James Harden and the Millsap Doctrine: minutes still don't matter

Daniel Shirey-US PRESSWIRE

Surprised James Harden is playing like a potential scoring champ in Houston? You shouldn't be. Part of his rise is a classic case of the Millsap Doctrine.

The Millsap Doctrine (n.): the theory that a players' per-possession production will remain similar given a boost in minutes played.

Back in 2008 and 2009, the basketball world was just beginning to get its collective head around the concept of efficiency, John Hollinger's PER and other advanced metrics. There was legitimate, informed debate on any number of pieces of the puzzle. One debate was over the efficacy of per-minute measures, which form the basis of PER. The skeptical side suggested that per-minute figures don't effectively represent a player's production because of several reasons: high production in low minutes could come against lesser competition (bench players), low minutes suggests a glaring weakness coaches can see but stats can't quantify and that given more minutes players would work less hard or not try to do as much. Valid suggestions, all.

But even when you considered those, the math showed that by and large players given more minutes would continue at similar levels of per-minute production. Kevin Pelton really led the charge on the math end, and I helped promote the idea that smarter people than I already espoused. We called it the Millsap Doctrine, based on its genesis: Paul Millsap doing incredible work as a starter while Carlos Boozer was injured in the 2007-08 season, which was totally expected if you'd looked at Millsap's per-minute stats when he'd been a reserve exclusively.

So really, I'm having trouble seeing why anyone is so surprised that James Harden has blown all the way up.

***

Okay, so no one predicted Harden would be averaging 35 points per game. And he won't all season long. But he's an easy 20-ppg scorer, and my suggestion he could challenge old friend Kevin Durant for the scoring title is looking less like a lark. He's a scoring machine. The funny thing is that he was a scoring machine when he was backing up Thabo Sefolosha in Oklahoma City. He just played fewer minutes than other scoring machines, and thus saw his total per-game scoring numbers sit lower than players he was clearly better than.

Harden averaged 31 minutes per game last season -- high for a reserve, but on the far low end for major scorers. He ended up at 16.8 points per game, No. 27 in the league. In per-minute scoring, though -- the common shorthand is the translate per-minute scoring into points per 36 minutes -- he finished within 1.5 points of the top 10. (There were a dozen players bunched up around 19-20 points per 36.) Clearly, despite the lower per-game scoring production, Harden was an elite scorer.

In Houston, he's playing lots of minutes and taking lots of shots. Part of the scoring explosion thus far has been the minutes growth: he's playing a league-high 41.7 minutes per game through three. Even if Harden's per-minute scoring production would have remained level from 2011-12 to this point in the season, he'd be averaging about 22 points per game. That's the Millsap Doctrine effect: a boost of up to six points per game this season. That's a huge amount that figures to stick around so long as Harden plays long minutes.

The rest of the difference is found in Harden's increased shooting frequency. This isn't explained by the Millsap Doctrine. In OKC last season, Harden had 14 shooting possessions per 36 minutes. This season so far, he's up to 23. That's a massive spread, and thanks to Harden's superb efficiency, it explains the other 13 points in his boosted per-game scoring average.

This piece needs to be read with that "so far" heavily emphasized: this is not a normal occurrence, for a player to boost their usage rate so massively and maintain efficiency. Dean Oliver pioneered the concept of skill curves in basketball. A skill curve is a basically a line showing recorded and estimated efficiency levels at varying usage levels. The rule is that increases in usage past a certain point result in lowered efficiency. It's the law of diminishing returns mixed with some basketball theory: if you take a good shooter and feed him all possible clean shots, he'll shoot them well. If you force him to take more shots, they are probably going to be worse shots (well-guarded, off-the-dribble, late in the shot clock), so his efficiency will fall.

That hasn't happened with Harden: through three games, his usage has shot through the roof, but his efficiency has remained high. The minutes gains -- the Millsap Doctrine gains -- are real, and will stick so long as he plays a lot of minutes. The days of James Harden averaging fewer than 20 points per game are long gone. But the gains made by massively increased usage while maintaining high efficiency -- let's call that the Harden Quandary -- remain unreliable for now. We'll check back in after 20 games to see if the magic is still there. If so, we may have to invocate the Harden Rule and get about finding more candidates.

***

The Hook is a daily NBA column by Tom Ziller. See the archives.

X
Log In Sign Up

forgot?
Log In Sign Up

Forgot password?

We'll email you a reset link.

If you signed up using a 3rd party account like Facebook or Twitter, please login with it instead.

Forgot password?

Try another email?

Almost done,

By becoming a registered user, you are also agreeing to our Terms and confirming that you have read our Privacy Policy.

Join SBNation.com

You must be a member of SBNation.com to participate.

We have our own Community Guidelines at SBNation.com. You should read them.

Join SBNation.com

You must be a member of SBNation.com to participate.

We have our own Community Guidelines at SBNation.com. You should read them.

Spinner.vc97ec6e

Authenticating

Great!

Choose an available username to complete sign up.

In order to provide our users with a better overall experience, we ask for more information from Facebook when using it to login so that we can learn more about our audience and provide you with the best possible experience. We do not store specific user data and the sharing of it is not required to login with Facebook.