Have Faith…in Regression to the Mean

.flickr-photo { border: solid 2px #000000; }
.flickr-yourcomment { }
.flickr-frame { text-align: left; padding: 3px; }
.flickr-caption { font-size: 0.8em; margin-top: 0px; }



The Red Seat, originally uploaded by Starving Photographer.

If I’d told you before the season that 21 games into the season that Buchholz would be our best starter – easily, that Darnell McDonald would have played more games that Ellsbury and be second in OPS to Varitek…on the team, what would you have said? What if I also told you that we would have scored 95 runs, but given up 113? That our vaunted defense would be -14 in defensive runs saved, tied for worst in the league with the Brewers? That our last six wins were one run affairs, against Baltimore, Texas and Toronto? That they’d leave us still one game under .500?

You’d figure we’d be further than 3 back from the Yankees, right? And that the Yankees would be in first place, not the Rays, correct?

But that’s why they play the games. This game is weird. I mean, how else do you explain Beckett’s identically terrible numbers this April to last?

In the wake of Buchholz’ gem tonight, you’d think I’d be leading the cheerleading squad. Not so much. We’ve stacked the deck against us, what with all this playing like shit. FanGraphs says we’ll finish in third, and frankly there are nights when it’s been tough to argue the point.

Still, the most important thing to remember is – as we talked about last week – is that the mean catches up with everyone in the end. For better, and for worse. So just like these folks with The Greater Good (and incidentally, if you haven’t seen that movie, you need to), I ask you to repeat after me: Regression to the Mean. Looking for evidence of that, I took a quick look at our BABIP figures for the season. And found it.

You know how hot JV’s been? Well, it just so happens that his batting average on balls in play is .400. Meaning that when he makes contact, he’s hitting about .110 better than is normal. Or in layman’s terms, he’s been pretty lucky, and is likely to get worse. Which is bad for us.

Good for us, however, are the BABIP’s posted by two of our coldest hitters: Drew and Ortiz. As it happens, they are 2 and 3 for lowest figures on the team Drew at .227 and .233 respectively. Meaning that they are likely to get better. Which is good for us.

As you might suspect, with the pitching, defense and offense all pretty awful in the early going, a regression to the mean would be just what the doctor ordered for us.

Will it be enough to make the playoffs? Who knows. It’s way too early to be writing off our chances on account of a six game deficit, but there’s no getting around the fact that the Rays and Yankees are good. Really good.

But have faith, my friends. We’ve had the second worst starting ERA in the league, two thirds of our starting outfield hasn’t seen the field in two weeks, and Beltre’s got close to one third of last year’s errors one eigth into the season. These too shall pass. It’s a long summer, and all we need is for our guys to start being our guys. Which we’re seeing signs of already.

In the meantime, buckle up and try and enjoy the ride.

Listening to the Sox in New England: Red Sox Affiliates, Mapped

New England Red Sox Affiliates
New England Red Sox Affiliates

Like a lot of people, I catch a large number of the Sox games from the car each summer. Whether it’s visiting family, a long weekend away, or traveling for work, few things pass the time more effectively than a ballgame. At least for me. But nothing is more frustrating than getting to a critical part of the ballgame and losing the signal and then the game because you can’t find the local affiliate.

The last time we had this problem, the fiancée found and bookmarked the Wikipedia list of Red Sox affiliates here. This weekend found us traversing the green and white mountains, however, where neither of us knew from the list what cities or towns were actually close enough to pick up. In spite of being wicked clevah, I’d never considered that this was a problem with a solution. Fortunately, my fiancée is quite a bit brighter than yours truly and suggested that I map the affiliates.

So I did. I’ll see if I can tweak this a bit going forward to improve the usability, but in the meantime, if you find yourself in New England short of a Red Sox radio affiliate, we hope the above is of use to you.

Two Men Enter, One Man Leaves: Buchholz versus Wakefield

.flickr-photo { border: solid 2px #000000; }
.flickr-yourcomment { }
.flickr-frame { text-align: left; padding: 3px; }
.flickr-caption { font-size: 0.8em; margin-top: 0px; }

Peter Abraham is right to be asking the question, because even the average fan at the game last night was asking the same question: who goes to the pen when Matsuzaka returns? My suspicion is that his answer – Wakefield – is also correct.

And to his credit, he looked at the issue from a number of angles. I wish, however, that he’d gone just a bit further with metrics he used. So let’s do that here.

