Statistical Analysis. Humor. Knicks.

Tuesday, December 23, 2014

Five Stats the NBA Should Keep (Part II)

[This is the second of a two part series. Part I contains stats 1-3, so just scroll down to read it, or click here for the scrolling impaired.

At the time of this writing there is a news rumor that Chris Webber got traded to the 76ers for Kenny Thomas, Corliss Williamson, Brian Skinner, and a bunch of junk from Pat Croce’s garage. Considering the fiasco over the Carlos Boozer non-trade, and that three networks picked up the false story that Shaq’s season was over, I’m going to wait until it’s official to comment.]

4. Defensive Shooting Stats (DFGM, D3P, DFT)
Allen Iverson uses a crossover dribble to get past Chauncey Billups. Billups follows him to the hoop, but ends up fouling Iverson on the layup, creating a three point play for the mercurial guard. The next time down the court Iverson fakes the drive and pulls up for the three. This time, Billups is not fooled and gets a hand in Iverson’s face. The ball clangs off the iron and Big Ben cradles the rebound.

Although 82games.com keeps track of opposing defensive stats, why doesn’t the NBA make it an official stat? Keep track of field goals, free throws and three pointers for the defensive player who is the primary defender. In the above example, Billups would have the following stats:

DFGM: 1
DFGA: 2
D3PM: 0
D3PA: 1
DFTM: 1
DFTA: 1

Or in other words, Billups allowed 1-2 from the field (DFG: 1-2), defended well against the only three pointer attempted (D3P: 0-1), and let his opponent have a free throw (DFT: 1-1).

Are there flaws in this system? Sure. Basketball defense is partially a team effort, and assigning credit or blame to an individual seems unfair. Using that same argument, how would we judge pitchers if the lords of baseball decided not to keep track of ERA because there are 8 other players who assist him in preventing runs? Maybe the difference between having a great defense in a pitchers park is similar to a shooting guard having Kirilenko or Duncan play behind them. As for the criticism that it’s impossible to judge who is responsible for allowing the score, 82games.com has found a way and RealGM’s Kevin Broom has started to keep track of his favorite team with defensive box scores.

Using only the current stats the NBA uses (blocks & steals) doesn’t give us a complete picture of a player’s defensive skill. Blocks don’t always correlate with defensive ability. The #1 team in blocks/game is Portland, who ranks 16th defensively. Chicago is the third best defensive team, but they’re 18th in blocks. As for steals, Kevin Broom has a theory that they don’t indicate much about team defense because “the best defensive teams force misses, and usually force some turnovers as well…[but] a steal happens on less than 1-in-10 defensive possessions.”

While some good defensive players (Ben Wallace) gets lots of steals and blocks, there are plenty of good defenders (Bruce Bowen) who don’t have any numbers to back up their ability to clamp down on an opponent. Study after study has shown that field goal percentage is the primary key to defensive ability. Defensive shooting stats will give us a better insight as to who is holding their own & who isn’t.

5. Possessions (POSS)
The Phoenix Suns are the NBA’s fastest team, averaging 99 possessions per game. According to points per game, they are the worst defensive team in the league, giving up 102.3 points per game. However, that’s an unfair label, because their opponents get more opportunities courtesy of the Suns nuclear offense. The Suns defense is actually 17th, when considering Phoenix’s fast pace. Similarly, Seattle’s offense doesn’t crack the top 5, because their team crawls at only 92 possessions a game. In reality the APBRSonics are the league’s second best scoring machine.

Of course I can say all that because I’m using an approximation of possessions. A possession ends when the ball exchanges hands between teams, either by made shots or missed shots rebounded by the defense (FGA – OREB), turnovers (TO), or free throws that end a possessions. It’s that last factor that causes a problem, because you can either shoot 1, 2, or 3 free throws depending on the type of foul. While .44*FTA is the current approximation of choice, it’s just that an approximation.

The NBA should officially keep track of possessions, seeing that they’re already doing so in the WNBA. More importantly, the NBA should make possession based stats part of the basketball vernacular. Saying the Suns allow the most points per game is a fact, but the implication that their defense is also the worst in the league is false. If Phoenix took full advantage of the shot clock like the Spurs or Pistons, their points allowed per game would drop dramatically. In fact their defense would be in the top 10 in points allowed per game (94.8) if they only had 91.5 POSS/G.

And why should the NBA stop with keeping track of possessions for teams? The NBA should keep POSS for players as well, split between offensive & defensive possessions. Maybe Ray Allen’s 24 points per game is more valuable than Stoudemire’s 26, because Amare the Great has more possessions in which to score? This would also be useful in seeing how coaching tendencies change with different players on the court. Maybe with Earl Boykins on the court, Denver uses up more possessions per minute, and therefore runs a quicker offense. Phoenix might slow things down with Barbosa & Hunter on the court instead of Amare & Nash. Making stats possession based will be a step in making a level field between slow and fast paced teams.

One comment on “Five Stats the NBA Should Keep (Part II)

  1. Pingback: KnickerBlogger.Net » Blog Archive » Five Stats the NBA Should Keep (Part I)

Comments are closed.