Reexamining the Mets, or When One Game Might Mean Something

Anyone who has ever studied statistics generally, and sabermetrics or other analytics specifically, will have at least a passing familiarity with the idea of sample sizes. I’m not a sabermetrician – nor do I play one on TV – but you can’t be a good trial lawyer without at least a decent understanding of statistics. And regardless of the statistic, it’s generally true that one game isn’t really all that useful. I mean, it’s useful in the sense that being entertained is useful, but it’s usually not useful in the sense that you can determine talent levels from it.

Enter the Nationals and Mets. They are both, in theory, Major League Baseball teams. They both play on the East Coast. And on July 31, as major league teams are wont to do, they played a baseball game. And when they were done, this had happened.

The final: Nationals 25, Mets 4
The goat: Abner Doubleday for inventing baseball
The stat: Worst loss Mets history
The streak: L1
The record: 44-60
The 162-game pace: 69-93
The emoji: ?????

— Anthony DiComo (@AnthonyDiComo) August 1, 2018

And as you might expect for a team that scored 25 runs, a lot went right for the Nationals. Daniel Murphy had three hits, hit two homers, and drove in six, yet had just a 0.10 WPA, because that’s what happens when your team scores 25 times. Tanner Roark, the Nationals’ starting pitcher, tossed seven innings of one-run ball and had two hits, including a bases-clearing double. Matt Adams and Mark Reynolds both homered for Washington, and neither of them even started the game. The Nationals walked eight times, had a hit batsman, racked up 26 hits and, as a team, hit .520/.593/1.000, which, in case you were wondering, comes out to a 311 wRC+ and a .646 wOBA. In other words, the Mets’ pitching staff allowed eleven more batters to reach base (35) than they got outs (24).

Now, to be fair, the term “pitching staff” should be used loosely here. Jose Reyes “pitched” an inning of garbage time, and he threw whatever this was to a decidedly unimpressed Mark Reynolds:

RECORD NIGHT: #Nats score a franchise-record 25 runs in their 25-4 rout of the Mets! Mark Reynolds supported the cause with this homer in the 8th! pic.twitter.com/EN6C8nK6eP

— Nationals on MASN (@masnNationals) August 1, 2018

But Reyes was only on the mound because the Mets’ pitching staff had, to that point, surrendered 19 runs on 21 hits before he entered the game. And that’s why this matters. The Mets declined to sell at the deadline. Yes, they dealt Asdrubal Cabrera and Jeurys Familia, but they held onto Jose Bautista, Reyes, and all four of Noah Syndergaard, Jacob deGrom, Steven Matz, and Zack Wheeler, evidently because they believe they can contend next year on the back of their starting pitching. But we’ve been down this road before, and while deGrom has been fantastic and Thor has shown spurts of brilliance when healthy, injuries and ineffectiveness have plagued the rotation overall since 2016.

And it’s true – deGrom didn’t pitch on July 31. But Matz did – and it wasn’t pretty. He lasted just two-thirds of an inning and gave up seven earned runs. It’s easy to look at his 1.66 FIP for the game and say he got unlucky. But Statcast tells a different story. This was Matz’s pitch location.

The Nationals did what you might expect with pitches in those locations. Of the eight batters to put the ball in play against Steven Matz, all of them hit the ball with an exit velocity of 95 miles per hour or higher, and five eclipsed 100 miles per hour. Granted Matz was, based on each individual batted ball’s hit probability, unlucky on a few of these hits, but you can’t give up several 100-mph rockets in the big leagues and not get burned.

And the Mets’ bullpen wasn’t much better. Jacob Rhame isn’t in the team’s A-group of relievers, but he’s given up six homers in 18 innings this year, two on Tuesday. Tim Peterson, whose performance was almost as bad as Rhame’s, got sent to Triple-A after he gave up his seventh homer of the year on Tuesday in just 20.2 innings pitched. Both Rhame and Peterson have FIPs over 6. Both are younger than 28, and the Mets are counting on both to be contributors to the 2019 bullpen. Neither has yet shown they’re big-league ready.

All of this is a microcosm for why, if we zoom out and look at the Mets with an objective, critical eye, this is not a contender’s pitching staff even with deGrom and Syndergaard in the fold. With Familia traded, the Mets have exactly three pitchers on their roster worth more than 1 WAR this year: deGrom, Syndergaard, and Wheeler. Seth Lugo is fourth at 1 WAR. After Matz, who at 0.5 WAR is fifth on the team even after Tuesday’s debacle, the Mets have only one pitcher even above replacement level — Drew Smith, who has tossed five innings this year and is at 0.1.

Compare that to the Nationals, who have 13 pitchers with positive WAR contributions, or the Phillies with 14, or the Braves with 12, and you start to see a pretty big talent disparity. The Phillies’ most-used relievers are Seranthony Dominguez, Victor Arano, and Tommy Hunter, who have combined for 2.9 WAR across 117.1 innings. The Mets’ top three of Robert Gsellman, Paul Sewald, and Lugo, who have exactly 1 WAR between them, all attributable to Lugo. In other words, with Lugo in the starting rotation, the Mets’ two best relief pitchers have both been replacement level. Overall, the Mets have the second-worst relief corps in the majors, at -1.5 WAR, worse than the -1.2 WAR posted by the 2017 Tigers bullpen (last in MLB that year) put up in the whole season. Mets relievers have a worse FIP and xFIP than the last-place Royals do this year.

But the problem isn’t just the bullpen. The Mets’ rotation is good, but it’s not deep. Even with deGrom and Syndergaard, the team ranks just eighth in rotation WAR. And that’s how a team with Jacob deGrom and Noah Syndergaard ends up with the 20th-best pitching staff in baseball by WAR, closer to the 29th-place Reds than they are to the fourth-place Phillies.

The goal of this is not to say just that the Mets aren’t a good team. Instead, there’s this idea that, thanks to the top of their rotation, the Mets are a potential pitching juggernaut if they can just stay healthy. But even if deGrom and Syndergaard were to combine for 10 WAR next year by themselves, this would still not be a good pitching staff. Wheeler is injury prone, and Matz, despite flashes, hasn’t shown he’s anything more than an inconsistent back-end starter.

The bullpen is a problem that’s harder to fix. Aside from Lugo, it’s hard to see anyone in that group that could be counted on as a contributor for a 2019 contender’s bullpen. And while contenders can succeed despite their bullpens, it’s hard to see that approach applying to the Mets. The Indians, for example, have managed to overcome a bad bullpen this year, but they also have Corey Kluber and Trevor Bauer healthy, and two of the best players in baseball in Francisco Lindor and Jose Ramirez. The Mets have an injury-prone Yoenis Cespedes, who’s already out for the first half of next year, and the unsettled potential of Brandon Nimmo and Michael Conforto. And it’s hard to see the Wilpons shelling out for a whole new bullpen — this isn’t a problem you fix by signing a single closer.

So statistically, one game doesn’t mean a whole lot in a 162-game season. But Tuesday’s game might have meant something. The Mets’ front office kept the team together because they thought they had a good pitching staff. But good pitching staffs don’t give up 25 runs. July 31 exposed the Mets’ pitching staff. Despite its stars, it’s a team weakness and perhaps always has been. And if that’s so, the Mets are a lot farther from contention than they realize.