The Season’s Biggest Upset

The season’s most lopsided game was a Clayton Kershaw start. That shouldn’t surprise you — the Dodgers were good, and Clayton Kershaw was great, and the Dodgers with Clayton Kershaw were fantastic. As a matter of fact, five of the season’s 10 most lopsided games were Clayton Kershaw starts. His opponents in those games: Bud Norris, Jered Weaver, Brandon Finnegan, Chad Bettis, and Tom Koehler. Four of those games, the Dodgers won. They lost the Kershaw/Koehler game. It was nearly the season’s biggest upset.

For every game all year long, we publish pregame win probabilities. At first, they’re based on general team projections, and the starting pitchers. Then they update to account for the actual starting lineups. Every calculation includes a home-field-advantage boost of four percentage points. This year, there were 58 games in which the favorite was given at least a 70% chance of winning. Our math predicted 42 wins. In reality, there were 43 wins. The odds work pretty well, provided you don’t take them too seriously. They’re wonderful estimates.

On April 26, Koehler and the Marlins went into Los Angeles and beat Kershaw and the Dodgers. Before the game, the Dodgers’ chances of winning were 74.3%. On July 22, Zach Eflin and the Phillies went into Pittsburgh and beat Gerrit Cole and the Pirates. Before the game, the Pirates’ chances of winning were 74.4%. There is no meaningful difference between these numbers. The calculation error is far greater than one-tenth of one percentage point. But, a difference is a difference. A leader is a leader. I have no choice but to designate that game on July 22 as the biggest upset of 2016.

When I looked up this information, I wasn’t expecting to end up with a Pirates game. They weren’t so good to be dominant, nor were they so bad to be dominated. They were something like average, so they were off of my radar. But, I don’t know why the odds calculations would’ve bugged out. The season’s biggest upset came in the season’s ninth-most lopsided game. And this was as much about the pitchers as it was about the teams. The next day, Tyler Glasnow opposed Aaron Nola. The Pirates’ chances of winning were 62.1%. The day after that, Jameson Taillon opposed Vincent Velasquez. The Pirates’ chances of winning were 61.1%. The game on July 22 was so lopsided because of Gerrit Cole. And, crucially, the game on July 22 was so lopsided because of Zach Eflin.

Cole, to be fair, had recently come off the disabled list. The projections didn’t know that, but in the moment, they projected a rest-of-season ERA around 3.23. As for Eflin? He was projected for a rest-of-season ERA around 5.13. It’s true that, over 11 starts in Triple-A, Eflin had kept runs off the board. But the year before, in Double-A, Eflin barely struck anyone out. This season, in the majors, 181 pitchers threw at least 50 innings as a starter. Eflin ranked in the 10th percentile in ground-ball rate, and he ranked in the first (lowest!) percentile in strikeout rate. He ranked in the 13th percentile in ERA-, he ranked in the eighth percentile in FIP-, and he ranked in the third percentile in xFIP-. Zach Eflin, as a rookie, was not very good. In his three starts after the game on July 22, he allowed 20 runs in 13 innings.

On July 22, he allowed zero runs in nine innings. The shutout was, of course, the only one of Eflin’s career so far. Shutouts these days are incredibly rare, and David Price, Max Scherzer, and Justin Verlander just combined for none of them. It was Eflin’s second complete game, but the first one came against the Braves, which hardly counts. Pitching against the first-half Braves was like pitching in some transition stage between Triple-A and the majors. The Pirates presented Eflin with a major-league opponent. He retired 25 of the 29 that he faced.

As I dive into the game itself, I recognize that there’s little in the way of suspense. You know the Pirates were supposed to win, and you know the Phillies actually won. They won 4-0, and Eflin went the distance. Right away, though, Cole tried to set a certain tone. He struck out the first batter on three pitches. He struck out the second batter on three pitches. Then he drilled the third batter, and Maikel Franco ultimately wasn’t able to finish the game.

Franco didn’t have a great season, but the Phillies’ lineup was terrible, and Franco was hitting third. When Cole effectively knocked him out, the Pirates’ chances of winning might’ve improved even more. At least, after they made sure the Phillies didn’t take advantage by pushing across a run. The Phillies were the underdogs, and then Franco was physically compromised. Things weren’t shaping up very well.

But then Eflin went to work. Say this for Eflin — he’s not some generic soft-tosser. He can pump his fastball into the mid-90s, and against the Pirates, he threw consistent strikes. After some of those strikes, Eflin thanked his lucky stars for the existence of the warning track.

The Pirates threatened to take Eflin deep, and they did so multiple times, but the balls would be just barely mis-hit. Leading off the second, John Jaso did manage to achieve a double. You’d think, at that point, the Pirates would have a pretty good chance of taking the early lead. But, as long as we’re talking about early leads:

Jaso went and took off, and Eflin had him picked. It’s true that Eflin was a rookie. It’s true that he could’ve been caught in a balk. But, last season, John Jaso stole zero bases, while getting caught four times. It’s easy to say now that Jaso shouldn’t have bothered. That’s results-based commentary, and there was some chance Jaso would get in safe. Yet the Pirates wasted a chance to score early because John Jaso tried to steal third base before the pitcher threw a pitch. Against someone who allows as much contact as Eflin does, I don’t know what the point of that was.

The game remained scoreless into the sixth. The longer the Phillies could keep things tied, the better their chances would become. As the sample of remaining baseball gets smaller, the chances increase that the game could turn on something fluke-y and stupid. Odubel Herrera led off.

A sure out became a sure hit. Not because anyone did anything wrong, and not because anyone did anything right. It’s just, early on in the game’s development, the fathers of baseball decided the bases should be raised, instead of buried and level. Three of the four bases, anyway. Herrera found one of the bases, and then there’s no telling where the baseball’s going to go. I don’t know the historical batting average on batted balls that bounce off of bases, but I’m going to guess it’s about 1.000.

Herrera reaching to lead off was bad for the Pirates. Things got worse when Cole hit Andres Blanco in an 0-and-2 count.

The fastball didn’t miss by very much, and it only slightly grazed Blanco in the leg. It was one of those hit-by-pitches where the catcher still catches the baseball cleanly. As Blanco turned his front leg inward, he also thrust it closer to the plate, increasing the likelihood of a free base. Blanco was only in the game because Cole had knocked Franco out with a different hit-by-pitch. An unlucky single and an unlucky hit batter set the table. The Phillies’ run expectancy in the inning at that point was 1.40 runs. They settled for one run, and it was all that they’d need.

In the bottom half of the sixth, the Pirates threatened to go deep once more. And after that attempt failed, they curled up and died. Just one Pirates batter reached in the final six innings. The Phillies added on in the seventh against Arquimedes Caminero, and the final score was set by a two-run Cameron Rupp homer in the ninth. The Pirates went down in the ninth on three consecutive ground outs. Eflin wound up with one of the lowest grounder rates in the league.

Shutouts are funny. Like anything good, they’re selective for baseball’s better pitchers, but every so often a pitcher from the lower class breaks through out of nowhere. On April 30, Wade Miley shut out the Royals. On May 14, Matt Andriese shut out the A’s. On May 30, Jeff Locke shut out the Marlins. On June 19, Jered Weaver shut out the A’s. On August 19, Kendall Graveman shut out the White Sox. And on July 22, Zach Eflin shut out the Pirates. We have some updated Steamer projections on the site. For 2017, Eflin’s projected for an ERA of 4.86. That puts him three-hundredths of a point behind Kevin Correia, but, on the plus side, it puts him four-hundredths of a point in front of Roy Halladay.