Buchholz easily wins if ERA’s the metric, 1.80 to 6.38, but as Abraham noted ERA doesn’t tell the whole story. So let’s look at FIP, which normalizes the ERA to account for variances in league, defense and so on. Buchholz’ FIP is 4.51, which certainly feels more like how he’s pitched. Wakefield, meanwhile, is at 4.54. Translated, this indicates that Buchholz may have gotten a bit lucky, Wake a bit unlucky. And that even adjusting for luck, Buchholz has pitched more effectively. The ratios are a bit more mixed: Buch has the edge in K/9, 7.20 to 5.40, but Wake is walking fewer at 3.44 BB/9 to 5.40.

One could make the argument, I suppose, that Wakefield has been profoundly unlucky when it comes to BABIP. In 2010 hitters are averaging .340 on balls in play. Given that his career average is .282, they’re hitting about fifty points better than they should.

But while that’s likely to regress to the mean, resulting in a statistical improvement, it’s not likely enough to make him a better choice than Buchholz. According to the stats, anyway.

Because as much as it might be easy to run a club by the numbers, Francona – bless him – knows better than that. There are a myriad of factors that will go into his decision, the numbers being but one of them. I think he’ll end up with Wake going to the pen, but it’s not as if the situation is cut and dry so there’s room for disagreement.

All that I wanted to do here was add some context to Abraham’s statistics argument, which seemed to me to be lacking.

WTF?

.flickr-photo { border: solid 2px #000000; }
.flickr-yourcomment { }
.flickr-frame { text-align: left; padding: 3px; }
.flickr-caption { font-size: 0.8em; margin-top: 0px; }



opening day, originally uploaded by sogrady.

You’re thinking it, I’m thinking it, and you can be damn sure Tito’s thinking it. If I’d known that the only home game we would have won to date would be the one I attended – that’d be Opening Day, for those who haven’t been keeping up with current events – I probably would have made an effort to get to more games.

But seriously, who would – could – have called this? 4-9, 6 games back of the division leading Rays. I sure didn’t. To try to ground the discussion, let’s look at where we are, and where we might expect grounds for improvement.

What’s Gone Wrong?

In a word, everything. And no, that’s not hyperbole. There is quite literally nothing we’re doing right at the moment.

The offense? Being 17th in the league in average is bad enough, being 20th in OBP is worse. Far worse. When the bright spot in your offense is a 9th place finish in slugging percentage, you’ve got problems. Frankly I was surprised we were only 15th in the league in strikeout percentage; Drew and Ortiz are between them striking out in 42.5% of their at bats. As an aside, I do find it interesting that everyone’s written off Ortiz while it’s just a slump for Drew, just as I find it curious that Lester’s just not good early while Buchholz is again being popularly consigned to the bullpen or someone else’s roster. The psychology of player evaluation is a really fascinating phenomenon. But we’ve got bigger problems to look at at the moment.

So the offense is not good. How about the vaunted defense? If anything, it’s worse. We’re 27th in Defensive Runs Saved, 18th in Fielding Percentage and we’re in such desperate shape at throwing out runners that there’s already talk of bringing up Mark Wagner, whose CHONE projected line is .232/.298/.341.

But at least the pitching is good, right? I wish. Our ERA could be worse, I guess, at 18th in the league, and actually our 25th ranking in FIP suggests that we’re actually lucky it isn’t worse. Not surprisingly, given those numbers, we’re bad at striking people out (27th in K/9) but much better at walking them: 10th best team in the majors at issuing the free pass. In case it’s not clear, being good at walking people is not a desirable skill.

Nor was there, as I had hoped before looking, any indications that the above numbers, both offensively and pitching-wise, are flukes. Our hitters’ BABIP is .287, and our pitchers’ is .279. Meaning that we have neither been exceptionally lucky or unlucky.

We are what we are, in other words. Except that we’re not.

What’s Likely to Go Right?

All of the above. The most plate appearances anyone on the club has is Pedroia at 57. That’s not as small a sample as the two games that had the writers penning Papi’s obit, but it’s statistically not significant.

It’s early. I know that’s hard to believe when we’re a few weeks into the season, already back by six games and with our offseason plan looking as intelligent as real estate investing circa 2010.

