View Single Post
06-01-2013, 11:56 PM
Registered User
Habiton's Avatar
Join Date: Apr 2010
Country: Canada
Posts: 906
vCash: 500
Originally Posted by DAChampion View Post
The problem with these analyses is that they don't [adequately] account for draft rank.

A pick at 6th overall is much easier than a pick at 18th overall. The batting average analogy to baseball does not work, as in baseball every player faces every kind of pitcher over a full season. In drafting, some batters only go up against the worst pitchers, and some only go up against the best pitchers. Similarly, the difference between 1st and 2nd round is huge, whereas the difference between 4th and 5th round is very small (by comparison). Finally, the concept of a "1st round" selection is misleading, as the difference between 5th and 25th overall is comparable to or larger than the difference between 25th and 199th overall.

Given that this calculation would only take a few hours to do properly, I'm surprised nobody has done it yet. I think there's a mismatch between the people who know statistics and the people who know hockey, and very few really know both. In this case, I have the stats knowledge but not the hockey knowledge to do the calculation properly.

If somebody with good knowledge of players drafted in the past 10 years wants to help me out, we can do the calculation together in a way that makes sense.

I think a good idea would be to take the 2003-2008 drafts, and give every player a rank between 1 and 10 in a manner that tries to be objective. I can then fit the numbers to an exponential model, rather than the usual fictional step-function models.
Hence the point of using averages. Once you get into the "advanced" statistics your opening yourself up to more error based on subjectivity. Clearly you think this is useless, although I would appreciate you not being so passive aggressive about it. It is a simple calculation based on a team with no extra advantages. Over the past 9 drafts we have had two top 5 selections and the others were mostly in the 15-25 range which I would say would be on par to be average. Sure you can ***** and complain about the first round and to a point I agree (not in this case due to our low amount of high selection). Third round +, even Second round + unless your in a fantastic position it really doesn't mean much at all.

Maybe next time you should try to post something relevant instead of attacking my methods, some of us are here to look at the stats and see what they really mean. Not a strict or advanced statistical analysis that 1% of the members on this board will ever understand nor should they but an easy way to see how Timmins has been more successful than the average scout in the league. If you cant see that after this, I will just have an additional member on my ignore list.

Habiton is offline   Reply With Quote