Quote
PHDram
So I know you're not asking me but I think this is an interesting question so im jumping in. Overall I like what pff tries to do. However this situation does point out some flaws in the system. Anyone who watched the game knows that both JR and Williams were fantastic but potentially in different ways. Im not sure how you address the JR rating other than saying its biased for DBs in favor of certain factors like INTs and PDs. Thus it undervalues pure cover corners. In having built rating systems like this myself I can vouch that its not easy so I dont fault PFF. However if I were PFF I would acknowledge the bias and use this information in order to build a better rating model.
Or can't they have two separate ratings? I would like a CB metric that only measures "stickiness" - i.e. how well is the CB blanketing the receiver.