No mirage in Arizonaby Chris Jaffe
August 20, 2007
This year one of the big surprises in MLB is the Arizona Diamondbacks. They had some hard times over the last few years, averaging 94 losses from 2004-6. (OK fine, the 111-loss 2004 really throws everything of whack. But they haven’t been good). Currently, they’re battling the Mets for the best record in the senior circuit.
Many thought they could be this year’s breakout squad. Here at THT, for instance they were the staff’s consensus to take the division. They had a fantastic ace in Brandon Webb, had retained former Mr. Everything Randy Johnson, and had an organization littered with promising young talent.
But even those optimists have been surprised. Sure, they’re winning, but they ain’t going about it in a particularly healthy way. At the moment I write this, they have scored 18 fewer runs than they’ve allowed, 537-553. Hmmm…
OK, let’s go back to Sabermetrics 101. There’s supposed to be some sort of relationship between runs scored and runs allowed. You can use some variation on the old Pythagoras formula to figure out how many wins a team “should” have based on its record. And you know, teams that can’t score as many runs as they give up really shouldn’t be on pace to win 90 games. Differences between Real W/L and Pythag W/L are dismissed as mere chance. Heck, in B-ref’s expanded schedules, the site flatly categorizes any variation as luck.
Yet the D-backs this year have done quite the job mocking sabermetric orthodoxy. Not only have they floated over their ordained record, but they’ve teased people into thinking they’ll regress. After bursting out to a 46-35 start (while getting outscored all the while), they fell apart and dropped 13 of their next 17. “A-ha! Mathematical certainly will not be mocked or long!” Well, accept in this case that is. Since then the D-backs have won 21 of 26. Sure they’ve outscored their opponents in that stretch, but only 140-116. They’re still wildly exceeding expectations.
Speaking as a sample size of one, this is why I love this game. Just when you think you can figure it out with numbers, the ball starts to bounce the other way. Right when you are so sure in yourself and your conclusions, John Mabry starts outperforming Jeremy Giambi.
How the hell are they doing it?
The awesome thing about the D-backs is that when you look closer at the numbers, what’s going on makes sense. Not only is this not luck, but there is a very really shot they can keep on at this pace for the rest of the year.
Added bonus: they’re doing it by flipping around conventional wisdom on how you’re supposed to defy the mathematical gods. The single most agreed-upon way to evade Pythagoras and all his equations is to do extremely well in one-run games. The 1974 Padres for example, won nine more games than they were supposed to. They did it by going 31-16 in one-run games, while going 29-86 (!) in other contests.
Arizona is doing well in one-run games, not doubt about it. But they’re 27-16 record in those affairs is only slightly better than their overall record. It can explain why they’ve won a couple more games than they should have. But dag nabbit, they’ve won 10 more.
But there’s a reversal. The revealing factoid with Arizona isn’t their one-run games, but their record in blowouts. They are 13-22 in those contests, while 58-31 in all other games. OK, now that’s not supposed to happen. When you play like a 106-win team in close games, you’re not supposed to morph into a 102-loss team in the laughers. Better teams are supposed to be the most likely to blow others out and the hardest to get blown out. And it defies logic to say a team that’s won nearly two-thirds of its close games aren’t that good.
Let’s think this through. Another standard explanation for why a team would exceed their Pythagoras mark centers of the bullpen. If a team has a great ‘pen, they’ll exceed because they hold leads. Yeah, that makes sense. And Arizona most certainly has a great ‘pen, but that still doesn’t explain it. The great reliever corps explains records in close games. It sure as hell wouldn’t explain why they stink so badly in blowouts.
Here’s where you have to flip it all around. The secret does lie with the bullpen, but it’s not with the fantastic core. Or rather, it’s not just with the fantastic core.