But before you take a leap from the Zakim, consider the following:

  1. Our hitters and pitchers alike will regress to the mean. For better and for worse. Pedroia’s sadly not going to put up a 1.159 OPS for the year, but neither is Drew going to put up an OBP of .233, V-Mart a SLG of .367, or Youk an average of .238. These things will fix themselves over time.
  2. No one likes to make excuses because of injuries, but remember that that two thirds of our starting outfielders this weekend were bench players. So when you see Hermida butchering balls in left, remember that he is not the starting left fielder, Ells is. And thankfully, he’ll be back because Beltre didn’t kill him. As will Cameron, after his current senior ailment – kidney stones – remedy themselves. Will that fix Scutaro’s jitters or V-Mart’s tendency to sail throws “just a bit outside?” Nope. But again: they’ll regress to the mean. Although in V-Mart’s case that’s not good news.
  3. A couple of folks have bitched that this is all the front office’s doing; if only we signed Jason Bay, we’d be right there with the Rays and Yankees. Setting aside the question of how a single player could fix all that has gone wrong thus far, there is the problem with the numbers. Namely, Bay’s. Thus far in a Mets uniform, he’s putting up a .217/.321/.283 line, while leading both leagues in strikeouts with 18 (narrowly edging our own Drew). Our left-fielders, meanwhile, have put up an unimpressive and still superior .240/.255/.400. So not only is Jason Bay not walking through that door, it probably wouldn’t help much if he did. Like everyone else, he’ll regress to the mean – which in his case means he’ll get a lot better – but in the early going, it doesn’t appear as if he’d be a difference maker.

What’s Not Likely to Improve?

I am worried with a capital W about the pen. The starters, I think, will ultimately be fine. Beckett’s been better, Lester’s history says he’ll be better, Lackey just had a bad start today, and between Wake, Buch and Matsuzaka – throwing well in Pawtucket, from reports – I feel pretty good. One through five (or six), we’ll have a chance to win most days, however much it doesn’t feel that way right now.

Likewise, our offense will hit. It’s looked brutal in the early going, but it is always does when 70% or 80% of your lineup isn’t hitting. Moreover, I think the front office will be aggressive if it looks like that’s a problem, and adding offense in season is always easier than adding pitching.

But we may be forced to give something away to get some help in relief, because there’s no real help available on the farm. Tazawa’s out with Tommy John. Richardson isn’t exactly lighting it up with a 1.80 WHIP at Pawtucket, and Kelly – for all of his poise – was born in 1989 (though I find his innings limits intriguing). I’ve been worried about our pen since the offseason, not least because our PECOTA projections were terrifying. And yes, I’m aware that PECOTA’s had its issues this offseason.

ESPN’s Jeremy Lundblad has the best breakdown of the issues out of our pen that I’ve seen. You really should read it, but the short version for the link averse? Delcarmen’s lost about three miles an hour off his fastball since 2008, and his usage reflects that. Ramirez Uno’s K rate is in sharp decline, and his walk rate is up. Also not good.

Oki’s still great versus lefties, but he’s become mortal versus righties. Which probably explains why he’s given up the 8th inning to Bard. And speaking of, while Lundblad’s not particularly worried about the young fireballer, I am. His K rate is down sharply in the early going, never a good sign, although that could just be a blip. His HR rate is higher as well. Like Keith Law, I’ve struggled to understand how people who watch him pitch right now think he’s ready to take Papelbon’s place. He just doesn’t command well enough yet.

Which brings us to Pap. You might recall that last season I, along with a great many other people, worried about the fact that Pap had basically reverted to a one pitch pitcher. Well, the good news is that he is indeed throwing the split more: 17.9% thus far, well up from last season’s 9.3%. The bad news is that his pitch selection may be impacting his performance – negatively. As Lundblad put it,

Papelbon has issued four walks in four appearances so far this season. In 2008, he didn’t issue his fourth walk until June 22, his 33rd appearance. Factor in just one strikeout, and this is the first time in Papelbon’s career that he has more walks than strikeouts.

Add it up, and I think the pen will be this team’s weakness.

Big Picture?

We’re going to play better, and things will turn around. I’m not going to guarantee 95 wins like a lot of people – the BP guys’ latest simulations are currently predicting an 80 win season and 24% chance of the playoffs, but we’re clearly better than that. Just as we’re better than this.

Will it be enough to make up six games in the toughest division in baseball? Who knows. But we can’t worry about that right now. Our only concern should be getting these guys playing better.

Meaning don’t boo them, it’s not going to help anything.