Cleat Geeks

Saber-Metrics Sunday: A^2 + B^2 = C^2 = WINS?!

Traditionally most people predict what a team is capable of based on basic knowledge of a team. You see conversations such as, “Oh (team X) has (players Y, Z) and added (player X), but they lack pitching depth so they will likely go 81-81.” Just as an example.

Then when a season is over, people evaluate what they did and see if they overachieved, underachieved or did exactly as predicted. Happens for every team every year.

The next topic I am introducing is, Pythagorean Win-Loss Record. This is an equation that uses the Pythagorean theorem (a^2 + b^2 = c^2) and uses various team stats to predict the winning percentage of a team. So if you were to calculate your teams Pythagorean Win projection using the most up-to-date stats, this is the formula:

win% = (runs scored)^2 / (runs scored)^2 + (runs allowed)^2

To put this in mathematical terms, you see the variables calculated as a relationship outcome as you would a right triangle. In other words, you are not looking at a ratio between runs scored and runs given up you are looking at the relationship of the two as they create the outcome, like how a right triangle’s sides are related to one another to create the complete shape. So all stats function as one as a team’s collective play function as a unit. That is key to understand.

Pythagorean record is always being adjusted as games are played. As current stats are placed in the equation, the different outcomes will show how well a projected team is doing. The initial projected winning percentage is almost never right on. Many teams over perform their projected record and many also underachieve it.

Now the important question to ask is, “How important is it really?”

Well it may not be perfect, the stats used are very basic and don’t really account “luck” in the outcome. But it still gives a basic idea of how good or bad a team is projected to be and how to measure their progress throughout the season. It should never be used as a “dead on” measurement of how many wins they will have, but more of an area of success a team should have. So if a team is projected to have a .560% then a team is probably predicted to have a .500 or slightly better record.

This is a really fun stat to use midseason to see how many wins “off” or “above” a team is to their Pythagorean win outcome because you can have a real good team (2016 Chicago Cubs for example) and still see they are still behind their Pythagorean win total despite still having a great record.

Leave a Reply


If you like this site or just simply want to school your friends because you got the information first.  

Join us on the field! Click on any of the links below.

%d bloggers like this: