## 39 Ways to Die

Sometimes, simpler is better. Sometimes, mathematically gymnastics are required to extract good from evil and separate the informational wheat from the noisy chaff. Joe Lemire of SI.com has a great article today, on a simple measure for a different type of win expectancy: batters faced.

Lemire leans on the work and experience of former Orioles pitching coach and current director of pitching development Rick Peterson in unveiling the mystical properties of the number 39 – as in 39 total batters faced for a pitching team is often the difference between winning and losing.

The Orioles analytics department showed Peterson (who in turn showed Lemire) that the difference in winning percentage plunges when a team faces 38 batters versus 39 or 40 (home teams only used for this study, as they registered the required 27 outs). As Lemire notes, it seems intuitive enough: more batters faced means more base runners and more run-scoring opportunities.

Click through for the full breakdown of how the Orioles manage their bullpen with this number in mind, trying to avoid facing the heart of the opponents order a fifth time in a single game.

No better way to test out sound mathematical theory than with a little ad hoc Small Sample Size Theatre! How did this shake out for the 2012 season? Any outliers or reasons for concern? (Line indicates league average of 37.90)

Oddly, the architect’s own Orioles pop up as one of the biggest outliers. Perhaps part of the reason Baltimore didn’t blow their brains out this offseason trying to chase the 2012 dragon.

This information should not be taken as gospel, as it represents only one year and lacks crucial context such as “offense.” Good defense and pitching teams don’t always win, as evidenced by the sadsack Seattle Mariners failing to capitalize on their own efficient run prevention.

The Rockies will always be an outlier because of their ballpark but that…that just seems wrong. Again, the run environment in Colorado makes this screwy…or does it?

Logo change indicates pre/post humidor

When the Rockies are good, they face fewer batters per game. It is just that simple. Pitching and defense always matters, no matter how many homers you think you can bash.

All in all, a very interesting piece you simply must read. So read it! Then check back: is there anything predictive in what we see above? Should Braves and Rays fans be ready for a bump in wins?

Team data courtesy of Fangraphs.

1. It would be useful if they could break this down by batters faced within the first 8 innings. A problem with including all 9 innings is that teams on the road that were behind in after 8 innings won’t face the additional 3 batters of the home team in the 9th. This will almost certainly make the relationship even stronger.

• “home teams only used for this study, as they registered the required 27 outs”

2. This is neat. Maybe not as exhaustively inclusive as WAR or whatever but the ratio of information to simplicity is very tidy. Low staff BB rate? Fewer batters! Your shortstop is a 37 wax statue? Fewer double plays, more crap singles, more batters! Plus i feel like i can keep track of this when i’m watching games without a calculator, while drunk. Wonderful stat.

3. I’m not sure that this is really groundbreaking, as teams that give the opposition less chances to score runs will usually fare better. 39 just happens to be that number. Better teams will generally face fewer batters because their pitching is better.

I don’t see how this is any different than Joe Carter got a lot of RBIs because the 3 in front of him were great at getting on base; good teams with good pitching will face fewer batters.

Am i missing something?

• “39 just happens to be that number.”

That’s the groundbreaking part. The number. It’s intuitive enough that you read it an think “well yeah” but had you actually thought about it in these terms before? I’m going to go ahead and assume not.

• “You’ve got to manage your bullpen [because] it’s critical that the 3-hole hitter doesn’t come up for the fifth time.” – from the SI article

You’re right I’ve never thought about the number, but shouldn’t pitchers always be trying to the least amount of batters? …working along the same lines as why the protection theory fails in practice? (Because pitchers who can pitch differently to get batters in front of the protected bat out would just do it anyways, all the time)…and better teams generally have better pitchers?

• I think that quote from the original article oversimplifies the main point, yes. It doesn’t change the fact that staying as far from 39 – by not giving away free passes or giving up outs on the offensive side – is in everyone’s best interests. The revelatory nature of this article is that the established orthodoxy is correct, just not for the reasons originally believed.

• Fair point, I can see that.

Although I’m far too lazy to do the research, I’d love to see how that ’39′ number changes based on the run scoring ability of the team, a 600-run team versus a 900-run team

4. That’s scary…..I just watched The 39 Steps last night for the first time and see this article today…..