I'm coming back to the Senior Prospects Metric (SPM) and I'm continuing to make small changes in it. I'll mention the changes I've made:
Height has now been built into the model. I've found the heights of all of the players and made the following assumpions:
Guard height. John Hollinger makes distinctions between short point guards and short shooting guards. Since it's hard to determine whether a guard is a point guard or a shooting guard just by statistics (he furthermore makes a distinction between "short" and "somewhat short" guards), I've simply penalized any guard who is 5'8" or less. These "short guards" get a penalty which is 3/4th of the penalites that follow. The reasoning Hollinger uses is that if a short guard is really good, the other parts of the metric will overcome the height penalty.
Forward/center height. If a forward center is shorter than 6'3", they will receive a penalty.
The 6'6" ers. We could call this the "Katie Feenstra" rule. Hollinger noticed that NBA prospects who were taller than 7 feet were generally a sorry lot. These big men had distorted statistics because their opponents were generally shorter than the players they'd face in the WNBA.
I've penalized any player that is taller than 6'6". Sylvia Fowles barely makes the cutoff at 6'6".
Guards who can't shoot from the perimeter. According to Hollinger, a guard should be able to make at least 25 3-pointers in a season, which is about 0.78-0.85 threes a game, depending on the length of the season. If a guard can't make that many threes, they get a penalty.
Bad rebounders. In order to make this one work, I had to drag in a formula from John Hollinger called "Rebound Rate". I am not going to drag out the formula, which depends on the player's minutes played, her rebounds, and her college team's rebounds and opponent rebounds. I'll just explain how it works.
Assume that during an average game against College Team's Average Opponent, both teams have 100 total rebounds (that's a lotta rebounds!) Rebound rate answers the question "how many rebounds would player X get out of those 100 rebounds?"
Of course, the Hollinger Rebound Rate he uses in the draft prospects article...isn't the same as the Rebound Rate formula that he uses everyone else. I had to depend on a converstion from Frisco del Rosario, for which I thank him.
The penalties assessed are for:
* guards with a rebound rate of less than 5.0
* guards between 5'11" and 6'1" with rebound rates of less than 7.5
* guards 6'2" or taller with rebound rates of less than 8
* forwards with rebound rates of less than 12
* centers with rebound rates of less than 13
With no further ado, here are the 2008-09 prospects reevaluated.
4 comments:
Nice adjustments. Will look forward to seeing how these play out over the year.
Have you tried comparing these to the 2008 rookie class yet?
Actually, yes, but I thought that the work and the discussion on statistical correlation would bore an audience of laymen.
I'll probably post something to the effect over the weekend.
If I had to make a guess, I'd adjust Sprague down by 25-33 percent for the "Amanda Rego graduation factor".
Hey , I think Charese Tudy Reed # 20 Xavier should be add to the list of WNBA 2009 prospects.
Post a Comment