Bucks Diary

Monday, October 06, 2008

Ranking the NBAs Power Forwards by WinRate

As part of my preseason buildup here on Bucks Diary I am using the Win Profile statistics that I spent a great part of the summer calculating (I'm only partially exagerrating -- I think it took me about 100 hours... but a lot of that time was spent putzing around) to compile lists of the most productive "projected" starters at each position on each team (according to Foxsports.com).

Remember, this is not a "30 best" listing, this is simply a listing that follows the results of my calculations. So there are some results that will be surprising. PLEASE take issue with anything you disagree with. Call me a moron, say this isn't no kind "of a blog"... (read the back comments to get the reference), whatever, I like controversy... I didn't spend 100 hours punching that stupid calculator just for the results to lay there like a fried egg on the pavement. My stats are certainly not the be all and end all of basketball statistics... and they won't get any better unless they get critiqued.

In fact, I came up with the concept of individual Defensive Win Score only after I made the comment that the Bucks ought to build around Mo Williams because he had a high Win Score and Pressey25 ripped me and said thats stupid, you can't build around a guy who plays no defense, and I thought about it and decided he was right, and then I came up with a method... for better or worse... that holds guys like Williams accountable on the defensive end. So far it works. We'll see.

I gotta fly, so no time to comment today on the Power Forwards list, and that accounts for the rambling post. I'll follow up tomorrow and do better with my prose. Sorry.


At October 7, 2008 at 3:01 AM, Anonymous Anonymous said...

Washington Bullets? Francisco Oberto?

At October 7, 2008 at 3:04 AM, Anonymous Sylvan said...

Keep up the good work. You are appreciated. By the way, send that metric out if you haven't ptented it, it's a good one that everyone should use

At October 7, 2008 at 4:09 AM, Blogger TCW said...

Anonymous, i ask for leniency on names... when you do alot of calculations, you tend to get sloppy with names. Not a justification, I know.

As for the Bullets, that is blog policy. Remember how half the world wouldnt recognize Communist China for years because they thought it sucked? I think the name "Wizards" sucks! And, more importantly, I think the name "Bullets" rules! Maybe because I grew up with it and associate it specifically with NBA basketball, but I just love it and always will. How could they change it.

As for the name change, as long as you knew who I meant, i made my point, and i suspect you did know.

At October 7, 2008 at 4:27 AM, Blogger TCW said...


Thanks for the nice words.

Im hesitant on a couple of grounds. First, the metric I'm using is essentially a non-academic amendedment to an established metric developed by serious academics and based on years of solid research.

Not that mine is therefore necessarily wrong because of that. In fact, I'm more and more convinced that its right... and that's the problem. It would put me in an almost adversarial position with the originators that I'd rather not be in.

Because if the original Win Score is right, my Defense-Incorporated Win Score is not necessarily wrong, but DI Win Score is right, then the Original Win Score almost certainly has to be wrong (look, for example at Troy Murphy... Original Win Score would have him rated far higher than I do. Am I wrong to downgrade him for his defense? So far the evidence does not suggest that I am.)

Not to get too long winded, but that's why this Bucks season is going to be so interesting.

Basically, the Bucks, by straight Win Score, should be a 25-28 win team again. They have some brutally bad Offensive Win Score players.

Now, enter Skiles. If he can get them into the 35-38 win range, I can almost guarantee those extra wins will be explained by Defensive Win Score (Players rarely improve their offensive output, but defense can be elevated).

So Milwaukee is something of a test tube. That's not the right word. What word am I looking for? Oh, who cares. You get my point. TEST CASE!! That's it.


Post a Comment

Links to this post:

Create a Link

<< Home