clock menu more-arrow no yes mobile

Filed under:

Happy stats

The comments in this thread below got to be a little depressing, until AJ got me thinking.  He pointed out what we should have been keeping in mind all along:

[W]e've been in just about every game and have had a chance to win, its not like we've been blown out 8 times.

Of course, there is a stat for this: Bill James's pythagorean winning percentage.  The idea is that luck evens out over time--most teams win or lose in direct correlation to how many runs they score and give up.  The Brewers have scored 58 and given up 51.  In other words, they should have a winning record, they've just been unlucky in the distribution of their rallies.  A corollary of this is that 1-run games and extra-inning games, while they are the most painful to lose, are not terribly indicative of the quality of a team.  Those results are, at least in part, due to luck.

So, without further ado...

If a simple version of pythagorean winning percentage is to be believed, the Brewers should have a winning percentage of .564.  That would be 7-6 right now; if the Crew keeps it up, that's a ...

...wait a minute, that can't be right.

...just double-checked, getting the same numbers again...

...is my calculator working?

If the Brewers keep scoring and allowing runs at their current pace, and are luck-neutral, they'll win 91 games.

How is this possible?  For starters, 58 runs in 13 games is an average of just about 4.5/game.  Not even quite league average, but considerably better than last year's 3.94.  51 runs allowed (that's runs, not earned runs) is a miniscule 3.92/game, which not only marks a gigantic improvement over last year's Obie-tastic 4.70, but would've outpaced the Major League best Cardinals, who allowed 4.07/game.  

This doesn't negate all of the complaints we've been making--it is generally thought that one of the reasons teams underperform their Pythag percentages is bad managing.  Particularly bad clutch hitting can do the job, too.  Right now, I'd say we've had more than our share of both.

But...say it with me: 91 wins!

Update [2005-4-20 17:14:31 by Jeff]: Amazingly, according to every pythag projection in the division, those 91 wins would put the Crew in 4th. StL is headed for 98-64; Chicago for 93-69; Houston for 92-70 (really miniscule difference from the Brewers: Milwaukee's at about 91.4 wins, Houston at 91.6); Cincinnati at a rockin' 68-94, and Pittsburgh checking in at a Tike-tastic 30-132.

Update [2005-4-20 17:34:36 by Jeff]: I've said this in the comments section and at BrewerFan.net now, but just so I don't have to keep saying it: yes, these games are disproportionately against the Pirates, but they are also disproportionately against good teams, like the Cardinals and Dodgers, who will likely be the best teams in the league this year. I would imagine that, by the end of the year, the weighted average of the winning percentages of the four teams the Brewers have played (weighted by number of games against each) will be a bit over .500. (With last year's winning percentages, that weighted average is .537, indicating that the 91 win prediction is skewed in the other direction.) In other words, the 91 win prediction, whatever it's flaws, is not grossly inflated because of all these games against the Pirates.