and a spin rate that has increased notably over this period, starting at about 2200 and making its way up to 2700 in 2019.
Finally, a cutter, something he added to start 2015. Thrown almost exclusively to lefties, and only broken out for special occasions as a put-away pitch. 90 MPH, 2400 RPM, and a lot of horizontal movement to the inside against a lefty, disguised well as what could just be a normal fastball before that point.
With that overview out of the way, we can proceed to the retrospective. This will cover just the first year of the five, as covering all five would be close to a book length affair, but I might revisit this topic for times in the future.
2015
On January 19th, 2015, this happened.
Source: Scherzer deal with #Nationals is seven years, $210M with half deferred. He will receive $15M per year for 14 years.
This was, at the time, the second largest contract given out to a Major League pitcher, surpassed only by Clayton Kershaw’s $215M extension given in the previous off-season. Judging by the tweets I had to sift through to find this Rosenthal tweet, the hot take manufacturers assumed this was a boneheaded decision by the Nats. They were obviously wrong, this is probably the best production anybody has done on a $200M+ contract.^1 A lot of the contract was deferred as well, making the present value of it more like $190M. On April 6th of that year, he made his Nationals debut. Let’s get into it.
Game 1: 2015-04-06 vs New York Mets
Scherzer had a very fastball heavy debut. 60 of the 97 pitches he threw were fastballs, one of which hit 98 (and perhaps not coincidentally was wild and ended up being hit for a single). He was mostly a three pitch pitcher here, fastball-slider-changeup. The slider was on when he used it, generating just as many whiffs as the fastball. Still though, this was mostly a game of fastball dominance. He struck out eight, walked two (both of which were cases of too many fastballs getting away from him), and while he allowed three runs, none of them were earned. The Nationals’ defense made a lot of bad errors, including a dropped infield fly. Once he got to the fourth time through the batting order he was showing legit weakness, he was taken out after 7.2 innings, and was handed the loss thanks to run support made up entirely of a solo home run from Bryce Harper starting his campaign that would be the best hitting season this side of Barry Bonds. Still, he was good and had a 0.00 ERA almost purely through fastball pounding.
Game 2: 2015-04-12 @ Philadelphia Phillies
This was a very similar game to Scherzer’s first. eight strikeouts, two walks, heavy fastball usage (70 out of 102, to be precise). The fastball accounted for four of those strikeouts, with two changeups, a cutter, and a curveball being used to get the others. He also broke out the cutter for the first time here, with the addition of added curveball usage, so when he wasn’t throwing a fastball it was unpredictable what was going to be given. It wasn’t as good as Scherzer’s previous start, with a hanging changeup that was punished by Freddy Galvis for a line drive single, and a hanging slider that was taken for a high strike instead of punished like it could’ve been. Nevertheless, Scherzer got out of things with only one earned run, and he never allowed anything more than a single. He was saddled with the no decision this time thanks to Nats bullpen woes, but they ended up winning the game in the end.
Game 3: 2015-04-17 vs Philadelphia Phillies
Scherzer’s best start yet. Despite him allowing an earned run in this one while he didn’t in his first, but hits vs errors are a weird and subjective judgement and you can tell watching him here that he was on his A-game. Scherzer dialed back the four-seamer usage a bit to let his changeup and curveball star, and the changeup and curveball were on-point. 4 Ks from the fastball, an additional 3 on the slider, and two on the change. Eight innings pitched with those 9 Ks, and no walks. This was great fun to watch, and 2015 Harper being 2015 Harper gave the Nats the easy win and Scherzer himself his first on the year.
Game 4: 2015-04-23 vs St. Louis Cardinals
Not a great game by Scherzer’s standards. He only got 4 Ks in seven innings, while he didn’t walk anybody he did hit Matt Holliday with a pitch, and held onto a changeup for too long for a wild pitch that allowed the first run to score. Only two runs allowed, but Scherzer’s defense, particularly a couple nice running catches from Harper, saved him from what might have happened otherwise. These Cardinals were a 100 win team, so it’s understandable having the first meh game of the year there. As for the specifics of his pitching, he was still mostly relying on the fastball, but the Cardinals clearly got word of him starting the season like that and were swinging pretty freely. As such, the only pitch of his that really felt at 100% was the slider, of which he only threw eight. Scherzer got saddled with the loss, and he had to leave the game after only 82 pitches due to him hurting his thumb while batting. It wasn’t an out and out bad game, but perhaps a sign of a need for some adjustments that have Scherzer rely on that great slider a bit more.
Game 5: 2015-05-01 @ New York Mets
Alternatively Scherzer could just supercharge his fastball and blow it by everyone for a ten strikeout game (eight of which were fastballs). Scherzer had a couple extra days rest thanks to the aforementioned thumb problem, and that may have given him enough in the tank to let the heaters fly. This one had a couple rough moments, the fastball-pounding let the ball get away from him for a walk and hit-by-pitch. Scherzer also gave up his first home run of the year here, but the pitch wasn’t an especially egregious one, it was a 95 MPH fastball on the outside edge as pictured. Just got turned around for a good hit. Otherwise this was just blowing fastballs averaging 94.5 MPH by a bunch of decent Mets hitters. Unfortunately the reason this Mets team won an NL pennant was mostly because of their pitching, and Matt Harvey (lol) powered the Mets to a victory and gave Scherzer the loss, making him a pitcher with a 1.26 ERA and a 1-3 Win-Loss record at this point in the year, if you want some evidence of W-L record’s shortcomings at being anything other than a team stat.
Game 6: 2015-05-06 vs Miami Marlins
This one wasn’t as bad as the stats look. Scherzer managed to convince manager Matt Williams to leave him in for an eighth inning after being at 101 pitches. He promptly gave up two singles and a home run to Giancarlo Stanton on a genuinely shitty pitch. 86 MPH slider right up there over the plate that was just crushed. If you take that inning out though, this was a 10 K, 0 BB performance that did allow two runs, but only because of bad sequencing on some back to back singles. Seven of these Ks were on on fastballs, three on sliders. This is representative of Scherzer largely being a two pitch pitcher here. His slider was great generating eight whiffs on 27 throws. Scherzer still got the win here despite allowing five runs, Harper had a three homer game. Scherzer’s shortcomings here were largely a symptom of a managerial style that’s largely dead, letting the starter go well past 100 pitches and into the fourth time through the order while he’s already allowed some runs. That wouldn’t happen in a regular season game today.
Game 7: 2015-05-11 @ Arizona Diamondbacks
Scherzer was pretty good, if in a more pitching-to-weak-contact way than his usual strikeout-or-flyball self. A lot of easy groundouts, 6 Ks (most of them on sliders), 1 BB, and one genuine mistake in a solo home run given up from a hanging slider thrown in basically the same place he threw it to Stanton last game. 94 pitches to get through seven innings scoreless outside of one mistake pitch, that’s pretty good. In terms of actually winning, this was a case where the Nats offense took it away, pullin out to a 10-0 lead after the top of the second. Scherzer has gotten the run support he didn’t have in most of the first games and is now 3-3.
Game 8: 2015-05-16 @ San Diego Padres
Scherzer’s best start yet, seven scoreless innings with 11 Ks. He had a shaky fourth inning, allowing a walk and a hard hit single, but was otherwise totally cruising. Scherzer did something he hadn’t really done a lot before this point, he pitched backwards. ten of the eleven strikeouts were from a fastball, most 95+ MPH. There were a lot of cases where Scherzer started a plate appearance with a slider or a changeup, and then blew a high fastball by the batter for strike three. This is interesting to see in 2015, because it’s often the norm now, with high four-seam fastballs being prioritized over sinkers, but at this point in 2015, sinkers were still all the rage. Travis Sawchik published a book three days after this game about how the Pittsburgh Pirates were using a pitching staff that was all about groundballs into the shift, and then about two months later something happened (we’ll get to it) that ruined that whole thesis, poor guy. Either way, the Nats won this thing, and Scherzer went to 4-3 on the year.
Game 9: 2015-05-22 vs Philadelphia Phillies
This one wasn’t as good as the stats look. Scherzer did get through eight innings of one run ball, and his start was good, but not as good as that would suggest. Only 6 Ks is somewhat abnormal for a game he goes this deep into, but that’s not the only issue. That issue would be the number of flyballs and line drives he allowed, fourteen and six respectively. He got away with allowing a sub-.200 Batting Average on Balls In Play in this game, but that type of contact doesn’t exactly lead to that all the time. Still, what happened happened, and Scherzer walked out with a W to make him 5-3 with a 1.67 ERA and 72 Ks.
Game 10: 2015-05-27 @ Chicago Cubs
Seven innings, no runs, 13 Ks. I could stop it right there really, but let’s go into a bit more depth. Scherzer was unlucky in one way and lucky in another, so I think this result roughly reflects true talent. He was lucky in the sense that he allowed six baserunners, and none of them scored, but unlucky in the fact that he got six baserunners out of just thirteen balls hit into play + one walk. It’s probably worth focusing on those thirteen strikeouts and how he got them. His pitch mix was firing on all cylinders for those, seven strikeouts on fastballs, five on either of his breaking pitches (curve or slider), and one on the changeup. The man allowed some hard hit balls, but Peak Scherzer was on display for much of this game. It’s a hint of the pitcher that we’ll get to see in the Truly Great Starts he had this season.^2 The Nats and Scherzer walked away with 6-3 win over a pretty good 2015 Cubs team. Scherzer is now 6-3 with a 1.51 ERA, the lowest it would get for the rest of the year.
Game 11: 2015-06-02 vs Toronto Blue Jays
Home run #1
Home run #2
This is Scherzer’s first unambiguously bad start. Dissecting one of these is a bit of a different experience from doing the same to a good start, we need to focus on what went wrong instead of what went right. First thing is first, Scherzer was facing the 2015 Blue Jays, my personal favorite MLB team of all time. The 2015 Blue Jays were an outstanding hitting team. The only better hitting lineups this past decade were the 2017 and 2019 Astros, which were the best lineups since the Big Red Machine teams of the 1970s, even if you expand the window out to the entire 21st century so far, the Blue Jays have a higher Weighted Runs Created Plus than all but four other teams. It’s not that big of a deal to have a bad start to them of all teams. That being said, we can still pinpoint some things that went wrong here, of which I see two. First off is that Scherzer seemed gassed, his average fastball velocity was about a full MPH below where it was in the starts preceding and succeeding this one. This had Scherzer stay away from using the fastball as much as he usually does, and he relied a little too much on the slider as a backup, throwing over 30 of them. The second problem is Kevin Pillar, he had recently been called up to the Blue Jays as a super late draft pick who managed to beat the odds and be their starting center fielder, and he had Scherzer’s number. In his first plate appearance, Scherzer threw a 95 MPH fastball on the inside edge to bring a 3-1 count full, Pillar took note of this, and when Scherzer threw him a similar pitch that caught a bit more of the plate, he hit it 364 feet for a home run. Fast forward to Pillar’s second plate appearance, and he manages to get a line drive single off of a fastball thrown high and inside at him. At this point Scherzer realizes Pillar knows what he’s doing, and he starts pitching around Pillar with a bunch of sliders on their third meeting. Scherzer proceeds to miss his spot and hang an 85 MPH slider right where Pillar hit his first home run of the evening, and, well, you can guess what happens, 4-0 Blue Jays, Scherzer quickly retires the pitcher Marco Estrada for the final out of the sixth, and then leaves the game to get his fourth loss on the season. Now 6-4, 1.85 ERA.
Game 12: 2015-06-09 @ New York Yankees
Remember the start in Miami? Serious shades of that here. Scherzer seemed a bit shaky at parts, most notably allowing a solo home run to Stephen Drew, but that was just a 96 MPH fastball that caught a little too much of the plate. The real issue here was the fact that Scherzer was left in so long, he had thrown exactly 100 pitches going into the seventh, and he was kept in. Maybe not what I’d do, but I get it, it’s a tie game, your bullpen is shaky, and Scherzer is one of the five or so best pitchers on the planet right now. Scherzer gets a groundout. Then he allows a single, he’s thrown 108 pitches, this is at the point where I’d really pull him, manager Matt Williams doesn’t do that. He allows a second single now, there are men on first and second with nobody out, Scherzer is at 114 pitches, he is now going into his fourth time through the heart of the batting order, he should be pulled, Matt Williams keeps him in. Scherzer allows a liner hit at 105 MPH off the bat that happens to go for an out on his 115th pitch of the night, Matt Williams doesn’t pull him. On Scherzer’s 116th pitch of the night, he allows a grounder from A-Rod that due to a throwing error allows a run to score and leaves men on second and third. Scherzer is finally pulled, the inherited runners leave him having allowed four earned runs on the night. Scherzer didn’t have his best stuff on this night, but keeping him in that long is just managerial malpractice, not just a pitcher’s failure. Scherzer got the loss, 6-5, 2.13 ERA.
