Pythagorean Winning Percentage/Pythagorean Record

Hi everyone, it's been a while since I contributed anything at all onto Brew Crew Ball, but since this place has consistently given me some of the best Brewers-related statistical analysis and baseball coverage (I lived in Florida for much of the past five years), I figured I'd add something that's been on my mind for quite some time now.

One thing I have been hearing a lot of lately is how the Brewers are not a real threat to do anything this season, and much of that criticism is based on the fact that the Brewers have had a losing Pythagorean winning percentage for much of the season, and have just now evened it out with Ben Sheets' marvelous pitching performance in a 4-1 win against Atlanta.

Now, onto what Pythagorean winning percentage is.  It's a statistical analysis tool that predicts "how good a team really is" by how many runs a team has produced and how many runs they have allowed.  The basic formula looks like this: 1/(1+[Runs Allowed/Runs Scored]^1.81).  As with any other statistical analysis tool, like BABIP, it provides a good gauge on what it measures (with PWP, the "real" quality of a team, and with BABIP, how "lucky" a team is with their hitting), but it is not by any means an end-all predictor, though I think BABIP tends to regress towards the mean quicker.

Let's look at one glaring example from this year and a few from years past to show that PWP should be taken with a grain of salt.  This year, the Atlanta Braves are sitting at 38-40 and are in 4th place in the NL East.  According to MLB.com, the Braves have a Pythagorean record (which you can synthesize by figuring out PWP and then multiplying it by how many games a team has played) of 43-35, which is the 3rd best in the NL and the 6th best in the majors.  Now, the Braves' road struggles have been well-documented in the media, especially their inability to win close games.  Atlanta has lost a mind-blowing 22 straight one-run games on the road, their road record this season is 11-27, and their overall record in one-run games this season is a pathetic 4-19.  Is a team that has this much trouble winning close games and road games really the 3rd best team in the NL and the 6th best in the majors?

Another example to take into consideration is this time of an "overachiever", the 2007 Arizona Diamondbacks.  The Desert Snakes had a Pythagorean record of 79-83 last year, the 16th best in the major leagues.  However, Arizona posted a real record of 90-72, the best in the NL.  The D-backs further cemented their legitimacy by sweeping the Chicago Cubs (who had a better PR) in the NLDS.  True, they were swept by the Colorado Rockies in the NLCS, but I attribute that far more to the Rockies unbelievable hot streak than the D-backs PR.

Another pair of PWP/PR anomalies came in 2006.  Remember the hype surrounding the Cleveland Indians in '06, where they were supposed to run away with the AL Central title and make a huge run in the playoffs?  Well, that dynamite group stumbled to a 78-84 record, good enough for 4th place in the AL Central, despite their Pythagorean record of 89-73 (4th in the AL, 5th in the majors).  An even more stunning example came in the form of the '06 World Series Champion St. Louis Cardinals.  As you remember, the Cardinals limped into the playoffs with a record of 83-78 and a PR of 82-79 (7th NL, 15th in the majors).  Who would have known that the majors’ “15th best team” would go on to win the World Series? 

Of course, this is not to say that the Pythagorean winning percentage is a poor indicator of a team’s projected success.  In fact, the 2007 World Series Champion Boston Red Sox had the majors’ best Pythagorean record last year at 101-61, and on average, the tool is a good one to use to predict how successful a team should be.  However, judging from the examples shown above, it does not take into consideration many intangible factors, like mental toughness, focus, and the will to win.  The overachievers (’07 Arizona, ’06 St. Louis) seemed to have these while the underachievers (’06 Cleveland, ’08 Atlanta) seem not to.  It may be too early to tell if the Brewers have these qualities in abundance, but their sterling 16-6 record in one run games and their improved play on the road indicates to me more that they do rather than not, and thus, deserve a bit more respect.  I will delve farther into the past on this matter in a future fan post, but for right now, I am in a hurry to finish this post, so I will leave it as it is.