2013 MIT Sloan Sports Analytics Conference

The 2013 MIT SSAC is happening. This is Daryl Morey's Las Vegas to his "Dork-Elvis".

ESPN has some coverage here with quips and quotes from the likes of Mark Cuban, Stan Van Gundy, Morey, and others.

Here is a paper on the value of 3's vs. 2's (if only it was as simple as it sounds)

Caution: reading this is like chewing on cardboard....old, dusty cardboard without anything to drink. Fortunately, they include a brief summary that includes most of their findings....but I do recommend reading it while considering the Rocket's offense...it helps visualize and understand their strange language of numbers.

Here is the summary:

We theoretically and empirically investigate the optimality of NBA shooting decisions in response to changing incentives over risk. The most robust theoretical requirement is that the gap in nominal efficiency between 3 and 2-point attempts must be negatively correlated with the offense’s preference for risk. We find adherence to our key test of optimality for the trailing team—as a trailing team gets in a more desperate situation (becomes more risk-loving), the efficiency of their 3-point attempts falls. The trailing team also tends to increase their fraction of 3-point attempts in proportion to their preference for risk, consistent with the ability to shift the offensive attack more flexibly than the defense can adjust resources. The leading team, however, violates our optimality requirement;

leading teams shoot fewer 3’s as their preference for risk increases and these 3’s actually offer higher average point value (consistent with a downward sloping usage curve). As a lead decreases, the leading team should become more risk-neutral, but teams in this circumstance actually tighten up and become more risk averse, contrary to what

their risk preferences ought to be to maximize the chance of winning the game.

We also find a strong motivational effect of trailing on the scoreboard—a given lineup sees a boost inefficiency for both 2’s and 3’s—combined with the sub-optimal shot selection of the leading teams this helps explain the surprising frequency of comebacks in the NBA and means clutch moments tend to occur more frequently that we’d otherwise expect. We show that for an average team it’s harder to score in clutch moments, but very good

offenses actually do better in the clutch, whereas bad defenses actually get worse. Taken together, this means good teams have an even greater advantage when the chips are down.

------------------------

Largely, this paper doesn't tell us much that intuition, reason, and experience don't already. The math supports the Rocket's approach to offense and further reinforces why we do what we do and why it is so successful. Perhaps, of all their findings the most interesting is the notion that teams leading down the stretch benefit from shooting more threes. Also, a team's defensive ability becomes less of a factor at the end of games vs. their offensive ability.

Something I feel they left out of consideration-- A referee's tendency to swallow their whistle in late-game situations....something James Harden needs to figure out. Those offense-initiated contact fouls go away at the end of the game--mostly.

Ultimately, they claim that a 3 point shot (1.5 x a 2pt. shot) does not always retain that value ratio. It can shift from 1.2 up to 1.8 depending on a multitude of variables. In analyzing those variables an offense can attempt to manipulate the game to maximize point/shot values. Thus, the "drive and kick" is actually a highly advanced basketball play with statistical data supporting its use as an offensive strategy.

If a team can limit/hinder the drive the ensuing kick becomes less effective as the game goes on....we saw this clearly in the recent loss to Washington. In the loss to Milwaukee player health limited the effectiveness of the drive (as did Larry Sanders) and, once again, as the game progressed the offense struggled.