Game 13: 2015-06-14 @ Milwaukee Brewers
Scherzer was this far away from a no-hitter
This is the first of Scherzer’s Truly Great Starts we get to go over here. A Complete Game ShutOut with 16 Ks, 1 BB, and just one hit that was a little bloop. These strikeouts came from a pretty representative sample of Max’s pitches overall, five on fastballs, two on changeups, and a whopping nine on sliders. Starting off the first inning Scherzer got a strikeout of Carlos Gomez, and then two routine groundouts, nothing especially special yet. The second inning starts with another routine groundout, but this is where Scherzer puts it into high gear. Over the course of fifteen pitches, he strikes out four consecutive batters. All but one on sliders, those sliders being great horizontal moving and high spin things. A few more field outs to get Scherzer into the fourth, where he proceeds to strike out an additional three straight, one on an 87 MPH changeup and another on a 97 MPH fastball blown past a Ryan Braun caught doing nothing but staring at it. A flyout followed by another three consecutive strikeouts. Scherzer is in the seventh now, and he allows this agonizing blooper that just escapes Anthony Rendon for the only hit of the night. He has to push on though, and he gets two flyouts and a strikeout on a changeup to get through the seventh. The eighth starts with a somewhat difficult for Ian Desmond to play groundout, and then Scherzer makes his only other fault of the day. After a nine pitch battle, Scherzer throws a curveball outside to walk Scooter Gennett. Scherzer still battles though, determined as ever, and strikes out four consecutive batters again, throwing some heaters at 96 even as his pitch count is in the triple digits. Finally, Scherzer gets a groundball off a changeup to complete the CGSO. Extremely close to a no-hitter, very close to a perfect game, but a terrific one nonetheless, this game hit the top of the Bill James Game Score scale, one that tries to be on a scale that maxes out at 100 to symbolize the ideal pitching performance. A perfect 100 here. What can we take away from this? Well I think two things, one is the identity of the catcher. Jose Lobaton is a bad hitter, so he usually doesn’t play in front of the better all-around player Wilson Ramos, but Lobaton is a noticeably better pitch framer, the ability to sell caught pitches that are borderline as strikes instead of balls. Framing matters a whole lot when you’re trying to run up a strikeout total, and I could just tell that Lobaton was a better framer without even needing to look the stats on it up (the stats confirm what I’m saying to be clear, this isn’t entirely pulled out of my ass). Three of Scherzer’s 16 Ks here were called, and to get to a two strike count in the first place you need some framing help, Lobaton provided that in spades. The second, and more direct, reason is that Scherzer was pitching to his strengths. He was more or less a two-pitch pitcher here, 98 of his pitches were fastballs or sliders. And as has become an increasingly popular pitching trend in recent years, if you have a hot high fastball and a deadly curve or slider, just those two can take you a long way. Scherzer’s fastball was going at a consistent 95, he was locating it well, the slider was breaking vertically less than it usually did, generating a pure lateral movement that Scherzer was, again, locating well. Put that together with a 50% groundball rate and some good infield and catcher defense, and you get something like this.
Game 14: 2015-06-20 vs Pittsburgh Pirates
Okay, this is the big one, the one I’ve been dreading. What follows is a description and an attempt to analyze the most agonizing Major League Baseball game of the last half-decade, hoping starting from the beginning can help a bit. Max Scherzer was pitching at home against the Pittsburgh Pirates, a team that would win 98 games only to get eliminated in one game by by Jake Arrieta and Kyle Schwarber, but the only thing that particularly matters is that they were one of the top five hitting teams in the National League that year. Scherzer had his fastball and slider working like they were in the previous game, and also had his changeup and curve working well. He had 10 Ks with at least one of them from all four of his main pitches. He had a lot of quick flyball outs to get through the innings quickly, and before you knew it he had thrown six perfect innings. Scherzer struck out two in the seventh with a flyout to to complete things. Fairly routine eighth inning until the last out, when Danny Espinosa throws out Tyler Moore at first by just a hair to preserve the perfecto. Ninth inning starts, Anthony Rendon makes a nice play in foul territory to get Scherzer the first out. On the very next pitch Gregory Polanco lines out to Rendon. Twenty-six batsmen have stepped to the plate, twenty-six batsmen have been retired. A pinch hitter comes up by the name of Jose Tabata, he battles, he gets to two strikes, he works the count to 2-2, he barely fouls back a 96 MPH heater from Scherzer to stay alive. And then, Jose Tabata does this.
Fuck. The next batter flies out to complete the no-hitter. 9IP, 0R, 0H, 10K, 0BB, 1HBP.
Rule 5.05(b)(2) of the MLB rule-book states the following:
[The batter takes first base when] He is touched by a pitched ball which he is not attempting to hit unless (A) The ball is in the strike zone when it touches the batter, or (B) The batter makes no attempt to avoid being touched by the ball;
If the ball is in the strike zone when it touches the batter, it shall be called a strike, whether or not the batter tries to avoid the ball. If the ball is outside the strike zone when it touches the batter, it shall be called a ball if he makes no attempt to avoid being touched.
I am bringing this up here because while it’s hard to tell from a still image, based on the actual video Tabata fairly clearly lowered his elbow to intentionally get hit by the pitch. This should have been ball three instead of the end of a perfect game. It’s extremely frustrating that it wasn’t, but the umpires didn’t call it that way, this history can’t be rewritten, and it’s possible Scherzer would’ve walked Tabata anyway even if it had been called a ball. Nevertheless, this game went down as the 289th no-hitter in MLB history and not the 24th perfect game. This game actually rates a bit lower on the game score metric than the previous start, 97 to 100, but I think this one is more impressive in the context of the Pirates offense in 2015. They are still both all-time great pitching performances. No pitcher has ever thrown two perfect games in their entire career. Max Scherzer came very close to doing it in back to back starts.^3
Game 15: 2015-06-26 @ Philadelphia Phillies
Feels weird to just have a normal game here. Well, not quite true, Scherzer was still perfect through 6.1 IP, but finally the workload caught up to him and he became tired and mortal. He was pretty clearly missing his spots a lot during those last couple innings of the start that really inched along slowly. I’m somewhat surprised that Matt Williams didn’t pull Scherzer after the seventh, I know he had only thrown 93 pitches, but consider the workload he had brought upon himself at that point, Scherzer had thrown 318 pitches in thirteen days and he was clearly tired and had allowed a run, it seemed like the clear thing to do. Either way, Scherzer was brought into the next inning, allowed a home run on this 93 MPH fastball to Ben Revere that just caught too much of the plate. Nationals won the game, Scherzer now 9-5 with a 1.79 ERA.
Game 16: 2015-07-02 @ Atlanta Braves
This one was a bit unfortunate. Scherzer was really good, he had all his pitches working, he had 9 Ks and just 1 ER with no walks heading into the ninth (perhaps not the best decision but he was under 100 pitches). Scherzer gave up an infield single, got an out on a sacrifice bunt, and then allowed a walk-off single on the stupidest possible hit, a ball hit into the ground that then bounced up and into foul territory. It’s impossible to rule whether that’s foul or not based on the angles provided in the video, and it’s not a reviewable call, but it sure looks like the ball was foul when it bounced over the third base bag. Scherzer had his curveball working in particular in this game, throwing it twelve times and getting two whiffs on it, though he did really get away with an 80 MPH curve thrown right down the middle that thankfully Kelly Johnson only flied out with. Scherzer doesn’t deserve as much credit as he would for this game in a context-neutral environment due to facing the worst-offense-in-baseball Braves with their best hitter, Freddie Freeman, injured because of a bruised wrist. Still, Scherzer pitched well to be unfortunately saddled with a walk-off loss to make him 9-6 on the year.
Game 17: 2015-07-07 vs Cincinnati Reds
Scherzer was gassed. His fastball, slider, changeup, and curve were all slower than the starts before and after this one, and his horizontal movement wasn’t what it usually is either. The obvious reason for this is the fact that coming into this game Scherzer had thrown 100+ pitches in nine consecutive starts, never taking any weeks off there either, he got six days of rest once in that span and four or five every other time. There is also a second reason other than fatigue though, the announcers said that it was very humid out at Nats park that day, and Scherzer may have been having trouble gripping the ball as a result. There’s no real way to know for sure if this was true or not without having actually been Max Scherzer on that night, but it’s an interesting hypothesis and might be part of the reason. Whatever the reason, Scherzer had a lot of pitches miss and a lot of fastballs right down Broadway including one Joey Votto pulled the hell out of at over 100 MPH over the fence. Johnny Cueto had an excellent CGSO performance on the other end, and Scherzer walked away knocked out in 4.2 innings with his record at 9-7 and his ERA raised by 30% of a run to 2.12.
Game 18: 2015-07-12 @ Baltimore Orioles
The Nats at this point were nursing a very narrow NL East lead over the Mets, so while Scherzer was slated to be the All-Star Game starter for the National League, he decided not to so he could pitch this game. And Max delivered, this game was fairly straightforward. Scherzer threw his fastballs, a whopping 72 of them, hard and on target, he had a some help from some good defensive plays, he didn’t walk anyone, and he got through 8.2 innings with only 2 ER. Those two runs were both from Adam Jones solo home runs, which were both on breaking balls well low of the zone that he just managed to muscle out of the park. Scherzer still walked out of this game with his tenth W of the season and his wish to help the Nats in their regular season race over an exhibition game fulfilled.
All-Star Break
For aforementioned reasons Scherzer didn’t pitch in the 2015 All-Star Game. As such, I thought it would be productive to use this break in the schedule to talk a bit about two points.
At this point, Max Scherzer had a healthy lead over all over Major League pitchers in fWAR at this point (Baseball-Reference doesn’t let you do partial year WAR splits). He led MLB starters in Fielding Independent Pitching, had an even 150 Ks for fourth in MLB, led MLB with an astonishing K/BB ratio over ten. Of course he had the almost perfect no-hitter under his belt, and he just felt lik this super dominant mound presence. If there was a Cy Young Award for each half of the season, Scherzer would have the best claim to the NL one in 2015. There are of course two halves to the season, and Scherzer did not win the Cy Young, but we will get into the starts that lost him that award (and the pitching run that gave the ultimate winner it) when we get to them. Scherzer’s megacontract was starting off Gangbusters with the Nats.
During this All-Star break, the most impactful event to the MLB run scoring environment since the end of the steroid era^4 was about to happen, unbeknownst to the players and possibly to MLB themselves. Beginning in the second half of 2015, the Major League baseballs, like the physical objects, started to be constructed in such a way that the air resistance they were exposed to was significantly lower. This allowed the balls to fly further in the air, and as such, increased the home run rate dramatically. In 2014, there were an average of 0.86 home runs per team per game, last year, which seems like it’s probably going to be the absolute apotheosis of this phenomenon, there were 1.39. Run scoring has increased by about three quarters of a run per team per game since then, and it’s entirely due to the home run rate increase, batting averages have gone down and strikeouts have continued to go up since then. This is really the start of the modern MLB environment, and when looking at raw numbers from now on compared to before this point, keep this in mind, and the aesthetic experience of MLB has noticeably changed over these five years because of this. Home Run/FlyBall rate, 2015 first half: 10.7% HR/FB rate, 2015 second half: 12.1% HR/FB rate, 2019: 15.3%
Game 19: 2015-07-19 vs Los Angeles Dodgers
Scherzer went six strong innings with just one run and a K/BB ratio of 8:1. That one run was on a wild pitch and it was a hot humid day against a Dodgers offense that was the best in the National League that year. That being said, the man on the other mound was Zack Greinke, who has a 1.30 ERA heading into this game (and probably should’ve won the NL Cy Young that year but we’ll get to that). Greinke went eight scoreless innings, and despite pitching great in challenging conditions, Scherzer was saddled with the loss to become 10-8. This was about as prototypical a pretty good Major League starter performance as you can find, and Scherzer wasn’t ridden for over 100 pitches this time, just 98, so that’s nice.
Game 20: 2015-07-24 @ Pittsburgh Pirates
Scherzer was bad. He was bad because he allowed three home runs, regardless of the other circumstances of your start or the era that you are playing in, that’s bad. The first home run was a fastball over the plate to a lefty, the second was a breaking ball over the plate to a lefty, and the third was another fastball over the plate to a lefty. My point is that the Pirates were running with a heavily left-handed lineup, and Scherzer as a right-handed pitcher is at the disadvantage in those matchups. Scherzer didn’t feel like he had the stuff for most of the day to really challenge those lefties inside, so he targeted outside, threw a lot of balls when he missed out, and when he missed in it spelled disaster on these three occasions. Also while it’s very silly to take a point about the leaguewide HR/FB rate increasing based on a single game, it does provide some illustrative rhetoric for the juiced ball theory that the second start back from the All-Star Break that Scherzer came back and not long after had a game in which he allowed seven flyballs of which three went out of the yard.^5 Either way, Scherzer was bad, though he uncharacteristically got a No-Decision (which ends a streak of seventeen straight games where he got the decision) instead on an L thanks to an ill-fated Nats comeback attempt in the sixth after he’d left the game.
