Comments on: Measuring basketball

By: Big East Hoops » Blog Archive » Basketball Prospectus - all about Big East Basketball

Tue, 13 Nov 2007 03:50:19 +0000

[...] The authors at Basketball Prospectus should not be new to anyone–John Gasaway (Big Ten Wonk) and Ken Pomeroy (kenpom.com) continue what they started as “part-time” hobbies (good idea!) at this Baseball Prospectus spinoff. I’m a long-time junkie of the original BP. The baseball formula worked very well, and there’s little reason not to expect the same for other sports. Donald might not agree, but I see a lot of potential for basketball research. There’s a huge gulf between what announcers say and what really matters. If baseball provides any indication, this gulf will persist for quite a long time, if not forever. Analysis will inevitably invade the mainstream — I’ll put the over/under on Jim Nance dropping “tempo-based ratings” in context at 5 years. [...]

By: Mister D

Mister D — Thu, 08 Nov 2007 18:29:45 +0000

It seems to me you’d have to come up with some quasi-qualitative unconventional stats. Like “shot quality” which could be some combination of the shot distance, the amount of pressure from the defense, and the player’s shot-making ability.

Very interesting to see what might come out of this field.

By: donald

donald — Wed, 07 Nov 2007 06:22:34 +0000

I understand what you’re saying (and thanks for the correct reference wrt tempo-free stats).

What’s interesting to me is that in baseball, you can reasonably consider each player independent of each other and determine good lineups based on individual player ratings. In football, this is more difficult to do, but I believe football is more about strategy than the individual talents of any one given player; as a result, it becomes more about analyzing the effectiveness of plays (which is still difficult to do).

With basketball, you don’t get either. The numbers still say something and can be used to understand what’s going on, but I wonder how they can be used strategically.

Anyway, great blog (we’ll keep an eye on it over here at Big East Hoops).

By: greyCat

greyCat — Wed, 07 Nov 2007 03:07:19 +0000

I have been an avid fan of D1 basketball for over 40 years and resisted applying quantitative analysis to the game. The stats, even the per game numbers told me very little about the player. I took a look at Ken Pomeroy’s tempo-free stats approach (pioneered by Dean Oliver), however and realized that using pace as the constant metric allowed me to understand a number of “things” I was watching on the court. It does provide a slightly different way of breaking the game down to see how the players interact (or don’t interact). The development of even more sophisticated techniques will bring more insight over time.

I love to watch the game, but I bring a slightly different way of organizing and interpreting what I see on the court. I does not diminish my appreciation of a well orchestrated offense (like the Princeton Offense used by your Hoyas), nor does it stop me from yelling at the top of my lungs for my favorite teams. It’s basketball for heaven’s sake, that will never change!

By: Big Willie Style

Big Willie Style — Tue, 06 Nov 2007 12:34:53 +0000

Great article. You’re right about football being as discrete as baseball. Basketball is totally, totally different. Like soccer, it’s the flow of the game that matters rather tahn individual plays. Sure, plays matter, and good plays matter alot, but it’s all about controlling the flow of the game, and how exactly do you find an exact science to that?