Point Differential, The Playoffs, And Ignoring The Numbers


I’ve gone on record as saying that I’m a guy who is a pretty big believer in advanced stats…but I don’t always understand the math behind it as I am not, well…much of a math guy.  But I’m willing to look into those numbers and divine whatever I can and listen to the analysis of the stat dudes like Prof. Hollinger and the dudes over at basketball-reference.com and what-have-you.

That’s a long-winded lead-in to linking to this post over at dallasbasketball.com discussing NBA Champions’ point differential.  Over the last 30 years, the average point differential for NBA Champions has been +7.1.  The highest was the ’95-’96 Bulls at +12.3 and the lowest was the ’94-’95 Rockets at +2.1.

The dallasbasketball fellas do some analysis, and throw out 10 thoughts on the subject.  Here’s an interesting one:

6. As a matter of comparison from this year’s number to Dallas’ point differential in the 2006 NBA Finals year: That club was +6.1. And maybe this is a good place to note the “intangibles”: Many Mavs people think this year’s team is more talented than that year’s team. But what about the opponents’ levels of talent and performance? Injuries? Trades? Happenstance?

Ok, I think anybody who watched the Finals in 2006 knew that the Mavs’ losing had nothing to do with point differentials or anything else.  That was the biggest bullshit Finals of my lifetime (not the biggest bullshit playoff series, of course…that’s either of those two Lakers series that we know and love so well).  Dwyane Wade was going to be an NBA champion whether he played like it or not.  I guess Stern decided that he’d rather have a team led by Wade win than a team led by a German.

Then they say this:

We think it’s important to go non-Trollinger on you here: This is a highly inexact science. Because this is sports.

Near as I can figure, a “Trollinger” is a stat geek.  But here’s the thing: this is not a “highly inexact science.”  No, it’s not “highly exact,” to borrow their nomenclature, because there are certainly variables involved in any of this stuff.  But while the numbers aren’t always exact, they certainly don’t do a lot of lying.  And the numbers say that a team that outscores their opponents by more points tends to win more games.  Der.  They go on to make a couple of points about intangibles and heart and bucking trends…and maybe all of that is possible.  But it sure as shit ain’t probable.  In fact, I’d be more than happy to make a bet on every NBA game with some stat-hater out there.  You can go with your “feelings” and I’ll bet based on point differential.  I’ll bet I end up with more cash than you…

Oh, and by the way…the Suns are 4th in the West in point differential.  The Mavs are last among playoff teams (and have outperformed their Pythagorean record [based on point differential] by seven games)Hmmm…