Game 21: 2015-07-30 @ Miami Marlins
Scherzer didn’t allow a single run, didn’t even allow a flyball that looked like it could go for a homer at any point, but he did walk a season high three batters (the fact that this is a season high is hilarious in itself), why? Well, I think Scherzer was nervous about the previous game where he allowed three dingers to balls in the zone when he was desperately trying to locate something on the edges, so he decided to be okay with walking a few guys if it meant avoiding the homers. Scherzer is generally good enough to avoid homers while still pounding the zone with fastballs, but it makes sense that he’d be a bit risk averse after the previous game, a home run several times worse than a walk after all (groundbreaking analysis here I know). While they didn’t score a run, the Marlins had a lot of well fought plate appearances, especially because their best power threat in Giancarlo Stanton had broken his hand in late June and would miss the rest of the season. The Nats offered very little run support just scoring one run all game, but it was enough, and Scherzer got to 11-8 on the year.
Game 22: 2015-08-04 vs Arizona Diamondbacks
There seems to have been a correction from the throwing outside the zone. Scherzer was just pounding the zone here with fastballs, 71 of them to be precise. It worked pretty well for all but the fourth inning, where Scherzer tried to mix up his pitches and missed the zone a few too many times, leading to a couple walks and a hit to allow three runs. In general though, it was fun to watch, a bunch of called Ks, struts, and nice locating on the edges. Scherzer got away without allowing a dinger, and was allowed to get away with the no-decision.
Game 23: 2015-08-09 vs Colorado Rockies
This game was really off-brand for Scherzer. He had the stuff, his fastball velocity was actually above its average and he hit 99 on the gun at one point. But Max really didn’t have the command, he missed his spot a lot, that only resulted in one walk, but it resulted in three home runs and eight total hits despite 10 Ks. Scherzer was pulled after just 84 pitches, and I have to imagine that this wasn’t just a function of him being a little shaky in a close game, but also that he clearly didn’t have his best stuff in a way that suggests he needed some rest. That’s going to become a broader theme throughout August, which is really gonna be the month that loses Scherzer the NL Cy Young, but we’ll get to that as things go on. Scherzer had been asked to pitch so many innings and so deep into games that even him, who might be the best workhorse in the game not named Justin Verlander, and it took its toll here. Despite that, the K/BB rate shows that stuff dominance, and some credit goes to Rockies hitters for being as good as they were though. This wasn’t an abject failure for Scherzer, just a sign of some negative trends, and speaking of those negative trends…
Game 24: 2015-08-14 @ San Francisco Giants
This is the apotheosis of stuff without command. Scherzer’s pitches didn’t have quite the movement that they usually do, but what really struck him here was just an inability to hit his spots. The first of two home runs he gave up (mentally adjust this to being like giving up three home runs because of San Fransisco’s marine layer) was a fastball on an inside corner, but it was actually meant to be on the other end of the plate, the time it spent breaking in was enough for Matt Duffy to hit it out. The second home run was just right down the middle, no excuse for that. Scherzer gave up base hit after base hit, he walked a batter, and overall had a nasty start that only lasted three innings and allowed six runs in San Francisco. To be fair to Scherzer, he was facing the NL’s second-best offense, and the BABIP was not on his side, it was over .455 on the night. But this is ultimately inexcusable. About two weeks after this game, Cubs pitcher Jake Arrieta would no-hit the Los Angeles Dodgers, being the biggest statement of his God-like 107.1 IP, 0.75 ERA second half that ended up winning him the Cy Young. This two week period was the moment the favorite shifted from Scherzer to Arrieta, and this game is the chief reason on Scherzer’s end.
Game 25: 2015-08-20 @ Colorado Rockies
Why don’t we sit back, relax a little, and enjoy the taste of some nice refreshing Coors? That’s a cheeky way of me saying that while the surface level stats for this start aren’t great (Game Score has it slightly below average at 48) the context makes it acceptable. Coors Field is famously the most hitter-friendly park in MLB, and Scherzer pitched in a way that made it clear that he was aware of this. Usually Scherzer is more willing to pound the zone with fastballs, here he was using all five pitches to avoid giving up the dinger, which he successfully did. This had the side effect of three walks though, and the home run threat is only part of what makes Coors hell for pitchers, the outfield is the biggest in the bigs, combining that with the thin air means even if the ball doesn’t go out, the BABIPs are a mile high.^6 So Scherzer allowed eight hits, and that causes a couple runs to score no matter what, Scherzer came into the seventh, allowed a leadoff single, was taken out, but had that runner score later in the inning, so he gets hit for 3 ER instead of 2. By the way, Scherzer got K #200 in this game, he still has a month and change to go.
Game 26: 2015-08-28 vs Miami Marlins
Inside fastball at 96, pulled for a dinger.
This start really tells you a lot about how weird baseball is. Scherzer started out pitching normally, but got some bad BABIP luck in the first inning that allowed a run to score, and then allowed two home runs later in the game on pitches that really weren’t that bad, especially the first one. But Max picked himself up and pitched a little differently, a bit more gas, a bit fewer fastballs, and suddenly everything started breaking his way. He got some really nice defensive help and some nice strikeouts, (as well as a foul ball pulled hard on a clear mistake pitch that he really got away with). The quality of the pitching might not have changed much, but after that home run Scherzer retired every following batter in a row until the end of the seventh inning, where he was pulled despite being on seven days rest and only having 85 pitches logged. The Nats were behind and in a tough playoff chase, so I get the instinct, but honestly that bullpen was bad enough that a somewhat-off-his-game Scherzer would probably be better on a per inning basis than whoever they had in the pen. Anyway Scherzer got saddled with his third loss in a row, sticking his record at 11-11 despite his amazing actual true talent this year. The 2015 Nats not named Bryce Harper and Max Scherzer were a mess.
Game 27: 2015-09-02 @ St. Louis Cardinals
It’s worth setting some context here as we enter September before getting into this game properly. First, the NL playoff picture in 2015 was a little wonky, the NL Central was this absolute group of death featuring three 95+ win teams, so any wild card path for the Nationals was closed off, they had to win their division. Something they were projected to easily do from the outset of the season, until the Mets had their best season in a decade and left them 6.5 games in the dust going into this game. The Nationals were projected to be the best team in Major League Baseball, they had the NL’s best hitter, the NL’s second or third best pitcher, and they were gonna need to ride them like hell over the coming month if they wanted to mount a comeback. Max Scherzer had to go back to how he was in the first half to make this work. No more 80 pitch outings for sure. Okay, on with the actual game.
This was Scherzer’s best game since July, even if the fact that he allowed eleven hits doesn’t show it. He allowed eleven hits because of a BABIP north of .500 due to a ton of bloop singles. Only two extra base hits got in, one of which was this HR off a fastball that caught just a tad too much of the plate. The actual numbers to pay attention to here are the 10 Ks and 0 BBs. Max used a bit of a weird pitch mix compared to his norm, throwing a lot of curves against a fairly lefty heavy Cards lineup, and they worked great. He also dialed the fastballs up quite well as the game went on. Scherzer’s ability to keep his fastball velo up deep into games isn’t quite as good as the true greats of that category like Justin Verlander, but he’s still good at being able to redline on pitch #86 if it is so needed. Scherzer was actually in line for the win here, so of course the Nats bullpen had to blow it. No harm done in the end though, the Nationals won the game.^7
Game 28: 2015-09-07 vs New York Mets
This would be Scherzer’s one bad start in September. Fitting that it came against the juiced-ball-miracle-Mets at their absolute peak. Peak Yoenis Cespedes hit a dinger, building off of earlier instances of that same act from Michael Conforto and Kelly Johnson. Two of those were off fastballs, though Johnson’s was off a way-too-high changeup. In general the Mets were just able to catch up with Scherzer’s hardest heat, he had to use his off-speed stuff to get through them. That opened him up to a lot of fly balls, some of which (inevitably in this juiced ball world) went out. Max still managed to get away without issuing any free passes though, so his control never really wavered. He did allow a balk though, which I’m only making note of because of just how infrequent balks are. Anyway, a no decision with five earned runs to the hottest team in the NL at the time, moving on.^8
Game 29: 2015-09-13 @ Miami Marlins
Clearly Max decided to brush up on his off-speed and breaking stuff after that Mets start, he used fastballs less often than he had in most of his recent starts, was comfortable with sticking his curves changes and sliders (oh my) out of the zone, and was paid handsomely for it. 8 IP of shutout baseball is the kind of thing that Scherzer hadn’t been producing a whole lot of since his no-no, but he delivered here. 6 Ks, 0 BBs, and no ball that really felt like it was in danger of going out either (though I suspect some of that is the nature of Marlins Park). He got four pop-ups too, which are basically strikeouts, so don’t sneeze at this start just because of its single digit strikeout total. I do however have to acknowledge that one of the five hits he allowed in this game was to the worst non-pitcher hitter of the 21st century in Jeff Mathis, which I’m pretty sure invalidates the start, better luck next time buddy. More seriously, Scherzer got himself an easy win to get back in the black with an 12-11 record.
Game 30: 2015-09-18 vs Miami Marlins
This start was fine, it was fine. Things started off a bit rough with this 92 MPH fastball that missed way inside to be right over the plate for Christian Yelich to smash out of the park and stick two runs on Scherzer immediately. So our man switched to off-speed stuff more heavily during the start. This time though he didn’t quite have the command he had of it in the previous game (also against Miami) and issued three bases on balls after a streak of four straight walkless starts. That being said, he still made it through 7 IP without allowing any additional runs. A lot of that has more to do with the BABIP Gods than anything though, there were plenty of balls in play that very well could’ve gone for hits that didn’t (also Bryce Harper nailed an overconfident Derek Dietrich trying to stretch a single out for extra bases), so Max probably deserved to have closer to three runs allowed. Regardless, two runs is what were allowed, and Scherzer still got saddled with the no-decision thanks largely to the efforts of the opposing pitcher, the late Jose Fernandez, in a game Washington ultimately won in extras.
Game 31: 2015-09-23 vs Baltimore Orioles
This start was very strange. The most obvious thing was Scherzer’s insane pitch count, 122. It’s not like he was trying to keep a no-hitter going, he already allowed two runs from Steve Pearce turning on a high fastball in the first. And he had finished the sixth with over 100 pitches already logged. I looked at an AP game recap and could not for the life of me figure out why Matt Williams left Scherzer in for that long. It’s something that would never fly even just a couple years from then in like 2017. Unsurprisingly Scherzer allowed a home run to Manny Machado when being stuck out there in the seventh like that on a 98 MPH fastball that I’m pretty sure Scherzer knew was gonna be his last pitch of the night regardless of whether it ended up in the catcher’s mitt or in the outfield seats, it just happened to be the latter. Max also used a pretty pure fastball-slider combo for this outing, something that I imagine had a lot to do with the fact that he was getting a lot of Ks in the game anyway, might as well keep it simple even going into the higher pitch counts. Either way, Scherzer got stuck with a loss here thanks to that Machado dinger, and he got himself stuck back at an even record at 12-12.
Game 32: 2015-09-28 vs Cincinnati Reds
It’s worth noting here before going into this particular game that a couple days before this, the Nationals were eliminated from the playoffs, they weren’t actually playing for anything in this game. That being said, baseball is baseball, and there’s no reason to think these teams weren’t giving it their all. This was Max Scherzer’s third Truly Great Start of the year, and as such I’m gonna go a bit more in depth on it. The first inning was nothing too unusual, a steady diet of high fastballs eventually got leadoff man Skip Schumaker to ground out on a nine pitch plate-appearance. This was followed up by 2 Ks on fastballs which were intermediated by a walk to Joey Votto, the 2010s’ greatest walk-taker. The second was rather uneventful aside from a rare cutter striking out pre-breakout Eugenio Suarez. The broadcast was also nice enough to play the top highlight from the previous day’s game in which Jonathan Paplebon tried to strangle Bryce Harper on live television, so that was fun! The third featured Scherzer issuing another free pass, this time to Tucker Barnhart, and Max getting bailed out by left fielder Tyler Moore catching a line drive that likely would’ve allowed Barnhart to score from second base. The fourth started again with another great outfield catch, this time from Michael A. Taylor in center. Scherzer followed this up with 2 Ks on a slider and a changeup respectively. He’s clearly gotten into a good routine here. The fifth fit into that same routine, a groundout followed by 2 Ks (on 97 MPH heat and a changeup), we’re through five no-hit innings now. sixth inning: infield fly, lineout, K on a fastball, no-hitter is still active. Scherzer started off the seventh with a beautiful 99 MPH fastball to strike out Votto, a terrifyingly close groundout following that up, a walk issued to Jay Bruce, and a much less terrifying groundout to finish the inning with the no-hitter intact. Scherzer’s pitch count had reached 98 though, and it had become clear that he’d need to claw his way through the rest of this to finish it off. Eighth inning begins with a nice strikeout off a 98 MPH fastball.