Here’s how it works in Arizona. The game begins. If the starter pitches well, then it’s no problem. Wait until late, and let one of their dynamite relievers—Jose Velarde, Tony Pena, Brandon Lyon, and Doug Slaten all have ERAs under 3.00. And Juan Cruz is well above league average as well. Keep in mind they play in one of the game’s great hitters’ parks in a league that averages over 4.5 runs a game. As long as the starter has a quality outing, they’ll win a lot of games.
But what happens when the starter has a bad game? What happens when Micah Owings gets shelled for seven runs in four innings? Or Livian Hernandez receives one of his many poundings? Do you really want to waste one of the big five in such lowly leveraged situations? You might have to just to eat up some innings, but you’re better off going to the mop up men.
And that’s where the secret lies. The D-backs aren’t 10 games over their expected mark because their relievers are so fantastic. They’re ten games over because their bullpen is bipolar. Arizona’s mop-up men stink. I don’t mean they’re below average—mop-up men by definition are below average—but even by the standards of last man on the roster they are terrible. They can turn any deficient into an insurmountable one.
Lemme break it down. Here are the numbers on the year for the Big 5 relievers and the remainder:
G IP H R ER HR W K ERA RA ERA+ Big 5 249 250.7 204 93 77 21 94 225 2.76 3.34 167 Dregs 86 106.7 127 91 84 21 51 76 7.09 7.68 65
NL relievers as a whole have an ERA of 4.05 and ERA+ of 107. By any standards their dreg relievers ain’t getting it done.
And now look at their full roster. Their offense, in plain English, stinks. There’s no excuse to be 13th in runs scored when you play in that park. This team depends on its pitchers to keep them in the games, because God knows their offense won’t catch up for them. In general, their pitchers deliver. Even their worst starters and merely league average, and they have the greatest bullpen core this side of Ron Gardenhire. Thus they can dependably win an awful lot of close ones. But when they fall behind badly they have virtually no chance of recovery.
This does confirm one thought on Pythagoras deviation. Others, including myself, have argued that managers have an impact no this differential. You’ve got to give Bob Melvin a lot of credit here. He’s putting the right guys in when it matters and sticking in the bums when it doesn’t. Sure, in this situation the difference between good and bad it’s obvious who are the good’uns and the bad’uns, but there are plenty of times managers have used their relievers improperly. Besides, given how incredibly his core has done, I have to believe he’s done a good job communicating with his hurlers and making sure they’re as comfortable as they can be in their roles, allowing them to thrive at what they do.
In the past I noted that a team should exceed their pythag mark if they have a really consistent offense. (If Team A scores five runs each game while Team B alternates between scoring 10, and zero, A will win more games). This flips that idea around. Having a really inconsistent bullpen where the quality of multiple relievers swings this wildly also will have an impact. Along those lines, all other things being equal a team with a horrible fifth starter should overachieve while a team with stable starting pitching – such as this year’s Cubs or the 2005 Brewers – should win less than they’re projected to.
What it means for Arizona
If Arizona’s going to lessen the difference in the actual and supposed records, it’s more likely it’ll happen by their projected mark rising up to reality rather than reality falling down to Pythagoras. The main engine is their inability to have typically terrible end-of-the-bullpen relievers. The easiest way to improve a team is to have a noticeable hole that needs filling, and this one can be filled with replacement level pitching.
They could also see a drop off from their Fab Five core relievers. My hunch is that’ll happen. Any time that many guys are doing that good, someone’s liable to drop off some. But remember, they’re not the real engine of this.
They should win the division even if they are outscored this year. And the funny thing is, they could do pretty well in the playoffs. There’s no really strong team in the NL this year, and back-end pitching matters far less when you’re only playing 4 games a week. With Brandon Webb and that bullpen, some minimal offensive presence could propel them to their second pennant title.
History instructor by day, statnerd by night, Chris Jaffe leads one of the most exciting double lives imaginable; with the exception of every other double life possible to imagine. Despite his lack of comic-book-hero-worthiness, Chris enjoys farting around with this stuff. His new book, Evaluating Baseball's Managers is available for order. Chris welcomes responses to his articles via e-mail. Oh, and now he's on twitter.