The Turnover Question and How It Relates to Wins Produced

Remember the buddy who I challenged to a Strat-o-Matic all-star tournament?  Well, on Sunday, he also asked me an important question.  He asked me why people think that scorers like Kobe Bryant and Carmelo Anthony have value.

After my initial response, I thought about this for a while.  Then, a thought hit me, based on my experience looking at Carmelo Anthony’s per-48 minute stats found at  What if scorers’ statistics are being brought down because they turn the ball over a lot compared to the average player, in part because they have the ball more than the average player.

I looked at the statistics for every player this season who averaged 25.0 points or higher per 48 minutes but produced fewer than .150 wins per 48 minutes in 1,000 minutes or more.  There were 25 players in all.  Here, I will list their turnover rates, Usage Rates, and Points Per Shot rates compared to the league average, rounded to the nearest hundredth.   Turnovers and Points Per Shot are from, Usage Rates are from    For the sake of the argument, I assumed that the average Usage Rate is 20% for all positions, although this may not be the case.  They are in order of points per 48 minutes:

  • Kobe Bryant: Turnovers 1.63 times average, Usage Rate 1.79 times average, Points Per Shot 1.00 times average
  • Russell Westbrook: Turnovers 1.44, Usage Rate 1.64, Points Per Shot 1.03
  • Carmelo Anthony: Turnovers 1.42, Usage Rate 1.59, Points Per Shot 0.98
  • Dirk Nowitzki : Turnovers 1.04, Usage Rate 1.46, Points Per Shot 1.07
  • Derrick Rose: Turnovers 1.24, Usage Rate 1.53, Points Per Shot 1.03
  • Kyrie Irving: Turnovers 1.40, Usage Rate 1.44, Points Per Shot 1.08
  • LaMarcus Aldridge: Turnovers 1.08, Usage Rate 1.35, Points Per Shot 1.02
  • DeMarcus Cousins: Turnovers 1.56, Usage Rate 1.49, Points Per Shot 0.92
  • Andrea Bargnani: Turnovers 1.19, Usage Rate 1.44, Points Per Shot 0.97
  • Deron Williams: Turnovers 1.61, Usage Rate 1.51, Points Per Shot 1.00
  • Tony Parker: Turnovers 1.15, Usage Rate 1.39, Points Per Shot 1.04
  • Louis Williams: Turnovers 0.61, Usage Rate 1.37, Points Per Shot 1.02
  • Al Jefferson: Turnovers 0.57, Usage Rate 1.29, Points Per Shot 0.88
  • Danny Granger: Turnovers 1.04, Usage Rate 1.30, Points Per Shot 0.99
  • Monta Ellis: Turnovers 1.53, Usage Rate 1.44, Points Per Shot 0.97
  • Tim Duncan: Turnovers 1.04, Usage Rate 1.31, Points Per Shot 0.97
  • David Lee: Turnovers 1.26, Usage Rate 1.30, Points Per Shot 0.96
  • Brandon Jennings: Turnovers 0.88, Usage Rate 1.30, Points Per Shot 0.94
  • Kevin Martin: Turnovers 1.04, Usage Rate 1.24, Points Per Shot 1.04
  • Jordan Crawford: Turnovers 1.50, Usage Rate 1.41, Points Per Shot 0.91
  • Marcus Thornton: Turnovers 0.88, Usage Rate 1.17, Points Per Shot 0.99
  • Amare Stoudemire: Turnovers 1.30, Usage Rate 1.27, Points Per Shot 0.99
  • Josh Smith: Turnovers 1.36, Usage Rate 1.42, Points Per Shot 0.92
  • Joe Johnson: Turnovers 1.00, Usage Rate 1.25, Points Per Shot 0.98
  • Drew Gooden: Turnovers 1.33, Usage Rate 1.29, Points Per Shot 0.92

I know that my methodology isn’t perfect.  I’m sure that time spent with the ball would be a better statistic than Usage Rate, but I know where to find Usage Rate and not time spent with the ball.  I only used one season’s worth of data.  However, I did have some interesting findings.

For one, only 4 of these players had turnover rates that were lower than average: Louis Williams, Al Jefferson, Brandon Jennings, and Marcus Thornton.  On the other hand, all 25 of these players had Usage Rates above 20%.  Then again, only 10 of these players had average or above-average Points Per Shot rates.

However, there was one especially intriguing and important discovery.  Of these players, only 6 (DeMarcus Cousins, Deron Williams, Monta Ellis, Jordan Crawford, Amare Stoudemire, and Drew Gooden) had Usage Rates that were a higher rate above average than their turnover rates.  This leads me to conclude that their Wins Produced numbers have been deflated, at least in part, because they have high turnover rates, which are largely caused because they use more possessions than the average player.

Now, don’t get me wrong-I’m not trying to defend Kobe or Carmelo.  Like I said; only 40% of these players had Points Per Shot rates that were above-average for their position.  However, their skill level according to Wins Produced may be decreased because of this phenomenon.  I believe that we should consider tweaking the Wins Produced formula to include Usage Rate-adjusted turnovers instead of strictly turnover rate.  When I say consider, I mean it should be tested out to see if it positively impacts the formula.  If it does, even by a barely significant amount, please include this.  This way, we truly know if Carmelo Anthony is as overrated as the current version says.  Thank you for reading, and please comment.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s