Before we talk about the next plate appearance, I want to briefly break stat nerd character and talk about aesthetic theme here, and I think if there’s been one for this Max Scherzer season, it’s been the near miss. Scherzer has had a near miss on two perfect games, he’s had a near miss with winning the Cy Young over Arrieta, the 2015 Nationals in general have had a near miss with the postseason after being projected to be the best team in the National League and getting the best hitting season of any player since Barry Lamar Bonds, and now, here’s another near miss. Tucker Barnhart lifts a liner over the infield to end the no-hitter, Scherzer ends the inning after allowing a run off of a fielder’s choice and another single. It was still a great start, but it was another case where Scherzer could’ve gotten more if things just broke 5% better. However, we still have one more near miss to cover.
Game 33: 2015-10-03 @ New York Mets
It’s the night game of a double-header in Queens, neither team has much to play for, the Mets have won their division and already have a date with the Dodgers in the National League Division Series that they would ultimately win en route to a National League Pennant. The Nats, as mentioned, have been eliminated. Many of the Mets starters are still going though, so this isn’t a super easy mode that Scherzer got to face here in his final start of 2015. The first inning was routine for Scherzer’s great starts, two strikeouts on fastballs with a groundout to cap things off. second inning for Scherzer featured a third strikeout, again on a fastball. We then have two more Ks in the third, on a fastball and a slider, then opposing pitcher Matt Harvey puts the only ball in play for the inning with an easy groundout. Granderson lines out in the fourth on something that second baseman Dan Uggla was perfectly positioned to grab. The sixth strikeout comes with another fastball toward Ruben Tejada, and Michael Conforto hits a flyball to left to end the inning, Matt Harvey was also pitching very well in this game, so even with these four perfect innings from Scherzer, the game’s still tied 0-0. Two more fastball strikeouts in the fifth to officially bring this into some no-hitter watch territory.
Now we have one final near miss. A throwing error from third baseman Yunel Escobar causes the perfect game to be done. So, spoiler for like 100 words from now, this game ends as a no-hitter. As mentioned before, no pitcher has ever thrown two perfect games in their career. Scherzer was one tiny thing going a bit different in each game away from doing it twice in the same season, and two small things going a bit better in another game from doing it three times. This is still frustrating to watch, but it’s still not as bad as the Jose Tabata situation. Moving on, the inning finished following a couple more outs on balls in play and the ninth strikeout of the game on a fastball to Curtis Granderson.
Scherzer, ever the focused man, didn’t let that phase him though, he had a no-hitter to keep going, and he got strikeouts ten, eleven, and twelve in the next inning without any real resistance. Then Max follows it up with three more strikeouts in the Eighth. Those 6 Ks in a row were on, in order: a slider, a changeup, two fastballs, a changeup, and a cutter. Ninth inning now, pinch hitter Yoenis Cespedes strikes out on a fastball, Lucas Duda strikes out on a fastball. And then on Max Scherzer’s 3359th and final pitch of the 2015 Major League Baseball season, Curtis Granderson turned a fastball for an infield fly that was caught by the same Yunel Escobar who blew the perfecto earlier, closing out his second no-hitter of the year to finish a dominant 2015 season.
Conclusion
First, here’s Max Scherzer’s 2015 season statline, with placements in the National League that year (for things he was in the top ten in) in parenthesis:
228.2 IP (3rd)/14 W (7th)^9/12 L (9th)/276 K (2nd)/34 BB (2nd lowest among qualified pitchers)/8.12 K:BB (led MLB)/27 HR (5th most)/176 H/74 R/71 ER/10.86 K/9 (2nd)/1.34 BB/9 (2nd lowest among qualified pitchers)/1.06 HR/9 (10th)/2.79 ERA (8th)/2.77 FIP (6th)/2.88 xFIP (3rd)/2.63 SIERA (2nd)/2.14 DRA (3rd)/71 ERA- (6th)/72 FIP- (5th)/50 DRA- (3rd)/6.9 bWAR (4th)/6.5 fWAR (3rd)/8.00 WARP (3rd)/All-Star (3rd time)/5th-place in NL Cy Young Voting
Now that’s a lot, but if you’ve made it this far into this post then I’m going to expect you know what most of those mean or are able to easily figure out what they mean. In short Scherzer was an elite control pitcher that when he was hot was likely better than any other NL pitcher in 2015, but much of his second half dealt with home run struggles as the juiced ball made itself known while Kershaw and Arrieta were able to cope. Arrieta won the Cy Young that year, though Kershaw probably deserved it if you take the recency bias of how wild Arrieta’s second half, the thing that was freshest in voters’ minds when they cast their ballots, was out of the equation. But Scherzer slides in comfortably at number three when you take into account the great environment that Greinke had on his side that year to produce his crazy ERA.
If I had to take some conclusions away from this season. It’s that Scherzer was very hot and cold in a season that’s largely remembered for his amazing hot bits, the near misses that I brought up before, and one last thing. Going into this I mostly thought of Scherzer as a power pitcher, but after watching every single pitch he threw in 2015, while I respect his velocity, I gained a greater appreciation for his command, his K:BB ratio was outstanding, and continues to be outstanding for time to come. And while Scherzer pitches quite conventionally (breaking balls as the putaway pitches, fastballs when behind in the count) his delivery is consistent and command stable enough that it still works really well for him, it’s the convention for a reason.
If you’re still reading this after all of this, congratulations, I hope you enjoyed it. This may very well be the longest piece of writing on Max Scherzer that exists on the internet, certainly about this specific season of his. I started putting it together in February when COVID was a thing but I had no idea it would even come close to jeopardizing an entire MLB season. Once that news hit my motivation to do this was killed for a while before I returned a month and a half or so later to finish it up. I’m quite proud of how this turned out, and there’s never been a better time to publish longform MLB writing, so that’s one small fortuitous thing about all of this. I will likely at least attempt to continue this on into a 2016 retrospective but not for a while. I’m probably gonna write about economics for my next thing, I haven’t decided what specifically to do yet, but I have a few things on my mind. Goodbye for now.
Footnotes
Mike Trout signed a $430M deal last off-season, and given that he is literally Mike Trout it’s likely that he’ll outpace Scherzer, but most of that still remains to be seen, so I think my statement holds.
These articles will be actually using a rigid definition of what a Truly Great Start is so I can spill additional virtual ink on them at their respective points. That definition is a start that has a Bill James Game Score of 80 or more. There are four such games that Scherzer had this season.
Amazingly, if Scherzer hadn’t allowed that one bloop hit in the start against Milwaukee, he wouldn’t have actually been the first pitcher to throw two no-hitters on back to back starts. Johnny Vander Meer pulled off this feat in 1938, the fact that something that crazy came even this close to happening a second time is wild, and a testament to just how much sample size and history Major League Baseball has.
Though the juiced-ball theory didn’t actually get fully put forward with confidence until this article in 2016, so nobody would’ve been arguing this at the time.
I do not apologize for this joke.
I don’t have a great place to stick this, so I’ll just mention it in a footnote placed during a game in which Scherzer allowed a lot of baserunners. Scherzer uses two pick-off moves, he has a slow lob one when he’s just checking on the runner, and then he has his real move that he only pulls out when he thinks the runner has a real chance to steal. It’s a bit bizarre, but he seems to be a good enough judge of when to use his A-move that it doesn’t really hurt him.
Worth noting here that the aftermath of this start was the only time all year Scherzer had an ERA greater than 3.00, so if you want a nadir of his season, look no further.
I realize I never really acknowledged it here because I did use it as sort of a general guide for the season, pitcher wins are trash, don’t use them as a way to judge the pitcher’s talent. I was more trying to give a sense of how screwed over Scherzer was by his team this year, which was something that happened quite often.
The Journal of Economic Perspectives (JEP) is one of the several academic journals published by the American Economic Association, and it is also the most interesting one to someone like me. The top flight American Economic Review (AER) is useful for the profession, it takes papers that are usually about a year old, runs them through the ol’ peer review process, and gets them properly published as journal articles. This is important to the actual reliable advancement of economics, but to an observer who is more familiar with econ than most laypeople (I’ve read a textbook and paper or two in my short time) but who isn’t really ready for the galaxy brain jargon and analysis that the most recent cutting edge papers cover, JEP is a lifesaver. It covers topics from all across economics as a discipline by using papers with the specific editorial goal of being understandable to someone with roughly an undergraduate level of knowledge with the discipline. Staffed by a variety of up and coming economists over its now 32 years of existence^1, (though keeping the same managing editor at the helm, Timothy Taylor, for all of that) JEP has been a quarterly constant of providing to the niche market of people like me and those reading this, that also happens to be totally open access. Everything in any JEP issue published ever is totally free as a DRM-less pdf, making it a great place to start to look at any econ topic in-depth. JEP issues are separated into symposia, a bunch of articles about the same issue from different angles, and then some miscellaneous articles about various, perhaps less covered, topics near the end. This isn’t a look at any JEP issue, this is an article about one in particular, the first.
The Symposium
The year is 1987, the economy is booming, the Wall Street high rises have cocaine, Miami has vice, the Major League Baseball has stolen bases, and Art Laffer was already a partisan hack. Whatever the merits of “trickle-down” which is definitely a very serious economic concept that is discussed in modern coursework, Reagan and a split Congress had to hash out a tax plan the previous year that pursued two goals, first, cut tax rates on individuals that already paid it, second, make it revenue neutral by making more Americans pay said taxes. Since the modern American income tax was established during WW2, there has been a slow trend as time goes on for more and more deductions and exemptions being introduced in tax policy until a reboot every 30 years or so. The tax base had a lot of time to shrivel up during the intermittent time and Reagan was out here trying to appease the base with a tax cut while also actually trying to make a true reform (as opposed to the 1981 tax bill, which was very much just a cut), and in retrospect, made a lot of progress on that front in a way that was genuinely good in many ways, though not approaching anything ideal. That tax reform bill was the subject of the only symposium in this first issue.
A total of seven articles are in this section and they all take a slightly different viewpoint.^2 Starting off with a pretty simple overview of the reform and more optimism than one can properly have in hindsight, arguing that this is the start of a reversal of a trend towards incresingly dumb and rent-seeky deductions and that while the act was amazing compared to what was previously, there was needed progress that was ahead of this, instead the trend that I mentioned in the previous paragraph seems to have taken hold, with it having taken 30 years for the most obviously bad part of the act to have gotten chipped away at. The mortgage interest deduction, a way to avoid full taxation on mortgage interest payments, is the perfect storm of something that is politically beneficial due to providing rents for a powerful class of people (entrenched homeowners) without hurting any other one group enough for there to be a proper movement against it, long time readers will remember a similar spiel I gave about protectionism a couple years back. Those compromises were necessary to get the tax act through Congress, but one would hope that they could’ve done more damage to it sooner. Moving on, the second paper from nobel prize winner James M, Buchanan^3 goes over the public choice theory angle of the law, the compromises it had to make to get passed, and who benefits from it. Then my personal favorite paper of the bunch, an article about the original income tax proposal from the US Treasury Department’s economists and how that differs from the final result. it’s a good look at how political compromises happen, and gets into the nuts and bolts of what the economic consensus about how to do income taxation was at the time.^4 Next we have a contrarian take about how this law, while a step in the right direction, doesn’t do a whole lot to reverse the broader trend towards less people paying federal income tax and that it’s not really worth patting ourselves on the back for it unless we get more progress in (narrator who sounds eerily like Mitt Romney talking about the 47%: they did not). It also comments on the general favorability of Reagan to cut taxes on the rich first and foremost, and trying to compensate for that with the not actually all that progressive act of a corporate tax increase. An article about corporate tax and how it affects the cost of various capital investments with some very fun and interesting I swear accounting math, with a general equalization of tax rates on different types of capital (other than the aforementioned travesty of the mortgage interest deduction) from the bill. An evaluation of how the tax reform might lower state-level public spending a bit based on some knock-on effects from elimination of some tax strategies and the lack of crowding out lower levels of taxation by lowering the federal rates. And finally an evaluation of how much these lower income taxes might incentivize additional working for individuals and households, which is to say by not very much. The tax reform was necessary, and it had a lot of good bits from being revenue neutral and also reflects an era of Congressional compromise that no longer exists. However the effects on inequality are still definitely pronounced as a Reagan-era trademark, the mortgage interest deduction is basically a war crime, and there were plenty of loopholes that weren’t closed.
