keyboard_arrow_uptop

This turned out to be pretty good season for prediction systems in general and our PECOTA in particular. PECOTA thought that Boston, Cleveland, and Houston would win the American League divisions, with Los Angeles and Chicago taking the National League West and Central, respectively. It had the Mets with 86 wins, the Nationals with 85, and the Giants with 84, so we were close with our NL East champion, if missing badly on the Wild Cards.

We caught a lot of grief for suggesting that the Dodgers would win nine more games than the Cubs; the gap is currently 12. Similarly, our prediction that the Twins would have the second-best record in the American League Central was met with derision, and they’ll be playing in the postseason for the first time since 2010. All in all, a pretty good season for PECOTA.

The Mets and Giants projections, of course, were whiffs, but they’re whiffs just about everybody else had, too. And here’s another one: We projected the Padres with the worst record in the majors, at 68-94. Again, this was consistent with just about every prediction out there. And it wasn’t just algorithms like PECOTA. Human predictors were in accord. Of the 62 Baseball Prospectus staffers who participated in our pre-season predictions, all but two had the Padres last in the National League West. There was no team in the majors for which there was greater unanimity for a prediction in the cellar.

They won’t finish with the worst record in the NL West; that’ll be the Giants train wreck. Nor will they have the worst record in baseball; the Tigers, Giants, Phillies, and White Sox have already clinched a worse finish. The Mets and Reds are likely to trail the Padres, too. The Braves, A’s, Pirates, and Marlins are within reach. This is not to say that they’re a good team. Entering play Wednesday, San Diego had a 70-88 record and trailed the Dodgers by 31 games. They were 14 games out of the running for the Wild Card. They’ll get a very high pick in the next June’s amateur draft.

But this is a team that had become kind of a joke. Their owner called them “miserable failures." The GM was suspended for failing to disclose traded players’ injury histories. He strip-mined the club’s farm system, making a series of trades prior to the 2015 season that failed spectacularly. They picked up three players from the very low minors in the Rule 5 draft and kept them on the roster all year. Their Opening Day starting pitcher, Jhoulys Chacin, had been released by the Rockies after the 2015 season and had compiled exactly 0.0 WARP over the prior five seasons combined.

So the fact that the Padres were merely bad rather than execrable came as a bit of a surprise. How did they do it? Before I get to that, let me make two points. First, if you’re a Padres fan, you probably know all of this. I’m guessing most of you (like me) haven’t been paying that much attention. Second, for a great take on the club, listen to this DFA Podcast with Bryan Grosnick and R.J. Anderson. (And while you’re there, subscribe! It’s a really good podcast.)

***

Austin Hedges, framer extraordinaire. Hedges put up consistently solid framing numbers in the minors, and moved into the starting role this year. He’s second in the majors with +23.8 framing runs, which combined with strong blocking and throwing, gives him 29.3 FRAA, easily the most among major-league catchers. He’s not much of a hitter, but his glove has made him the Padres’ MVP according to WARP, with 3.8.

They Got Legs. The Padres are second in the majors in BRR. They’ve been the best at advancing on grounders and among the best at advancing on fly balls and hits. Cory Spangenberg ranks eighth in the majors with 5.4 BRR to lead the way. The Padres don’t get on base a lot (last in the majors with a .299 OBP), but once they’re there, they scamper around better than almost anybody.

The New Outfielders. The Padres have five players with at least 2.0 WARP. Hedges and Spangenberg are two. Wil Myers, you probably figured him as well. The others? One, center fielder Manuel Margot, turns 23 today (happy birthday!). He is familiar to the prospects crowd; we ranked him 18th in all of baseball before this season and 14th before the 2016 season, so his .270/.321/.421 rookie season isn’t a surprise.

The other, left fielder Jose Pirela, is more out of nowhere. The 27-year-old leads the team with a .302 TAv but has been limited to 83 games, partly because he was at Triple-A El Paso until June, partly because a finger injury ended his season earlier this month. He compiled a .568 OPS over 59 major-league games spread over three seasons prior to 2017, so his .288/.347/.490 breakout—he leads the team in all three slash stats (min. 250 PA)—is a pleasant (if partially .343 BABIP-fueled) surprise.

(The third outfielder, Hunter Renfroe, has been okay—.275 TAv, 1.6 WARP—but has fallen short of Margot and Pirela.)

The All-Star. The Padres’ representative in Miami was not face-of-the-franchise Myers; it was left-handed reliever Brad Hand. Hand, whom the Padres claimed off waivers after he was designated for assignment by the Marlins last April, led the majors in appearances last year and was elevated to the closer role after Brandon Maurer was traded to the Royals in July. His 2.21 ERA is the eighth-lowest among National League relievers with at least 50 innings pitched, and it accompanies a 3.16 DRA and 3.09 FIP. He’s arbitration eligible next year, and a free agent in 2020, so it was a bit of a surprise that he wasn’t dealt at the trade deadline, but his 5.1 K/BB ratio and 32 percent whiff rate give San Diego a reliable back end of the bullpen.

No-Name Pitchers. Six Padres pitchers have generated at least 1.0 WARP. Of them, two—Dinelson Lamet and Craig Stammen—didn’t even warrant a mention in the 2017 Baseball Prospectus Annual. A third, Kirby Yates, got only a one-sentence Lineout in the 2016 Annual. (And yes, if you were wondering, Lamet received a coveted Vogelsong Award in May and Stammen got one in August.) Lamet has a 4.57 ERA and 3.77 DRA in 21 starts, striking out 139 in 114 1/3 innings. Stammen has a 3.10 ERA and 3.53 DRA while leading the team with 78 1/3 relief innings. Yates, picked up off waivers from the Angels in April, has struck out nearly 40 percent of the batters he’s faced in 54 2/3 relief innings, establishing himself as the team’s eighth-inning setup man.

