Tuesday, May 3, 2011

Arguing against the "Kobe isn't Clutch" conventional wisdom

I told Michael Grange I needed more than 140 characters to make this argument, so I moved from Twitter to this dusty ol' blog.

A quick primer: Kobe Bryant's poor late game performance in Game 1 against the Mavs renewed the "Kobe isn't clutch" argument. This argument is based on one key set of statistics that show, on game-winning shot attempts, Kobe shoots right at the league average (somewhere just below 30%).

First warning sign for me: Everyone seems to agree with this argument (except for Laker fans). This rule comes from my penchant for gambling but whenever so many people seem to agree (so vehemently) about a particular issue, that usually tells me the truth probably isn't so concrete.

Before I present some counter-statistics, let's look at two weaknesses that pop up when you look at the numbers these conventional wisdomers always cite (note: "wisdomers" not a word).

First, who decided that field goal percentage is the be-all & end-all when it comes to "clutchness"? As I said to Michael, this is like ripping on the NHL leader in game-winning goals because he also had the most potential game-winning attempts (he didn't buy this argument, but you might).

If you've played team sports at any level, there's a pecking order on each team in terms of who you look to throughout the game. But, there's also a pecking order late in the game that doesn't always perfectly correspond with that first list. Some players who don't look to score for most of the game tend to rise up in big moments. Conversely, some big-time scorers tend to shy away from the big moments, and prefer to act as a decoy or passer (more on this later).

Secondly, who decided that game-winning shots is the best measure for "clutchness"? This strikes me as a very narrow-minded definition. The best analogy for the people who make this argument are the people who believe in the (overrated) value of a closer in baseball. Isn't there just as much value in hitting a shot to put your team up 2 with 90 seconds left? Or 2 minutes left?

With that in mind, let's look at 82games.com Clutch statistics --- defined as 4th quarter or OT, less than 5 mins left, neither team ahead by more than 5 points.

This season, Kobe had a league -leading 50 pts per 48 minutes under those conditions. Here are his pts per 48 minutes (and league rank) over the past 4 seasons:

10-11 49.8 (1st)
09-10 51.2 (2nd)
09-09 56.7 (1st)
07-08 51.8 (2nd)

Some more points about this year's statistics:

- Some players thought to be more "clutch" than Kobe in most circles still shoot a higher percentage than him, but take significantly fewer attempts per 48 mins.

K. Bryant 39 fga, 40 fg%
C. Anthony 30 fga, 46 fg%
D. Nowitzki 24 fga, 46 fg%

As I've said before, detractors of Kobe will say he forces too many shots, I say most of the other players shy away from big shots (quick aside: D. Rose averages 37 fga per 48 mins in clutch... does anyone say he forces shots?).

- Kobe also takes hits for the fact that he has just one game-winning assist, and this is another bullet used as evidence for the "forcing too many shots" argument. As it turns out, Kobe averaged 7.6 assists per 48 minutes this season in the clutch. Here's how that number ranks with the 10 other players that average at least 40 pts per 48 mins in clutch (min 70 mins):

Derrick Rose 9.8
Kobe Bryant 7.6
Russell Westbrook 7.0
Dirk Nowitzki 5.6
LeBron James 4.9
Monta Ellis 4.0
Kevin Durant 2.3
Carmelo Anthony 1.3

So Kobe creates more for teammates in the clutch than Russell Westbrook, LeBron James, and Kevin Durant. Are all those players considered ball hogs as well?

- Finally, Kobe's net plus/minus in the clutch over the past four seasons: +192.

In conclusion, this argument was not meant to prove that Kobe is the "best" or "most" clutch player in the NBA. What this argument intended to do was disprove the theory that Kobe is simply an average player in the clutch.

I think I've successfully accomplished that wish. And it only took me about 14,000 characters.