I've already written about Strasburg's debut, but it's something that's probably worth writing about at least twice, don't you think?

Since the performance absolutely failed to cool down the hype, it seems there are some people looking to pour water on it on their own. One of the things you hear on occasion today is, "Well, it was just the Pirates."


So far this season, the Pirates have maintained a .192 strikeout per PA rate. That's just barely above the league average of .187. That's, um, an extra strikeout every 200 plate appearances. (If you remove their showing against Strasburg, the Pirates had a strikeout rate of… .188.) The Pirates' problem on offense isn't their inability to make contact so much as it is their inability to make good contact.

And in 24 PAs, what are the odds of the Pirates putting up 14 strikeouts against a league average pitcher?

One in a million.

No, I'm serious, I did the math. The Pirates' strikeout per PA in last night's game was 4.875 standard deviations above the team's average, looking simply at the standard deviation we'd expect from randomness. That works out to around a one in a million chance.

And do you know what's really fantastic? Normally when we talk about odd baseball occurrences that are 4-5 SDs above (or below) the mean, the issue at hand is cherry picking. You're looking at this game because you already knew the outcome was unusual.

But with this game – someone told you ahead of time to watch, didn't they? Someone told you this game was going to be special before it happened. And then it was.

Don't blame the Pirates, kids. That was truly an impressive pitching debut, against ANY team.

You need to be logged in to comment. Login or Subscribe
Interesting note and brilliant title to the article. Thanks.
Not really related to the strikeout rate, nor exactly defending the Pirates' lineup here, since this is the lowest-scoring team in MLB and Doumit was sitting, but if you look at the weighted average TAv for the lineup they put out there (yeah, I know, that's back-of-envelope type calculation): Andrew McCutchen .307 Neil Walker .316 Lastings Milledge .242 Garrett Jones .276 Delwyn Young .258 Andy LaRoche .239 Ronny Cedeno .238 Jason Jaramillo .194 Weighted by league-average PA-per-slot: TAv= .261. If you include: Jeff Karstens .170 Ryan Church .211 ... the TAv drops to .254, still much higher than their season rate of .240. This is due mostly to having Clement (.198 TAv) and Iwamura (.217) out of the lineup. Not saying they have a *good* lineup, but the ballpark and the early struggles of two regulars certainly are contributing factors to their bottom-of-league placement in runs/game. As Colin showed, they aren't really a high-K lineup, nor are they nearly as awful as their reputation (due, in large part, to having such an excellent offensive player batting atop the lineup).
I did the math and got a chance of 1 in 59,388 of exactly 14 of 24 and 1 in 50,463 of at least 14 of 24. Of course it took me a good 45 minutes to remember stats 101 and I still probably screwed it up... It was an impressive feet in any event.
I concur exactly with 1 in 50,463 for at least 14 of 24 assuming a 0.188 probability. The standard deviation and Z-score calculations from the post are right, though, so I'm gonna guess that the normal approximation of the binomial distribution isn't valid for expressing the magnitude of a tiny probability the way it is used in the post.
Great point, as I found myself wondering why some people were raining on this parade.