Bill James introduced the idea that one can estimate the amount of wins a team should have from looking at their runs scored and runs allowed totals. It's not perfect but it's a good indicator of whether a team over or under performed. Over the years it has been very accurate. The difference can sometimes be chalked up to managerial ability but mostly it's luck. Teams with excellent bullpens often over-perform due to winning more close games.
The formula is:
Win% = Runs scored (2)/Runs scored (2) + Runs allowed (2)
(Note: The formula wouldn't copy to this page so I wrote it in. (2) = squared)
This is the breakdown for my teams this season:
| League | Runs Scored | Runs Allowed | Pyth. Wins | Actual Wins | Difference |
| | | | | | |
| B8 | 870 | 695 | 97.67 | 88 | -10 |
| F18 | 947 | 731 | 100.26 | 102 | 2 |
| W9 | 536 | 780 | 51.32 | 54 | 3 |
| A12 | 962 | 754 | 99.11 | 98 | -1 |
| Champs | 798 | 649 | 96.3 | 87 | -9 |
| 99T2 | 1014 | 652 | 113.19 | 107 | -6 |
This kind of fits my general perception that I'm not a particularly good game manager. I like the GM side of things much better and think I'm better at assessing and accumulating talent than utilizing it in the actual games. The only league where playing closer to the Pythagorean estimate would've made a difference was B8 where I would've made the playoffs instead of missing out by a few games.