Incorporating Opposing Alliance Information in CCWM Calculations

[wgardner]

Originally Posted by Ether

Attached are A, b, and T for 2015 OPR, CCWM, and WMPR.

Also introducing a funky new metric, EPR, which uses 3 simultaneous equations for each match:

1) r1+r2+r3-b1-b2-b3 = RS-BS
2) r1+r2+r3=RS
3) b1+b2+b3=BS

... and solves all the equations simultaneously.

How do you interpret these new EPR values? They look closer to an OPR than a WM-measure as #2 and #3 both only factor in offense. How will you measure its performance? Perhaps by comparing the overall residual of all 3 combined vs. other ways of predicting all 3 (e.g., using WMPR for #1 and standard OPR for #2 and #3)?

[Ether]

Originally Posted by wgardner

How do you interpret these new EPR values?

I just tossed it out there for fun. Since equation1 is a linear combination of equations 2 and 3, I question its usefulness.

Originally Posted by wgardner

How will you measure its performance?

Compute the residuals of actual_alliance_scores minus alliance_scores_computed_using_xEPR:

residuals = b₁-Aopr*xEPR

...where b₁ is column1 of the provided bopr.