Other Articles
First off we have a basic explanation of Bayesian priors that seems targeted at undergraduate classes (explicitly called “A Note for Teachers”). Not much to say about it really, take priors and evaluate the chance of something being a correct identification using the chances of both the priors and the accuracy of the identification strategy. The main example given is of a taxi who hit someone with a witness who saw it, one should not just take into account the accuracy of the witness at identifying the type of taxi hitting the pedestrian, but the overall relative frequencies of the different types of taxis to be out at night. While the small articles are being taken care of, there’s a recommended readings article that’s in plain enough language I feel no need to summarize all the readings mentioned. A look at some economic modelling/game theory puzzles of which I could figure out the first two (the ones that were explicitly labelled as warm-ups and had answers in the articles I could check with) and then got too big brained for me. And a very silly article about internal American Economic Association politics about whether an astronomer who at the end of the 19th century wrote some important-for-the-time economics texts should have been honored at the AEA’s 100th anniversary celebrations. Onto the meaty stuff.
The Economic Revolution in the American South is a genuinely fascinating look at the evolution of labor and production in the south, and how it went from feeling like almost a different country to America to about in line income-wise with the rest of the nation. The inflection point was New Deal policies that hurt the south’s employment level in the short-run providing a sort of accidental shock therapy that worked out for the people still left in the south as more and more people left. The establishment of the national minimum wage and more granular industrial policies that increased unemployment in the low-skill very low wage jobs that were so common in the south as they became less worth paying the minimum wage for. So a ton of people had no real choice but to just move lol to the north, midwest, and west. The remaining people (who were generally higher-earners anyway) had a lot of leverage against The Powers That Be all of a sudden, and combined with the increased connection with the north allowing northern companies to realize the insane arbitrage opportunities by investing in this much lower-wage lower-price country within a country that has literally zero trade barriers with you and start investing there helped equalize the south with the rest of the country. The paper also considers the paths not taken, and how there might have been southern-specific technological advancements taken and given the south more efficiency in certain areas in much the same way Japan overhauled its manufacturing systems to launch themselves into being a rich country throughout the 20th century. It’s very interesting and fairly approachable, you should read it. Next up are the two most influential articles by far from this inaugural issue. We have to talk about behavioral econ.^5
Anomalies
Richard Thaler has a Nobel Prize. For him to get that Nobel Prize, he first got hired by a very prestigious econ school. For that to happen he built up his reputation in the econ community through a series of articles published over a period of four years finding little issues with markets, finding regular biases that precluded them from being as efficient as possible, on their own not that big of a deal, in combination a suggestion of human psychological fallibility that make strong versions of the efficient market hypothesis hard to swallow. First up, The January Effect stocks, especially the risky kind, increase more in January than in other months in a way that’s hard to just explain away with it being a new tax year, Japan’s fiscal year operates on a different schedule (also just doesn’t have its capital gains taxes work the same way) and the phenomenon persists, same with Britain. The increased returns that risky stocks have relative to what a traditional pricing model might suggest are all in January. Thaler doesn’t attempt to make a particular explanation of this, just pointing out its existence.
Finally, the final paper we come to is also on behavioral econ, a pretty comprehensive review on models of choice under uncertainty, the models that existed in mathematics before econ was even really a thing, the classical econ view, the more modern perspective, and then all the issues that have popped up with it. It cites Khaneman and Tversky a lot, and very much feels like a bit of a lightning round version of a lot of the content in Thinking Fast and Slow. It talks about the value people seem to place on “winning” in its own right even though it seems to lead to what from a purely monetary perspective are intransitive preferences, somewhat hilariously represented by this graph. The way that people seem to have non-linear preferences for these sorts of gambles, as the probabilities get closer to certainty people need higher amounts of additional certainty to make doing the gamble more attractive, and vise-versa (as probability of a good return approaches zero people need more and more approaching to zero to make them more pessimistic). And the ways you can frame things that are identical expected value-wise to get different reactions out of people. The conclusions of all of this are far reaching, but the author concludes the solution here is to not throw the baby out with the bathwater and figure out how to adapt existing economics models to these psychological revelations.
Conclusions
The biggest one is how the general style of academic econ papers has shifted a lot in the past 32 years, a lot of the papers here are much lighter on the math, especially stats, than the econ papers of today. Even a lot of the mathier articles do not discuss regressions, and there’s just more of a reliance on case-studies than today. That being said, the editorial consistency JEP has kept over the years feels nice. Their managing editor being the same the whole time surely helps, and the overall tone and formatting is basically the same as it was three decades ago. JEP is a great open-access resource that people should take advantage of and I hope shining this light on its inaugural issue was enlightening.^6
JEP’s symposia are good at presenting some differing viewpoints without needing to contact any straight up quacks to talk about the topic, a recent(-ish) symposium on free trade in the US had Dani Rodrik present a counter argument, and he’s one of the only critics, if not the only critic of modern free trade discussion in economics that one should bother listening to.
One of the chief members of the board of economic imperialists alongside Gary Becker and the behavioralists.
Reading this also gives you the sense that income tax discourse hasn’t really changed that much at all, it’s a little more nuanced with regards to corporate and capital gains taxation, but we’re pretty much having the same arguments 32 years later over income taxes vs VAT, what a progressive consumption tax might look like, whether it’s the right choice to tax capital gains as income, are corporate taxes worth having at all, etc.
Something I believe I’ve never covered in one of these posts, so this is uncharted territory for me.
Two final things: One, this is definitely the post of this sort from me that spent the longest time in the cooker, I reworked the exact way this post would work a couple times over the space of about four months. If the tone at the beginning seems a bit inconsistent from that of the ending, it’s probably an artifact of that. Two, I partially did this so that I now reserve the right to do one of these about the next issue of this journal in case I don’t have a specific topic I really want to write about at some point, given that my pace of writing these (n = 1, but still) is slower than the actual publishing of these journals, it seems like a pretty much infinite content well.
Zachary Barth is the creator of a series of engineering puzzle games he refers to as Zach-Likes, he and his studio Zachtronics have been making these amazingly creative and tough-as-nails puzzle games since 2011, and they are a combination of both obscure and me liking them enough (you could talk me into putting one of them on the 10/10 games list I did in the summer, but I decided against it) that they deserve some real in-depth written coverage that I haven’t seen for them. As such, this is the start at an attempted project that will try (and almost certainly fail but I might as well take a shot) to do a retrospective look at every single one of the Zach-Likes in detail level-by-level. Starting with Zachtronics’ first commercial title: 2011’s SpaceChem.
What is a SpaceChem
The biggest problem with SpaceChem is that it’s terrible at explaining itself, and this first world, Sernimir II is naturally the tutorial, so I must tackle the worst part head-on. SpaceChem’s intro tutorial is literally a link to an external YouTube video that doesn’t work that well as a tutorial first and foremost because games are fundamentally interactive, and making a tutorial that the viewer can’t interact with is questionable to say the least.
The later tutorials are in the first few levels, but they’re too fixed and noninteractive while also throwing way too much at the player at once, as a result, this particular part will not go level by level through the first world and instead attempt to explain SpaceChem in text form. The bulk of SpaceChem takes place inside of these magical chemical reactors that can directly manipulate individual atoms and molecules. If nothing else that should tell you that this game isn’t really about chemistry, it’s about engineering, it just happens to use chemistry’s iconography as a setting for its engineering puzzles. At the most basic level SpaceChem reactors have inputs of atoms/molecules from two possible windows on the left-hand of the screen, which can be manipulated and bonded together using the bonder pads on the floor and the “waldos”. The waldos are two machines, one colored red, the other blue, that pick up the atoms and do various things with them using tracks and assorted visual programming instructions placed by the player. Once the atoms are manipulated how the player sees fit the atoms can be outputted in one of two windows on the right-hand of the screen, with different missions requiring various input atoms/molecules to be turned into various output atoms/molecules. This is a fairly simple premise once it is actually just stated concisely, but the level of complexity it offers from the interaction of these systems is endless. The Zach-Likes are different from most other puzzle games in that they have no fixed solution, they are just solved by whatever the player can bodge together with varying levels of efficiency (which are then placed on public histograms showing your efficiency relative to everyone else’s), and that makes solving them feel all the sweeter. The commands that are given at the start of the game are ones to change the direction of waldo tracks, to input atoms/molecules from a part of the left of the board, to output atoms/molecules at a part of the right of the board, to grab an atom, to drop an atom, to add a bond to atoms on bonders, to remove a bond from molecules on bonders, to rotate molecules, and to sync one waldo up with the other at a given point along their tracks. That’s a lot presented to the player at once and it’s genuinely overwhelming if you haven’t played one of these games before, let alone if you’ve never so much as opened up a programming text editor. With that explanation out of the way, let’s talk about the two puzzles that don’t hold your hand to the point where you can’t understand anything and just let you play the game. This is where the game shows that it is actually good. (I should note that Zach-Likes do technically have a story in them, but they’re so segregated from gameplay that they might as well not exist for the purposes of this article series).
A Brief History of SpaceChem
This one is quite simple as one would expect, take inputs of hydrogen and chlorine, give them a single bond to make hydrochloric acid, output in bottom-right corner. Not particularly different from the previous tutorial level, all that changes in this one is the fact that the output is going in the lower-right corner instead of the upper-right corner, the real progression of this puzzle is forcing you to figure out how to use the game’s mechanics without help. Which makes this a potential first stumbling block for new players. The only thing one really needs to worry about here though is the possibility of the two waldos getting out of timing with each other and messing the process up after a few runs through. After all, you need to output ten completed molecules to solve the puzzle, not just one. A couple well placed sync instructions will fix that though and then you can be well on your way to finishing the assignment. As long as you have a hang of the mechanics this one is fine, it’s just the game is really bad at conveying said mechanics.
Removing Bonds
This one is the introduction of the idea of removing bonds, and also has two inputs for a puzzle that requires one output. It seems unnecessary to introduce those two concepts at once instead of first having a very easy puzzle in which one just needs to remove bonds and distribute the two separated atoms between two outputs and then doing this puzzle, but this is what we have. Anyway once you have the idea of how to remove bonds under control, the difficulty is putting each atom individually into the same output corner without them colliding. This can be done with a clever sync instruction that gets each waldo to carry one of the two fluorine atoms individually to the output, or with one waldo taking both one at a time, depending on if one wants to optimize on space or speed. If you’re really clever unlike me you can place the bonders right on the output window and not have to worry about carrying them after the fact and optimize both. A solution that I thought of just staring at that screenshot while writing about this puzzle and then implemented in the game, if that’s not a testament to the Tetris Effect type feel that Zachtronics games can have I don’t know what is.
Anyway join me next time at some undefined future date when I look at the game’s second world: Sernimir IV.
Lighter topic here because it’s summer and I feel like it. This is a list of video games I personally consider the best I’ve ever played with a few sentences on each, and is designed to update over time as I play more. For now this will just include games I would consider true ten outta ten mastapeeces, but over time I will probably introduce sections for the 9s and the better 8s when I want to do the writing legwork. With that being said, let’s get started, games are listed by year in categories of whether they’re 10/10 or 9/10 or etc.
UPDATE 1: HITMAN (2016) ADDED TO LIST
The True 10/10 Peaks of the Medium
Super Mario Bros. (1985)
Look, it’s Mario, I don’t have many reasons to explain why this is here that haven’t been repeated by much better writers than I thousands of times. I think the best way to really appreciate how this game may be the most influential of all time is to play any contemporaneous platformer. They all (including 1981’s Donkey Kong, an earlier work by Nintendo’s lead game developer Shigeru Miyamoto) force the player character’s jump to take a rigid physical arc, Super Mario Bros shed this realism in favor of the liberating versatility of Mario’s jumping, which remains the most pure fun to be had in video games as the series continues to this day. You can hold the button longer to vary height, adjust to your wants in mid-air, and come in at varying ground speeds to add distance to it. The level design was a sorely needed example of how to teach a player purely through level design in an age when most games felt utterly incomprehensible. Mario’s good, this isn’t revelatory, but it’s worth revising the original to see just how well it manages to hold up compared to most games of its type.
The Legend of Zelda (1986)
Then the same development team turned around and did it again less than six months later. The Legend of Zelda has aged a lot less gracefully than Super Mario, and it’s very easy to find it annoying if you go in it with a modern mindset. It finally clicked for me when I discarded any internet walkthroughs or hurry to be where I needed to get, and just explored to see what was there in the game in front of me. That pure exploration still can outclass a lot of mindless checklist Ubisoft Montreal developed open world games (with apologies to Assassin’s Creed II). The dungeon design has not aged well full stop though, they’re largely these twitchy combat challenges in a game that does not handle that at all well with very unsatisfying feel of swinging Link’s sword that would get massively improved in future top-down installments. Still though, as a counterpoint to the linear structure of Mario with an exploration based game in its proper historical context, it’s well worth a revisit.
Super Mario Bros. 3 (1988)
This is the first game on the list that I can recommend with no “well if you consider the time” points. Super Mario 3 is the original NES hardware pushed to its absolute limit with a world map allowing a degree of selection and depth to the levels Mario was traversing through that weren’t available in the previous games. Mario’s pure jumping is improved with a more in depth running system allowing Mario to reach a defined max speed with a changed jump animation to match. A variety of new power-ups including the ability to fly, tons of bosses, secrets with more depth than the original Super Mario’s warp zones, and a host of far more creative level design idea. Super Mario Bros. 3 is a fun-producing machine and the peak of what the Nintendo Entertainment System was leading up to, and the last game Nintendo’s A-Team would develop for the system as a result. It’s still one of the best 2D platformers ever made three decades later.
