Observing NBA Point Guard Performance with Pitcher-Type Statistics

Charles BennettSenior Analyst IAugust 22, 2012

Tony Parker had 29 quality performances in 2011-12; Derek Fisher had 3
Tony Parker had 29 quality performances in 2011-12; Derek Fisher had 3Tom Pennington/Getty Images

The basketball position most analogous to a pitcher in baseball is the point guard. Pitchers control the tempo of the game when they are on the mound; point guards control the tempo of their team's offense when they are on the floor.

Of course, the analogy isn't perfect. Pitchers generally are only important on the defensive side of things; point guards are important on both ends but gain most prominence on offense.

But since they are somewhat similar, stats should exist to track point guards' performance at the position. There are a plethora of pitcher's stats, but very few point guard stats beyond assists.  

I think it goes without saying that adapting something like ERA or WHIP wouldn't work. The analogous statistics are offensive or defensive rating, which don't tell the whole story, or player efficiency rating, which isn't a point guard-only stat, and indeed it somewhat penalizes point guards.

Another stat commonly used to track pitchers is record. Of course, win-loss isn't perfect; a good pitcher could have a good night and lose 1-0, or a bad night and win 14-10. I do think a point guard's win percentage is an important stat, but it suffers from the "cheap win-tough loss" phenomenon.  

There are point guards like Mario Chalmers who played on teams that won a lot but didn't help their team. On the other hand, the great Steve Nash played on a lottery team last year.

The "fixed" version of win-loss for pitchers is quality start. Quality starts entail going six or more innings and allowing three or fewer runs.

The problem with a quality start is that it penalizes certain types of performances. In the case of players like Derek Fisher, applying quality start to point guards doesn't always account for their floor leadership.

So should there be a stat similar to quality starts for point guards? I think there should, and it should combine aspects of the wins and the quality start.  

Let's hypothetically call this a "quality performance." First off, as with quality start, you need a minimum amount of playing time. Thirty minutes seems about right.  

You'd also need a minimum performance threshold of what designates a halfway-decent game for a point guard, while not penalizing high-performance games with "fewer than" stats. I think that's five points and five assists.

Finally, add the aspect of marshaling the troops to win the game and you're good to go. Do all those things, and you've got a quality performance.

Here are some numbers on how some of the NBA's point guards did in the quality performance stat, with the number of games included:

Point Guard



Chris Paul, L.A. Clippers



Tony Parker, San Antonio



Rajon Rondo, Boston



Derrick Rose, Chicago



Russell Westbrook, Oklahoma City



Steve Nash, Phoenix



Deron Williams, New Jersey



John Wall, Washington



Jose Calderon, Toronto



Kyrie Irving, Cleveland



Mario Chalmers, Miami



Jason Kidd, Dallas



Derek Fisher, LAL/OKC




As you can see, good point guards on good teams do well in this stat, poor point guards on poor teams do not, and the rest split the difference. It's also fairly well correlated to performance on All-NBA Teams, with the exception of Westbrook, who fell short in the assist column too many times. Thus, I think it is not a bad stat for determining the success of point guards.

Also, an interesting thing to note: Deron Williams was not particularly penalized by playing on a sub-.500 New Jersey team, as he recorded a quality performance in all but one of New Jersey's wins. The same applies to John Wall, who had 19 quality performances on a Wizards team that won just 20 games.  

By contrast, even though wins is an aspect of the stat, the fact that Miami's Chalmers didn't always meet the minimum assists (or even minutes) standards puts him relatively low on the list. Kyrie Irving was partly hurt by his team not winning, and partly because he had a number of 27- to 29-minute games. 

Jason Kidd scores low because he only played 30 or more minutes 21 times in the 2011-12 season, and he only scored in double figures nine times.

Bottom line: A pitcher-like stat such as quality performance needs to be adopted for point guards. "Quality performances" is that stat.