Error in Pythagorean Expectation

Error in Pythagorean Expectation is the difference between its calculated winning percentage to the winning percentage that actually occurred.  Many sites like to claim Pythagorean Expectation is correct and reality is wrong; that a team “should have” won what the Pythagorean expectation formula states.  Mathematics can only estimate.  If there is a difference between reality and math, the error is in the estimation, the math, not reality.

The normal Pythagorean Expectation formula was posted in the previous two posts.  Here is a variation that claims to be more accurate.  We’ll see.

103b25df1a391f10f097154971ebcf8a2

The above formula merely replaces the exponent of 2 with 1.83.  There is another version called the Pythagenpat which has a formula for the exponent states as folllows:

b6e29685ac174dd4578fe2d3da0d92b9

Number of games played for a full season is 162.  The following table lists all 3 of the above variations from the year 1990 to 2013.

Year Pythag Pythag1.83 PyPat AVG
1990 173.11 168.55 168.26 384
1991 105.72 109.49 109.04 384
1992 173.73 173.48 173.14 452
1993 191.06 190.38 190.54 552
1994 150.40 144.80 149.12 374
1995 211.14 210.20 211.40 440
1996 139.26 142.73 140.42 436
1997 203.34 196.08 198.33 434
1998 179.06 182.30 180.60 640
1999 139.90 140.38 140.63 644
2000 154.86 149.55 154.19 512
2001 198.33 196.63 194.98 630
2002 210.68 212.34 211.39 738
2003 197.64 203.77 201.97 650
2004 182.86 191.84 185.99 688
2005 223.62 222.74 221.49 516
2006 179.09 179.98 180.23 490
2007 199.49 196.29 196.82 488
2008 200.32 194.59 197.39 554
2009 227.94 232.22 227.06 570
2010 163.49 149.12 150.44 532
2011 209.69 200.12 201.86 566
2012 161.24 167.61 166.40 608
2013 194.33 187.61 188.95 624
TOTAL 4370.3 4342.8 4340.64

How error is calculated was shown in the previous post.  The error numbers in the above are a summation for the entire league for each year.  The last column is:

AVG = Sum(abs(Team WAA)) across all teams in the league.

The above shows that PyPat version has the least error but not by much.  Overall error seems to hover around 6-7 WAA/team/year where WAA is measured as (W-L).  This means that if Pythagorean Expectation says a team should be at 81-81, its margin of error is WAA/2 or between 84-78 and 78-84.

Pythagorean Expectation only uses runs as inputs.  Although the sheer number of runs scored and runs scored against should be good estimators, the timing of those runs also matter.  A team that under performs its PE has problem scoring in clutch situations and scores a lot in non-clutch situations, and vice versa.  Some players excel when their team is way ahead and others excel when the game is on the line.   The modeling of this timing and differentiating clutch and non clutch players has to be done by a different model.    PE can only make a rough  estimation based upon the above brute force analysis.