Super Mario World (1990)
Then Miyamoto took a two year development cycle (something that while being relatively short today, was more or less unheard of at the time) and topped himself with a new, better Super Mario, for a new better platform. Super Mario World is the greatest 2D platformer ever made, it hasn’t been surpassed in the years since, no matter how close indie developers have gotten. Taking the Super Mario Bros. 3 formula and bumping it into the 16 bit, adding the sheer liberating power of the cape, the greatest power up the series has ever had, and adding the wrinkle of Yoshi to add extra depth to the platforming, and a basic control scheme that just feels so good. The New Super Mario Bros. games have tried to recreate the magic found here, but they’re all just a bit short of it without enough interesting ideas brought to the table to dissuade one from playing this. Seriously, play this game right now if you haven’t, it’s a masterclass in mechanical design and level design in 2D.
The Legend of Zelda: A Link to the Past (1991)
I swear this list isn’t just a chronology of major Nintendo releases, it just starts that way.
A Link to the Past is a much more structured take on Zelda to the original, and it loses something in that transition towards greater linearity. That being said it more or less created the formula that most of the series would use from here on out. Go to three dungeons to get three MacGuffins to take apart some seal which allows you to get a powered up sword, a new direction, and some twist that the rest of the game will now use. That formula is influential enough that another game on this list that isn’t even part of the Zelda series follows it pretty close. The moment to moment gameplay loop in this game is largely just a much improved version of the original Zelda. There’s an overworld and there are dungeons, you carry a sword, a shield and some secondary item you happen to have equipped. The dungeons are a lot better thought out here though, the puzzles are much more intricate, the dungeon item has been formalized into a proper thing with a massive chest requiring a specific key containing it, and the bosses feel much more interesting. The combat has gotten a lot of well needed polishing up, and it overall makes the experience much easier to recommend to come back to today than the janky-ass Zelda 1 which is more there for impact than modern playability.
DOOM (1993)
Our first non Nintendo release of the bunch, id Software, an American dev led at the time by the Johns Carmack (lead programmer) and Romero (lead designer), more or less single-handedly made the First-Person Shooter a bankable genre. DOOM is a game with zero frills, the plot is that you are the Doom Marine a man who is tasked with killing demons, so kill demons you shall, the entire plot is reduced to a single digit number of textboxes laid between chapters of the game. To talk about how DOOM plays, we need to talk about game feel. Game feel is the audiovisual (and sometimes touch-based given vibrating controllers) feedback that differentiates two actions in games that are totally mechanically identical, but have a different texture to them. DOOM is a masterclass in this and would have a legitimate claim to being the greatest of all time at it if not for its successor (we’ll get to that later in the list). Mechanically, there is nothing making the DOOM shotgun feel so good with its pump action, explosion on firing, and auditory feedback, it could just make a small flash and do damage to the enemies, and that would be incredibly boring and remove all the edge from it. Tools like what the shotgun uses are employed all around the game to really sell the feeling of being this demon slaying badass going through these mazelike levels. Those levels are the biggest roadblock to enjoying the game today, shooters just aren’t designed that way anymore, and you can sometimes get stuck in the later levels for quite a long time if you don’t know what you’re doing, so I’d recommend just playing the first episode if you’re not that dedicated to seeing the entire thing.
Mega Man X (1993)
Perhaps the most famous and influential YouTube video about game design, Egoraptor’s Sequelitis episode 2, is about this game, so really you should just watch that. I’ll try to do the game justice here though. Mega Man was a classic series of six NES games that had tight controls and fantastic level design, but were a little basic when it came to the mechanics. Mega Man X was the transfusion of energy that the series needed to go from great to excellent. There were these upgrades you found throughout the game, most notably the dash, which changes the entire nature of the moment to moment gameplay. While the old Mega Man games were largely played at a fixed pace with rigid screens separated from each other, X uses the dash to have dynamic speeds with faster paced gameplay on levels continuously on one screen. It’s a great deal of fun that does a great job of showing you that without really tutorializing directly to the player.
Super Metroid (1994)
This game is the peak of video games as a medium up to this point. More or less codifying the genre of Metroidvania (with apologies to Metroids 1 and 2, but you were too obtuse) this game is a perfectly executed masterpiece of interactive storytelling and communicating everything through level design and mechanics, with a lot more player freedom granted compared to any game with this degree of polish before. Super Metroid follows Samus Aran through the planet Zebes on a mission to save a baby Metroid from abduction by Space Pirates. The story isn’t that interesting, it’s how its told that’s so good, the only explicit text the player receives is at the beginning with everything else being told through environmental storytelling, a technique that games have gotten really good at in the last 15 years or so but was nearly unheard of at the time. The level design is open but only as much as it can handle to make sure the player knows what they’re doing, the upgrades all feel great, and the mechanics are solid with an ability to master them to a degree where you can completely sequence break the game using mechanics the developers put in totally intentionally like wall jumping, bomb jumping, and shinesparking. One of my top 5 favorites ever.
Chrono Trigger (1995)
The swansong for the 16-bit JRPG before the added storage space and CG animation capability of the PS1 caused most of the genre to turn into these 50+ hour messes with a plot that made zero sense. Chrono Trigger is a collaboration between the Final Fantasy and Dragon Quest creators and it keep things much more simple than what is to come. You are Chrono, generic teenage anime protagonist meant as a blank slate character, you discover the ability to time travel, you also discover that there’s an apocalyptic event that seems inevitable to happen in the future, your (and all the charming characters you run into along the way’s) goal is to go between time periods trying to stop that apocalypse before it happens. That simple overall structure allows the characters to shine through excellent pixel art and probably the best version of the active time battle system found in this era of JRPGs put to use, random encounters aren’t a thing, and enemies instead take up fixed spots on the real in game map that you can then engage in battle with using attacks that actually take into account that spacial set-up. Magic attacks can be combined into double and triple attacks with different combinations of party members allowing for a lot of experimentation and an incentive to keep switching between all your party members instead of slipping into a single orientation. The game is about 20 hours long which is a much preferable length to the “please insert disk 27” of the PS1 Final Fantasy games, and also unlike those games the game pulls off its story with some self awareness and a sense of fun that future FFs other than IX and XV really need an injection of.
Super Mario 64 (1996)
Now this is how you make a jump into a new dimension. Mario 64 is a full 3D game from 1996 that still holds up great, something that couldn’t be remotely said (and even then those games have aged worse) for any other game pre-1998. Mario 64 is Mario turned sandbox and 3D, and everybody else entering 3D for the first time apparently didn’t know what they were doing except Nintendo’s core team. It’s been mentioned to death but it’s still worth talking about that Miyamoto spent the first several months of development just working on transferring Mario’s movement to 3D and it shows. Mario’s basic jumping feels just as revolutionary in here has his 2D jumping did in Super Mario Bros., he can do a standard jump but also has a bunch of new acrobatic moves to adapt his versatility to the third dimension. Double jumping, triple jumping, wall jumping, backflips, long jumps, and side flips, they all feel great. The later Mario games would perfect this control scheme, but 64 did the most legwork of any. Play a game like Bubsy 3D to see how badly you can screw this up, or Crash Bandicoot to see how far Mario elevated jumping (pardon the pun) above the most basic 2D->3D mapping. Also it has the best level design in any 3D platformer ever which probably should be mentioned but I don’t want to be here all day so I’ll leave it at mentioning that their toybox design style makes them incredibly deep and memorable despite underwhelming total size by modern standards.
Half-Life (1998)
The moment a genre of games stopped being DOOM-clones and became the First-Person Shooter. Half-Life is Valve Corporation’s debut game, and as a result there was little pre-release hype. It managed to get published by an ill-fated Sierra that was on its last legs at that point though, giving it just enough of a platform to become the smash hit it did. Half-Life is all about immersion and player projection onto the protagonist, the game has no real cut-scenes, you are Gordon Freeman, Gordon Freeman never talks, and you never leave control of his body. Instead of the traditional maze level based system of DOOM and its clones, the world of Half-Life is one continuous linear stretch, allowing for much tighter control of pacing than those games, and no developer does pacing like Valve. New and interesting weapons and situations are introduced at a constant rate that never allows more than the occasional dull moment, with the first confident and successful dive into a full 3D PC shooter after Id’s Quake took some tentative steps there and RareWare’s GoldenEye 007 showed how it was done on consoles. The game starts you off with an uncharacteristic for the genre introductory segment that’s about 20 minutes long in which you do not do any violence, but it takes off from that, and as Gordon Freeman runs around the Black Mesa facilities and into the hostile alien world of Xen, the game never really stops its balance of thoughtful level design, varied weapons and enemies, and tightly-paced story.
The Legend of Zelda: Ocarina of Time (1998)
It’s the most critically acclaimed game ever and Miyamoto’s swansong from being the true creative lead of Nintendo’s high profile efforts, capping off an unparalleled 18 year run from 1981-1998. You take your silent self-insert protagonist from a weird elf child to time traveling warrior God possessing every item in the game’s universe in the space of about 30 hours without a tenth of the jank Western RPGs trying to accomplish the same sort of experience at the time had. It introduced the targeting system to make third-person 3D melee combat manageable the series still uses and that another series we will be getting to later basically steals in whole cloth. The dungeons are great, the overworld is great, the boss fights are great, and I want to move on from talking about this before I actually try to say anything remotely original about this game, just an FYI to play the 3DS remake if you can, if only for the Water Temple being turned from a menu navigating nightmare to the best dungeon in the game.
Deus Ex (2000)
If DOOM got the PC first-person game off the ground, and games like Quake and Half-Life raised the bar for technical and narrative innovation, Ion Storm Austin‘s (read: Warren Spector and Harvey Smith, not John Romero) was one last triumphant leap over the bar to cap off things before Halo came out and completely rewrote the book on FPS design. Deus Ex is only kind of an FPS, it is a first-person game in which you can shoot things, but it’s also a stealth game, and an action RPG, and a narrative-driven action-adventure. The shooting actually feels pretty weak in comparison to all the rest of it, this truly is a game that can only exist on PC, the default reload key is semicolon, and it’s really its own genre of game, the immersive simulator, a genre focused almost entirely about giving wide open levels that run on some traditional action game rules, but maximizing player choice within those. Deus Ex might not be the most polished of those games, and as you work through a thick, pulpy narrative where every “it’s after the USSR fell but before 9/11 so American culture doesn’t have a common villain to focus on and as a result we’ll invent our own” conspiracy theory (that really has not aged well in the face of the Alex Jones types that exist now) is true, through levels that start great but become less in depth as time goes on, you get the feeling that this game probably could’ve been about half as long without losing much. This game might be the clearest legacy pick over modern playability standards here, but it deserves it.
Metroid Prime (2002)
Welcome to the sixth generation of consoles, and this is probably the point in this list where I show off my Nintendo myopia the most, there is another console science-fiction FPS that came out to start this console generation with a bang and I have not included it because despite having played it I didn’t find it that great, but this is my list dammit. Metroid Prime is the result of a collaboration between American and Japanese creatives working at the Texas-based Retro Studios and Nintendo’s main offices, respectively, trying to adapt Super Metroid into a First-Person Shooter medium, and they succeeded with flying colors. While Super Metroid’s map is organized in this perfect way to make its pacing perfect in a way that prevents this game from quite surpassing its predecessor, Prime makes up for it with a sense of atmosphere and style impossible to get with the hardware limitations of the SNES, it’s that core Metroidvania sense of exploring dangerous worlds unknown brought to the highest level of polish it would ever receive, the combat feels like it’s taken a hint from Ocarina of Time with a lock-on system that allows you to do some Serious Sam style circle-strafing on a console without issue, and because this was before the other console FPS I have neglected from the list that standardized how FPS games on console would control had fully established itself with its right analog stick for aiming thing, to aim freely you need to be standing still and holding the R button, which works surprisingly well. Otherwise, it’s the Metroid experience but on a new planet with better presentation, atmosphere, and boss fights. I could muse for a bit about how modern Metroid seems to work best when Americans are given the chance to work close with it and what this says about the culture it reflects, but that is for another day.
