Kuku, the stats you showed are really interesting. Can you calculate the variance, too? That will help us in some extend to understand how significant the stats from three games is.
I just reviewed the Rockets vs. Jazz on league pass. It is obvious even to the most casual observer that the Rockets came out flatter than a pancake. Kudos to the coaching staff for motivating their butts at halftime. A 27 point second half swing. That is absolutely insane. Word.
See my previous post back in April. http://bbs.clutchfans.net/showpost.php?p=7799145&postcount=261 However, there is a multiplier of 2.57 for every Point Differential (PD) which can substitute for Morey's Pythagorean Expectation formula and still come up with close enough results. PD*2.57 + 41 = expected number of wins (without taking SOS (strength of schedule) into consideration). For example, Rockets had a PD of +3.5 last season. The expected win would be: +3.5*2.57 + 41 = 50 wins (matches with basketball-reference expected wins of 50 by Rockets) http://www.basketball-reference.com/teams/HOU/2013.html Our margin for 1st half was +0.3 and if we played the same during second half, our PD would be: +0.6 Expected wins: +0.6*2.57 + 41 = 42.5 wins If we played both halves same as second half, PD would be +5.4 Expected wins: +5.4*2.57 + 41 = 55 wins When the PD is too big for the multiplier, you have to revert back to Morey's Pythagorean Expectation formula. I admit I made a safe guess that, with +20 PD, expected wins should be 75+ wins. It's even uglier vs Clippers. Until coaching staff makes some changes, the trend will continue.
LOL. Do not attempt to mask a paucity of data leading to an incredible margin of error. I have attempted to be circumspect in my analysis of your conclusions. This ends now. It is not Morey's Pythagorean Expectation theory. It is James's Pythagorean Expectation theory to be applied to baseball. Morey just took it and bastardized it for basketball. It has also been done in the NFL and even in the NCAA. I have seen no MIT Sloan articles supporting its application to basketball. And realistically it is a minor theorem that makes no difference in basketball games. Just a back end analysis for stats geeks attempting to quantize player value. If you keep up this insane line of unsupported analysis I will take a serious look at your methodology and your underlying assumptions. Should be fun don't you think???
Way to cherry pick my post. I stand corrected that it was Morey's Modified Pythagorean Expectation Theorem. Happy? 82-game sample size is too small? Is that what you are saying?? If you are referring to this year, please see many of my previous posts. Just in case you missed it, SMALL SAMPLE SIZE!
Morey's Modified Pythagorean Expectation Theorem? I think that my previous analysis of the concept fully expresses my feelings about the Theory. Would you please cite your 82 game sample size. I do not remember it from your posts on this thread. It is well after midnight here. Looking forward to tomorrow. Good night to you and all.
Definitely. It is almost impossible to win against a team that shoots that well. Best to you and yours.
