FanPost

A Different Perspective on the Nationals and Pythagoras

A lot of the talk on the Nationals recently has focused on their exceeding their Pythagorean record thus far through the season. Pythagorean record was developed by Bill James and combines a team's runs scored and runs allowed to attempt to predict a team's future record and what the team's record "should have been." The formula for the Pythagorean equation is as follows:

W%=[(Runs Scored)^2]/[(Runs Scored)^2 + (Runs Allowed)^2]

The Nationals equation looks like this:

W%=[(132)^2]/[(132)^2 + (148)^2]=.479%

Over 32 games, this predicts a 14-18 record (rounding), in other words a complete reveral of the Nationals record.

Now, obviously, there are flaws in the system-if a team gives up 11 runs in one game, that is going to negatively affect their pythagorean record to show a lower win expectancy. That 11 runs isn't representative of how to the team usually plays, it's an outlier, an abberration and skews the predictions. So I'm going to try and figure out the largest outlier that has caused the Nationals to post an 18-14 ‘real' record and a 14-18 Pythagorean record.

The first word one everybody's lips is probably ‘Marquis', he was mine too. Let's start with him.

Game 1: L, 4-8

Game 2: L: 4-7

Game 3: L, 7-11

That's a run tally of 15 RS and 26 RA (a .249 Pythagorean W%).

By subtracting that total out of the National's runs scored/allowed, we get 117 RS and 122 RA, a .479 PW%. Out of 29 games (32-3 Marquis starts), that procures a 14-15 record, out of 32, a 15-17 record.

So clearly, Marquis was not the largest factor in the four-win differential, he was worth only one.

The second thing I elected to look at was the Phillies. The Phillies are the class of the National League, they've won the pennnant for the past two years, they're more or less the best team in the National League, and teams probably shouldn't expect to fare well against them.

Game 1: L, 1-11

Game 2: L, 4-8

Game 3: W, 6-5

Game 4: L, 4-7

Game 5: L, 7-14

Game 6: W, 7-5

In the six games the Nationals played against the Phillies, they scored 29 runs and allowed 39, a PW% of .360. When subtracting that total, we have a season total of 104 RS and 98 RA. This yields a .530 PW%. Out of 26 games (32-6 Philly games), that produces a 14-12 record and out of 32, a 17-15.

So what does this mean? It means that the Nationals are playing more or less to their Pythagorean record against the league outside the Phillies. The Phillies cause problems for everybody, and as this shows, also for a team's Pythagorean record.

For kicks, I've decided to look at how the team would fare with facing the Phillies or trotting out Marquis for three starts. 89 RS, 72 RA. The PW% for this comes out to a .604, or a 19-13 record. (on a side note, that's a 98 PW team) The statistical anomaly of Marquis and the gauntlet that is the Phillies having been removed, this shows us how damned good they've been playing.

Obviously I understand that this is 'playing with numbers', too small a sample size, that I can't just throw out Marquis and the Phillies, I can think of a lot of criticisms for this analysis, but it's just to provide my own perspective on the Nats and Pythagoras.

All FanPosts on FBB consist of content created by site users without editorial oversight by Federal Baseball, and do not necessarily represent the views of the FBB editorial staff.

In This FanPost

Teams