Basketball analytics make Gophers look good

• Blog Post by:
• November 8, 2012 - 1:51 PM

So the NYT’s Nate Silver pretty much schooled America with a play-by-play of the presidential election before it happened?

So after making all the talking heads look like fools, he has something like a 99 percent success rate over two elections?

Hmm … there might actually be something to this “math” theory after all, eh?

Perhaps this is the time to start taking a gander at the more analytical rating systems in college basketball, too, and just how they view the Gophers.

First, a reminder of where the Gophers stand in the two most commonly used polls:

In the media basics, the Gophers have been mostly forgotten, with a couple exceptions:

Athlon: Not ranked
Lindys: No. 31
ESPN: Not ranked
SI: Not ranked

Now, let’s look at the “other” rating systems. Ken Pomeroy’s tempo-free analytics have long held a niche in the basketball community and has had a steady following of basketball nerds -- er, fans. His system uses a complex calculation of different factors including who the team has beaten and how it has beaten them and who it has lost to and how it has lost.

This year, RealGM’s Dan Hanner has also released his own new projection system for Basketball Prospectus that looks at historical data inputs and, like the Ken Pom rankings, uses the Pythagorean formula to calculate a number of different factors. There’s a 16-step process that projects playing time, individual performances, possession distribution and calculates adjusted offense and defense.

So. Much. Math.

I won’t get into all the details because I don’t want to bore or confuse you guys (I totally understand it, though; seriously, it’s not over my head at all).

But hey, after Nate Silver, should we all be doing this?

This is how those two systems view the Gophers:

Ken Pom: No. 26

Interesting, too, that Hanner has Wisconsin at 33 (!) Michigan at 44 (!!)

Hm … suddenly more intrigued? (Even if only for use in debates with your buddies?)

The thing with these ratings – and any ratings – is that right now, we’re at the beginning of “election” season. Candidates are still warming up and you never know who is going to surprise you with something stats never could have suggested or how they are going to shoot themselves in the foot (or fail to shoot effectively at the basket).

And generally, even predicting a game in sports is not the same as predicting an election based on polling data, as this interesting blog on Big Lead Sports noted.

We don’t have as much information now as we need to REALLY know what will happen.

Of course, that’s the beauty of basketball, right? You can pore through the statistics and read all the analysis and come armed with numbers (and yes, that part is fun too), but in the end, you have to just watch the game.