Unlucky Wang?

A week ago, Chien-Ming Wang faced the Red Sox, and pitched brilliantly, last night he faced those very same Red Sox, and pitched horribly. So I went looking for what went wrong.

Watching the game last night, it was easy to assume that Wang didn’t have command of his sinker, and as a result was falling behind batters more often than he usually does. Except it isn’t that simple.

Last night Wang faced 24 batters, and in his previous start agains Boston he faced 29 batters. In both starts he had eleven Plate Appearances that were resolved when he was behind in the count.

In the start last week, the Red Sox got on base ONCE in those 11 situations, and last night they got on base 8 times in those situations. So Wang wasn’t really pitching from behind all that more often last night, the problem was he was surrendering walks in those situations, which he usually doesn’t do, and when the balls were put in play yesterday they were going for hits at and extremely high rate.

Game    BABIP
11-Apr  0.074
16-Apr  0.474

A fluky thing for sure, two extremes in back to back games. And it illustrates why Wang has to continue working on his strikeout pitches.

I just wasted a bunch of time.

I did this project this morning, where I looked at every major leaguer’s 2007 stat line and ranked them according to Batting Average, SLG, OBP, OBP-Batting Average (more on that in a minute.) Then I calculated the standard deviations of each data set, then I ranked players who were +1 or +2 standard deviations above the category mean. Then, I calculated the correlation coefficients for 12 different offensive metrics(correlated to runs.)

See, I’m totally not making this up. FreeRunn is the sum of BB, HBP, and IBB. And if you don’t want to lose, don’t get caught stealing a bunch.

Then based on that calculation I weighted OPS so thta SLG got a tick more weight than OBP because SLG tends to correlate to Runs a bit more than OBP. Then I created a weighted OBP that prioritized BA slightly over walks and other non hits (because hits correlate more strongly to runs than walks or HBP or IBB.) Ok, then i re ranked all the offensive players from 2007 using a new weighted OBP and weighted OPS and I found out, after all that, that Alex Rodriguez and David Ortiz are really good hitters.

Just for fun though, the top 6 players based on my weighted OPS metric are:

Ortiz: 1.112 (Regular OPS 1.066)
A-Rod: 1.112 (1.067)
Magglio: 1.089 (1.029)
Chipper: 1.081 (1.029)
The one who got away Pena: 1.027 (1.037)
Matt Holliday: 1.070 (1.012)

The conversion hurts Pena, and helps all the other guys. I should point out that this analysis is all bunk because I didn’t adjust for park, or league.

I should also point out, because I think it’s cool, that Chipper and Magglio were the only two players in baseball who were more than 2 standard deviations above the mean in the following categories, BA, OBP, SLG, OPS.

Finally, I realize that above I promised more on OBP-BA. Here’s the thing, if you had 2 guys with .450 OBP you’d be like, holy crap my payroll is going to be HIGH. But if you wanted to know which one to keep, wouldn’t you want the guy whose OBP was more made up of hits. I would because hits are sometimes doubles or triples, and walks are always base cloggy singles (kidding of course.) But you always only get 1 base on a walk, and dudes never go first to third on a walk. So, love OBP, it’s my thing, but I love OBP that is a result of hits slightly more than OBP that is a ton of walks. Like I’d rather have 2007 Magglio than 2007 Helton even though they had identical OBPs because Maggs had a higher BA, higher SLG…so I was just trying to isolate those guys that have OBP that is more a function of hits than non hits. I realize too that BA is pretty variable year over year, and that’s why OBP is better from a planning standpoint. But Hits correlate more strongly to runs than walks do, at least in 2007 they did.

Contact Rate

WasWatching has an interesting post about contact rate.

Ron Shandler in the USA today had this to say about contact rate:

So, when it comes to batting average, perhaps the skill that we should be tracking is the rate at which a batter makes contact.

Contact rate (at-bats minus strikeouts, divided by at-bats) is a statistic that is far more stable and projectable. League level rates run about 80%. Our .300 hitters often come from those with contact rates greater than 90%. Batters with rates less than 70% typically have trouble keeping their batting average above .250.

First of all, what I think this article is saying is that hitters who make more contact tend to have higher batting averages. Um, great.

But what I think could be interesting is applying contact rates to specific situations, like the post-season. I’ve always sorta thought that hitters with low batting averages yet good OBPs were something of a liability in the post season. More specifically, a lot of hitters, clustered together with low BAs were a problem. Because as long as pitchers didn’t throw them meatballs, they could capitalize on the passivity of the approach.

So maybe in the postseason, you want guys with high OBP AND high BA. And for that matter you want them to have highish slugging too so they are efficiently moving guys around the bases.

So basically, I’ve just talked myself out of thinking there’s anything interesting about contact rate.

Nevermind.

So random

Chad Moeller goes 4-4

Wang give up 8 runs in 4 innings.

The Red Sox give up 15 runs.

If this isn’t enough to illustrate the dangers of drawing conclusions from small sample sizes, I don’t know what is.

The only certainty in life is that Giambi own Mike Timlin. Owns him. Like the way the teenagers who play Call of Duty 4 online own me.