Doctor MJ wrote:WestGOAT wrote:eminence wrote:
SRS isn't too complicated: It's just MOV+SOS. For example the 2006-07 Spurs won games by an average of 8.43 points per game and played a schedule with opponents that were 0.08 points worse than average, giving them an SRS of 8.35. For an NBA regular season it very rarely separates from MOV in a meaningful way.
The calculations seem circular to me, MOV for Spurs is 8.43 and when you adjust for Strength-of-Schedule, which is basically the opponent's average SRS, then you have the Spurs SRS, but how do you calculate the opponent's SRS?
The ratings keep changing untill they reach an "equilibirum" somehow:From: https://www.pro-football-reference.com/blog/index4837.html?p=37Uh oh! Everyone else's ratings just changed again, so we've got to run through the same procedure again. And again. And again. And eventually the numbers stop changing. When that happens, you know you've arrived at the solution.
I was just wondering what formula/calculations you need to exactly apply.
Are you coder? What we're talking about here is basically a while loop - keep going in circles until the exit condition is reached. In this case the exit condition pertains to how much the weights are still changing. Keep going until it settles down.
If you're wondering how things "magically" reach equilibrium, well I think the reality is that you're talking about waiting for the change to fall below some threshold and perhaps waiting for to stay there a certain number of iterations as opposed to truly waiting for change to reach zero.
Also, if you're doing stuff in a spreadsheet it's not the most conducive to loops, but there are things like Solver add-ons you can use.
Hope that helps.
If it helps, I've done it for college seasons, and run it through Excel 10 times. By the 10th time, the difference in ratings between the 9th run and 10th are pretty small, so I stopped there. My results came out real close to S-Ref or Sagarin. For full season NBA the strength of schedule adjustment doesnt get to be too big, and there are enough teams so it doesnt matter too much. In an 8 team league it does matter a little, because the best team and worst team dont get to face themselves. Im guessing 1976 ABA with one real bad team out of 7 might be best example.