Quote
PHDram
You came up with your own way to asses the draft which is fine but it still has a number of subjective elements. Most notably as you point out is the difficulty in defining a hit. However my main objection is your dample. If you only picked good teams to assess your sample is biased. Without the bad teams we cant get an average. Moreover a reputation for drafting well does not actually mean they did. Its curious as to why no one has ever published a really good methods paper on grading a draft.
As I said all of this went through long real good debates and I tend to stand by it. I think you just took the word "reputation" out of context. Brady has a reputation for being good in comeback situations. And guess what, it's true, the numbers back that. I looked at teams that had a good reputation for good reason---the teams I looked at DID draft well, and that's why I looked at them. Forgive me but IMO the "sample size" argument is the most misused internet trope. The point was to find a standard based on teams that did do well--so, what is doing well? You look at good teams, voila there's a standard. And it has nothing to do with "bias"--- I set out to find the standard set by good drafting teams and I did. Good drafting teams hit at 40% or better. What I DIDN'T DO was a comparative study of hit percentages for all teams and make a distribution chart. But I wasn't looking for that. The only argument that might dissuade me would be if someone does the work and comes up with what they claim is a better metric.
Now if you don't LIKE the 40% number for any reason at all, that's fine...completely your prerogative. To me it's useful in clarifying draft discussions a bit. I think a lot of people have unreal expectations for what it means to draft well.
...
Edited 1 time(s). Last edit at 03/22/2019 08:55PM by zn.