Paper Mario: The Thousand Year Door (2004)
My personal favorite game of all time because of nostalgic memories, even if I acknowledge there are games better to people who didn’t grow up with this game, a switch of the usual JRPG monotony of cringe-inducing angsty anime plots next to a battle system most akin to spreadsheet management, this game is colorful with a stupid sense of humor and a twist on the usual turn-based battle system, with action commands one must hit in the middle of battle to block enemy attacks or increase the power of one’s own. This game is a sequel to the original Paper Mario, released in 2000 as one of the last Nintendo published titles for the N64, and while I have more personal memories with the original, this one does everything at a more interesting level. The Thousand Year Door is split between a bunch of largely self-contained chapters with these storylines that provide a vehicle for humor in one of the best Japanese-to-English translations done for a video game, the standouts among those chapters are the third, in which Mario rises up through a professional wrestling circuit and has to take down corruption from the inside, and the sixth chapter in which Mario has to solve a Murder on the Orient Express style mystery. It’s the good parts of JRPGs with a lot of stupid levity injected into it to get rid of the bullshit, what’s not to like?
Half-Life 2 (2004)
This is how you make a sequel that takes the original game and jumps off of it into a new level of performance. That phenomenal sense of pacing is perfected here with gunplay that’s less hectic and DOOM-based as the original, but has more thought and punch with the types of weapons you have as you go around this post-apocalyptic world in the aftermath of the mess you made previously. Starring in the weapon show is the gravity gun , allowing you to pick up physics objects in the newfangled Valve Source Engine (it was novel at the time okay!). The level design is unique and varied with a memorable story that, while not particularly deep, works perfectly as a compliment to that gameplay, all but one or two segments of the game are memorable in a way that really sticks with you. The ending of the game’s breaking of the standard FPS rules in a way that I won’t spoil but I can assure you works very well. After this we got too underwhelming episodic sequels leaving us hanging with cliffhanger endings for the last 12 years, this remains the last proper chapter of the series, and it stands as that at a great level.
Portal (2007)
In 2007 Valve released the Orange Box, a big physical retail release of Half-Life 2 with its 2 episodic sequels bundled with it, a new and radically different Team Fortress sequel, and a small little experimental game left there as a bonus. That experimental game was Portal, a masterclass in everything Valve was amazing at during their peak and the best video game ever made.
Portal is very short, only taking about 3 hours to beat, it’s a puzzle game directed around exploring one relatively simplistic mechanic to its full consequences, the ability to fire a system of two connected portals using a gun, left click fires one of the portals, right the other. This mechanic brings some simple puzzle solving mechanics, but also has second order effects once things like momentum, movable cubes, and turrets come into play. You complete these puzzles in test chambers for the mysterious Aperture Science while a robot voice talks to you in a way that first comes off as just pre-recorded, but as time goes on you come to realize that this robot is self aware and a real character, GLaDOS. New mechanics are introduced at a perfect pace that aligns with the placing of story beats until the final test chamber, in which you encounter the most famous plot twist in video games, the cake is a lie. You hop out from GLaDOS’s attempt to incinerate you and slowly go in the back corridors of the decrepit Aperture facility until you finally take the fight to her base, you proceed to take her apart one piece at a time, and then you get cake after escaping to what is presumed to be the post-apocalypse world of Half-Life 2, having experienced the peak of what the medium has to offer. Portal has perfect gameplay-story integration, the puzzles are fun, the dialog is funny and interesting, the difficulty is ratcheted up at a level of mastery not seen in any other game, enough to make Miyamoto go out and commend it on its ability to do this. The pacing that Valve had learned such amazing control over in the Half-Life games prior to this comes out as the distinctive feature here as they’re forced to pare it down to three or so hours, all good stuff, no filler. It would take a lot to top Portal for greatness in the medium of video games, it would probably require something similarly tight with something actually profound to say on top of that, Portal isn’t exactly a game with some sort of moral, but it’s so mechanically and narratively elegant that it doesn’t really have to.
Super Mario Galaxy (2007)
It’s been a while since we’ve last checked in on a main series Mario game, his previous escapade in Super Mario Sunshine had a lot of good ideas but struggled with a ton of frustration in the execution (hello blue coin hunting!) that prevented it from making this list. Galaxy, however, is Mario’s peak of just pure mechanical fun. 11 years of development from Mario 64’s control scheme have created a game feel for Mario that just feels amazing. The cavalcade of moves from 64 is back, with the addition of a Wii Remote waggle based spin that gives you some extra air time and works as a short-range attack in a way that easily maps onto how people react to falling down in a Mario game (seriously watch people who aren’t familiar with video games try to play a Mario game they uniformly shake the controller when falling). Level design here is more linear than 64 and Sunshine, but the pure platforming is so good they make it work. The gimmick here other than motion controls is Mario operating with his usual moveset in micro-gravity that slings Mario around between planets, it’s fun, it’s energetic, it’s Super Mario.
Limbo (2010)
Most of the other games on this list use their mechanics to give the player pleasing feedback to attempt to be, for lack of a better word, fun. However, viewing games as only a medium for delivering pleasant experiences limits the emotional range of the medium, most of the energy to making games that aren’t traditionally fun is towards horror, a genre that is very much not my thing, but there’s still a lot of other feelings the uniquely kinesthetic experience of game feel can express, Limbo is an exploration of those. Limbo isn’t horror per se, but it certainly isn’t fun, it’s interested in giving you the game feel of an ultimately uncaring world of purgatorial, well, limbo. Everything here wants you dead, not because they’re evil, but because that’s just the only way to survive in limbo. The underexposed monochromatic style adds onto this dark nature in a way that is never communicated using any actual language other than what comes naturally to the interactivity of games as you die over and over again trying to solve the puzzles in this puzzle-platformer, and to put it in pure view what kind of world this is, when you complete the game, your reward is minimal before you get booted back to the start, you are stuck in purgatory here, and the only winning move is not to play. Limbo was the best game to come from the wonderful experiment of the Xbox Live Arcade’s Summer of Arcade program, in which some of the best indie games from an era in which the scene was burgeoning were released using Microsoft’s platform. A spiritual sequel was released by the same developer in 2016 with Inside, but it just came off as a flabbier version of this without the same punch and unified artistic vision, no reason to play an inferior version of Limbo when Limbo exists.
Portal 2 (2011)
Portal 2 loses something in the transfer from tight experimental game on a release being paid attention to for other reasons to massively hyped AAA tentpole with thrice the length of the original, but worse than the greatest game ever still leaves it in like the top ten all time. Your escape at the end of the original has been retconned into a recapture and being in a coma for several thousand years, the facilities have decayed, and the writing is less subtle but funnier in my view (this is probably a British vs American humor taste thing, you’ll probably find the understated sarcasm of Portal 1 funnier if you’re British, and the cleverly written gags of Portal 2 funnier if you’re American) with an English accented robot named Wheatley guiding you around as you accidentally manage to wake GLaDOS back up. From there the first third of the game plays more or less just like more of Portal 1, as you go through more test chambers with a more openly antagonistic GLaDOS narrating the experience with some new mechanics (lasers that you can direct, trampoline pads, and bridges made of light) added to spice things up. Things seem to be headed to a formulaic ending as you escape with Wheatley and have another boss fight with GLaDOS, then you install Wheatley to be in charge of Aperture Science, and he promptly gets corrupted by power and sends you and GLaDOS (now operating via potato battery) down a shaft to the second act of the game, where you encounter these three different colored goos to interact with as you discover the darkly comedic history of the Portals, with original Aperture leader Cave Johnson leading the way, then you make your way back to Wheatley and take him down, install GLaDOS back as leader, and have her send you out for a true escape. It’s more Portal, but more of the best game ever with clever new mechanics and similar puzzle sense is good enough by me. After that, you have a quite difficult co-op mode which is pretty meaty and not that different in length to the game proper, and then that is the end of things. Valve has not released a proper single-player game in the eight years since Portal 2’s release, most of the writers for Half-Life and Portal have left the company, and most of their efforts are towards Steam and support of their two massive esports titles, Counter-Strike: Global Offensive, and Dota 2, they’ve recently gone back to their ways of acquiring other developers to work with their touch at least, with Firewatch developer Campo Santo going under their wing for an upcoming release of In the Valley of the Gods. In terms of what people ultimately want though, a proper Half-Life and Portal 3, total radio silence, at this point it’s mostly an embarrassment that the company no longer wants to talk about, leaving fans without any sort of closure, we still wait though, hopeful that one day we can once again put on the HEV suit or Long Fall Boots, ready to get back into what this great developer can deliver next.
Dark Souls (2011)
Most games follow a design ethos of being fair, ideally the mechanics of the game are communicated well enough through level design that a first-time player could theoretically get through the game without ever reaching a failstate so long as they’re technically skilled enough, the cards are laid out on the table for the player to read clearly. Dark Souls did not get this memo, the way Dark Souls teaches you about a mechanic in it is to kill you because you don’t understand it, thus communicating the sense of the world Dark Souls takes place in, but also teaching you a valuable lesson of what not to do in a particular situation. To match with that, the narrative and sense of direction in the game is a mess to find at first blush, but what a lot of readings of the game miss is that this isn’t because Dark Souls is trying to be this impenetrable mess of a game shutting out those that can’t fully master its Monster Hunter-y brand of weighty third person swordfights. Dark Souls is actually trying to deliver a very inspiring message using the power of the internet, the way to get through some seemingly insurmountable obstacle of a boss fight or navigational obstacle with the power of jolly cooperation, you’re meant to summon players from the internet or NPCs for the toughest bosses and to look up walkthroughs to figure out what the hell to do. In a way Dark Souls is the best Legend of Zelda Nintendo never made, the combat feels better than any of the 3D Zelda’s manage and the level design has this winding doubling back on itself feature usually located in Zelda dungeons. There’s a reason Dark Souls is the entrant from its series that makes it to this list, and it’s not that it did this first, Demon’s Souls was a trial run of sorts at this style of game that came out two years prior and won some Game of the Year honors. Dark Souls 1 is elevated over the rest because of its world design, it’s actually a Metroidvania, you explore this ever expanding interconnected world looking for ways to get tools to unlock access to new areas, and the game’s lack of a fast travel system for its first half really hammers this home, Dark Souls isn’t a series of levels, it’s Lordran as a whole, not unlike the Metroid planets of Zebes and Tallon IV or Dracula’s Castle from Castlevania: Symphony of the Night. That’s what truly finishes the Dark Souls experience into this supremely engaging experience of using the internet’s cooperation to uncover the lost secrets of this difficult to traverse long-dead world, and getting to the top of it through a lot of improvement of not just an RPG character you grinded up, but your actual skill.
Fez (2012)
This might be the most controversial pick on the list, and I generally don’t subscribe to the general hate surrounding Phil Fish, so keep that in mind as I praise this game I guess. Fez is an ostensibly simple game about going through 2D areas with the best indie pixel art I’ve ever seen, with the wrinkle of these areas not really being 2D, there’s four different perspectives that change how the levels go. As time goes on you learn about some real fucking obtuse puzzles with an entire alien alphabet you need to decode. There’s not much to get into here without spoiling puzzles or getting into the mess of Phil Fish controversy, so I’ll just say it’s a beautiful experience that does my favorite thing of running with one mechanic as far as you can take it and it’s worth checking out.
Papers, Please (2013)
For all of the fun and interesting games I’ve listed so far, outside of Deus Ex (and even then it’s pretty silly) there’s been a lack of games listed that actually really say something, Lucas Pope’s solo developed Papers, Please very much has things to say about authoritarianism and anti immigration policies, they aren’t nice things. You are an immigration officer in a fictionalized version of Cold War era East Germany, you make shit wages, and are constantly forced with this ever changing endless bureaucracy in which your well being is always in jeopardy and you struggle to feed your family. The core gameplay loop is not really fun, you’re forced to manage documents for potential visitors and immigrants, but the moral decisions you’re forced to make of whether to follow code or to let someone in to see their family for the first time in years are what make it interesting, with multiple endings to go along with the choices for these. The atmosphere of oppression in this game is strong, and provides a genuinely solid political point about immigration policy being made within the fiction of a video game, something that the immaturity of a lot of the medium almost never manages to do beyond the most surface level. And it communicates this feeling almost purely through the way the gameplay loop forces you to think and feel about what you’re doing, truly something that only a game could accomplish.
DOOM (2016)
After the 1996 release of Quake, Id Software started going downhill. John Romero was forced out, leaving behind the anarchic story premise ideas and design ethos that he had brought with him, other designers kept leaving over that time as John Carmack remained directing the technology of Quake 2 and DOOM 3, games that were more impressive as game engine tech demos than entertainment. In 2009 ZeniMax Media (Bethesda’s corporate arm) acquired them, and in 2013, Carmack finally left. A DOOM 4 was in development hell for a long long time, and it looked to be a generic Call of Duty style shooter that lost everything that made DOOM real, but somewhere along the line they restarted development into an attempt to not merely recreate the aesthetics of DOOM, but to recreate the entire experience for a new generation. What we got was the pleasant surprise of the decade, DOOM 2016 takes the game feel of the original and rams it up to 11 with the power of modern graphics technology for a gory experience that has the best game feel I’ve ever experienced. The entire game works as an antithesis to the regenerating health cover-based shooter that requires a lot of waiting around and can be more tedious than anything, the motto of DOOM 2016 is rip and tear, a pure experience of kinesthetically pleasing gameplay and the stuff an immature edgy teenager’s dreams are made of. There’s a legitimate argument that DOOM 2016 is just too devoid of substance to really be on this list, but I think that robs games’ own intrinsic worth as a medium, games aren’t uniquely good because of an ability to deliver audiovisual messages in a way film already could, it’s the ability to have you be the participant in that action. The traditional narrative of kill demons doesn’t need to be that complex to make the game feel and ludonarrative of the complex pinball ways you take down those demon’s to be interesting.