***

Does this mean the Padres are on the verge of contention? Well, no. They’re going to lose 90 games, more or less, this year, and they play in the same division as two, and likely three, playoff teams. They’re above average only at catching the ball (fourth in the NL at park-adjusted defensive efficiency), and they’re 13th in TAv, 10th in reliever DRA, and 14th in starter DRA. Their third-order winning percentage (based on their underlying statistics and quality of opponent) is six games worse than their actual percentage, which suggests they’ve been lucky to win as many games as they have.

But they haven’t been the worst, or the second-worst, or even top-six worst team in the game. That may be suboptimal in terms of positioning themselves for the amateur draft, but it’s a lot better than PECOTA expected.

You need to be logged in to comment. Login or Subscribe
jssharo
9/28
Regarding tanking to get high draft picks... my impression has been that unless there is a Bryce Harper type beast out there (and there is none this year), the odds of getting a good or even franchise player are pretty comparable whether you have like the 1st or 7th pick. So maybe the Padres fans got to have a little more to cheer about this year at no significant cost to their team's future?
mainsr
9/28
There's no question that the MLB draft is far more of a crapshoot than those of the NFL, NBA, or even the NHL. But the data do suggest that a no. 1 pick returns greater value than a no. 2, which returns greater value than a no. 3, etc. However, as you suggest, the data are really noisy, and when Nate Silver did his seminal research on this issue 12 years ago (link below), he defined the top tier as precisely the top seven that you suggest. So yes, maybe putting out a merely bad club rather than a terrible one didn't hurt the team's long-term prospects. But if the idea was to appease the fanbase, it didn't work well. The Padres drew 2,629 fewer fans to PetCo this year than in 2016, the fourth-steepest decline in the majors. http://www.baseballprospectus.com/article.php?articleid=4368
jfranco77
9/28
Most drafts, the #1 pick comes with a slot value higher than any player will get. So what it really gets you is a better pick 31 (or 35 or 42 or whatever). I think that matters more than the guy you get at 1.
mainsr
9/28
True, good point. The research I recall looked at excess value--i.e., WAR generated above salary--and no. 1 picks were the best ROI, even considering the signing bonuses they get. Getting a good, say, 31 pick is valuable, but once you get past the first round there's a pretty big crapshoot factor.
newsense
9/28
You really can't compare PECOTA projections with real outcomes because PECOTA doesn't account for trades. If PECOTA correctly identifies a team as good the team will outperform its projection because it will trade prospects to become better this year. The opposite is true for bad teams. This is why PECOTA makes projections instead of "predictions ".
mainsr
9/28
Good point about projections vs. predictions, but there are a whole lot of Royals and Orioles fans from this decade who would beg to differ on whether it's fair to compare PECOTA to outcomes. In the case of the Padres, trades weren't really a factor this season. Nor were they for the spectacular faceplant that was the Giants. But your point's fair.
newsense
9/28
The PECOTA projection for the Padres is only off by 4 games. They only look successful compared to the Giants
mainsr
9/28
Yes, but PECOTA expected them to have the worst record in baseball. As I type this seven teams--SFN, DET, PHI, CHA, CIN, NYN--are worse.
newsense
9/29
That's because of my original point, those teams collapsed (mostly) because they traded away many of their good players
mainsr
9/29
Dammit newsense, I live in the East, and I've got a lot of things going on this weekend, and you're making the SDP-SFG series appointment baseball for me (fortunately only one night game)
dethwurm
9/28
This is an interesting rundown, but I really feel like it should have been mentioned that the Padres have, by an *enormous* margin, the worst run differential in baseball (-204, next-worst is the Tigers with -156). They've got an awfully good case to be the worst or second-worst team in baseball this year (the dessicated corpse of the Giants dynasty somehow actually has worse 2nd and 3rd order Winning Percentages despite a huge advantage in actual runs scored, though only by tenths of a percent "ahead" of the Padres, who are second-worst), despite the relatively non-terrible record. Some of that is relative overperformance in one-run games (18-18 vs. overall .440 Winning Percentage), but that certainly doesn't explain all of it (noteworthy: 36 one-run games is actually fewer than most teams). I honestly haven't watched them at all so I'm really curious how they've managed this - blowout losses and a lot of 2-run wins?
mainsr
9/28
Great thought, Pat! I checked. Yep, you called it, blowouts. Through last night, the Padres played 38 games that were decided by six runs or more. That's tied with KC and ARI for the eighth most in the majors. Their record in those games was 6-32, far and away the worst in the majors--Blue Jays at 8-20 were second worst. No team lost as many or won as few games decided by six runs.
CrashD
9/29
But hey you finally got the Orioles right. We owe you a Natty Boh!
mainsr
9/30
Natty Boh and the O's...You have a deal, provided that you buy the third round after I buy the second.
FlemSnopes
10/03
Why did PECOTA get the Nats wrong by 12 games (a major discrepancy)?
mainsr
10/03
A couple things. First, I looked at every Nats player who had 2.9 or more WARP this year. There were ten. Of them, the only ones who didn't beat their projections were Harper and Turner. Some of the discrepancies were pretty large: Scherzer 3.9, Murphy 3.8, Rendon 3.7, Gonzalez 3.2, Roark 3.0, Taylor 3.0. There were disappointments, too, of course; Adam Eaton didn't quite get to the 3.4 WARP that PECOTA envisioned. But overall, a lot of good players were excellent. Second, and this is a little circular, the NL East was dogmeat. There were eight NL teams that had 77 or fewer wins. The Nats got to play 19 games each against half of 'em. The Mets, of course, were way worse than projected, but so were the Phils and Braves. So that helped the Nats. They played 120 games against teams with losing records, the most in the league.