Hitman (2016, and also early 2017, it’s complicated)
DOOM wasn’t supposed to be good, neither was this, Hitman was a series best remembered for its PS2 entries that’s only 7th gen entry was Absolution, a game in endless development hell that came out (I’m told) as more of a generic cover shooter than a true Hitman game. This game was released episodically as an always online live-service package mess under a publisher in Square Enix that was losing confidence in Danish developer IO Interactive, and sold poorly as a result. But in 2017 the full game of the year edition and whatnot was released, and IO Interactive was sold off to its own management (a painful but ultimately fruitful process of letting them go indie) word started to spread about how great it was, and people started buying it for actually quite good sales figures. I had never played any other game in the series before diving into this reboot a few months back (I have heard 2006’s Blood Money is really good, but I have yet to go back to it), and with a cool 40 hours now logged into it, I can now firmly put it on this list as a true triumph of level design. Hitman is theoretically a stealth action game like the Metal Gear Solid games before The Phantom Pain, and the game takes place on a series of 6 levels in which you are tasked with killing some set target(s) and make an escape, and 6 levels in an MGS style seems like a rip-off, then you see the level design. All 6 of these levels reveal the genre Hitman actually is, it’s an immersive sim. Every level is designed in that oh so satisfying immersive sim style of being like an actual location with a ton of alternate pathways to get places. The different ways you can chain together strategies to kill your targets makes this game makes this surprisingly smart for a game centered around killing people, it can become this wild puzzle game during higher difficulty or challenge runs. The way this game gets its playtime from you is trying the same levels over and over again to do some challenge, kill a different target than the usual one, or to play it at the high level professional difficulty. The highest levels of these challenges turn it into one of the wildest video game experiences you can have, with all these emergent stories coming from levels that rely on you running around and interacting with systems rather than there ever being some strict choreography you must follow, and it does this all without feeling bloated.
Super Mario Odyssey (2017)
I’m really running out of new things to say about why the Mario series is so bursting with joy and creative energy to routinely land its biggest releases spots on this list, but Mario feels as good as ever in Odyssey, and he can throw a cap to possess enemies in certain ways to get to objectives. The real star of the show here isn’t the cap though, it’s the fact that Mario has gone full Banjo-Kazooie style open world collectathon, with 880 unique moons in the game instead of the standard 120 stars, levels are these open concept puzzle boxes that Mario can platform around to get as many moons as possible, with tons of levels of depth people can get into the game for how many of the moons they feel like getting, with only 124 of those moons required to beat the game, and tons of bonus content included to get power users like me to stick around. It’s just oozing fun everywhere with Mario being his usual cheerful easy-to-learn-hard-to-master self. Super Mario Odyssey came at the end of Nintendo’s Wii U long slump as the final hurrah to announce that the Switch was real beyond just having one game in the form of Zelda, and Nintendo was really and truly back on top.
Celeste (2018)
Celeste is a game that feels like it was expressly designed with my interests in mind. It’s a balls-to-the-wall hard 2D platformer that still has frequent checkpoints and has a protagonist that deals with very similar mental health issues to me. This is a great example of how to avoid getting labelled with the infamous buzzword of ludonarrative dissonance by having gameplay about overcoming difficult challenges where the only solution to overcome your self-doubt about them is through persistence, and a story about that very same thing, these two aspects of the game are constantly firing on all cylinders. It’s not for everyone as a result, if you don’t relate to that sort of story at all or are just sick and tired of pixel art indie games this probably isn’t for you, but for me Celeste is fantastic. The central platforming mechanic of the game is a mid-air dash that can go in 8 directions and is recharged upon hitting the ground or hitting these recharge tokens here, the game’s seven main chapters each have a central gimmick that operates really well with that dash and is explored as far as it can be explored by the end of it, with a bonus 8th chapter adding to that, with some additional B-side and C-side levels to really take the difficulty up as far as it can go. It’s really all a great way to use traditional gameplay mechanics to tell a story that fits that mold, as opposed to the usual arty game standard of sacrificing the well thought out gameplay traditions to fit the mold of the story.
Conclusion
That’s it for this list now, I expect it to update over time as I play more games (there are two games I have on the mind that could make it onto the list soon given more time with them). In addition I might add an additional section for 9/10 type games or games have some noticeable flaws that prevent them from being transcendent but I still love anyway. This is already over 7000 words though, and I’m okay with letting this sit where it is now, it’s a pretty solid recommendation list. If you want to give me feedback on this idk @ me on twitter or something, though I’d ask arguments with me be about what is on the list instead of what isn’t, there are a few games that are very critically acclaimed that I very consciously didn’t include for problems I have with them, but nine times out of ten it’s not on the list just because I haven’t played it, not out of hard feelings.
With all that in mind, you should go play some of these video games.
Note: This is an adaptation of a post I made for reddit that didn’t get that much traction due to covering a player from a small market team, so I’ve decided to move it here
An Explanation
Near the beginning of the season this year the guys at MLB Advanced Media working for Statcast/Baseball Savant announced that they had made video for every single pitch since the start of the 2018 season available and searchable on the site. I’ve wanted to do something with this for a while, so this is my first attempt to do that. I’ve wanted to do a pitch-by-pitch breakdown of a relief outing for a while, and this gives me that chance.
Enter Kirby Yates, he was pretty good in the first month this year, no homers allowed, sported a 14.06 K/9, and got 14 saves to his name. All of this gave him 1.0 fWAR in March/April this year, the most of every reliever. It’s worth noting that these 14 saves are probably inflated compared to his actual talent, the Padres have played him as the most generic-ass 1 inning closer possible and have also played in a lot of close games, saves are a bad metric, I’m sure you’ve had this beaten into you before if you’re reading this. Before getting into a specific appearance of his it’s worth giving a brief overview of his pitching style (stealing a lot of material from Ben Clemens for this bit). Kirby Yates is a right handed relief pitcher that as mentioned is used exclusively as a one inning closer with 2 main pitches, a four-seam fastball that sits about 93-94, flashes 96 from time to time, he locates it well and it gets a good amount of spin, but otherwise it’s relatively unspectacular on its own. Where Kirby Yates has really separated himself is a splitter that he developed in 2017, before he was a fastball-slider guy like many other relievers, but neither of those pitches were amazing and generated a lot of fly balls. With his splitter though, that changed, it spins pretty slowly giving it an almost changeup like quality and allowing it to drop off a table when it needs to. That 1-2 punch and a deceptive delivery to make the pitches hard to tell apart has been a rising tide to lift all his boats, with even his four-seamer seeing much improved results (36% whiff rate this year vs 25% in 2016 via Baseball Prospectus) now that he keeps hitters guessing with that combo. He’s also gotten pretty weak contact against his splitter, putting his ground ball rate over 50% for the first time this season. In conclusion, good command, good 2 pitches with mixing and tunneling and whatnot, and improved ability to restrict quality of contact over a large sample.
April 12, 2019
Now for what I actually came here to do, take a look at one of his best outings in April and go pitch-by-pitch to look at what he’s doing and commentate on what this says about him as a whole. I have selected his outing on April 12th in Arizona against the DBacks, where he retired the side in order to secure the save and flashed some higher fastball velo than usual, let’s get started.
Starting off here with a pretty simple fastball where he just missed his spot against Wilmer Flores. Maybe he held onto the ball for too long, it doesn’t look like it from me glancing at the exact frame he releases the ball, my guess is his grip was weird or his wrist snapped in a weird direction. Nothing much really to talk about here.
It didn’t work the first time, so he tried it again. Hit his spot exactly this time and taken for a called strike by Flores despite being right over the plate. This is likely due to him not knowing whether it would be the four-seamer or splitter and deciding to not risk embarrassing himself by taking the swing, that being said Flores isn’t that aggressive in general, swinging at less than 65% of pitches in the zone, a common tactic for catchers not confident in their power or BABIP ability, moving on.
The first splitter of the outing, a really good one that has a ton of break and is done a disservice by the slanted camera angle in Phoenix making the inside break of it look more obvious than it actually is, Flores still has the bat speed to catch up with it (I have to imagine he was sitting on the splitter given him not swinging at the fastball right down broadway last pich) but for nothing more than a foul tip that isn’t that much better than a whiff.
Back to the 4-seamer, this time hits 95 on the gun while missing his spot a little higher than as intended. Didn’t particularly matter, Flores did catch up with it but not fast enough to produce more than a foul back.
Splitter again, he hit his spot, but Flores’ passive approach paid off here and he lays off. It also breaks a little earlier than Yates probably wanted it to. Not a big deal, count’s just even.
He actually misses his spot by so much here that it goes out of the zone the other way (I think he let go of it a little early), this does allow Flores to catch up to it but just for an ineffectual grounder to Tatis. One down.
Yates hits his spot against David Peralta well and Austin Hedges frames it well to get a called strike on what was definitely an inch or two out, worth noticing here that the catcher hasn’t called for a high fastball yet here like wot all the cool reliever kids who throw 98 are doing the last couple years instead opting for the strategy that was en vogue earlier this decade of fastballs lower in the zone to try to get people to hit it into the infield. It makes sense, he has basically average velo, and the splitter gives him enough whiff potential without having to blow high fastballs by people like Josh Hader or Justin Verlander. The fact that a command first guy is up there with the top relievers in MLB right now even in this environment where everybody seems to be going high velo fastball-slider is interesting.
Yates tries to go for a 3 pitch strikeout with the same pitch 3 times but this time he misses his spot and it goes outside for a ball. How much horizontal break this had vs just heading in that direction from the start is hard tell because of the aforementioned camera angle. This particular one bothered me so much that I went to use another Baseball Savant thing with their 3d pitch viewer and it appears to mostly be break, just less than the camera would have one believe. The count is now 1-2.
After doing the same thing 3 pitches in a row Yates changes tack going for his first high fastball that he actually meant to throw. Peralta doesn’t bite and he still misses a bit above where the catcher was signalling. Yates seems to have an issue with commanding his four-seamers once they get into the 95+ range, maybe he’s actually not trying to throw them that hard and that’s just what happens when he lets go too early, I don’t know. Now I kinda want Eno Sarris to interview him so he can get into the weeds of his pitching style now that I’m invested invested in it, anyway.
Same pitch as his first 3 this plate appearance, holds onto it too long and throws it in the dirt. I have nothing interesting to say here, I guess Hedges does a good job of grabbing it as soon as he realizes where it’s headed.
I’m starting to think that the scouting report on Peralta had something on his ability to hit fastballs low and away because this is now the 5th out of the 6 pitches in this plate appearance that were thrown at that spot. Peralta’s able to just catch up with this one to stay alive, but hits it foul.
We now conclude these 7 straight four-seamers to David Peralta with a perfectly located high fastball right on the outside corner with just enough horizontal movement to get Peralta to whiff. 2 down, time for the final act.
Adam Jones steps to the plate and we see the return of the splitter. The pitch goes where Yates wants it on the corner and while Jones does still make contact, it’s a very low exit velo foul.
Very similar high fastball thrown to Jones as the one that got Peralta and it works here too. I do have to wonder if a younger circa 2014 Adam Jones would have been able to make contact with that, but I’m not here to focus on the inevitability of our progress towards death that so much of pro sports already reminds us of, so I’m not going to actually research that question.
For the crescendo of Yates’ performance he flashes a splitter so good Statcast actually classified it as a changeup. It drops like a fucking rock, Adam Jones looks like a fool, and the Padres win the game.
Conclusion
Not really sure if I have much of a thesis statement here, Kirby Yates is good. Is he actually the best reliever in baseball? Probably not, I’d give the honor to Josh Hader if I had to pick someone because of his monster strikeout rate and overpowering stuff. That shouldn’t rob Yates of what he’s done over this year (and much of last) though, and it’s probably *relieving* (geddit) for Padres fans to make their return to being a good team with a shut down closer who they can rely on to not blow leads.
Ultimately this was just a way to write a piece that tested out the baseball savant video feature and really got into the inner mechanics of a pitcher’s outing and the general outlook. I might do this to whatever reliever leads MLB in fWAR for May, provided that person isn’t also Kirby Yates, a real possibility. I’ve thought about doing something like this for a long time (about a year and a half) and with the help of Mike Petriello and the other Baseball Savant guys I finally did.
/cdn.vox-cdn.com/uploads/chorus_image/image/56874759/usa_today_10278006.0.jpg)
Arthur's Economics and Sabermetrics Blog 
