For years now, I have been arguing that the basketball community uses an outdated and flawed definition of positions that can create problems. Since we have a league of players with massively different offensive abilities that are uncorrelated to height and weight (think of Dirk Nowitzki, LeBron James and Shaun Livingston), using what a player can do on offense to decide what position he plays can lead to mistakes and mismatched talent.

Over the past few weeks, two of my favorite writers on the Warriors (Tim Kawakami and Ethan Sherwood Strauss) have asserted that the Golden State Warriors had been better without David Lee. While this case has some merit, I see a larger underlying issue that affects the debate: the Warriors have played David Lee out of position for a vast majority of his tenure with the team.

While David Lee has many problems on defense (anyone who has either watched the Warriors or seen or heard about Kirk Goldsberry’s presentation at Sloan 2013 has a pretty good idea), many of those issues have the root cause of Lee playing a position that allows opponents to take advantage of him. While he possesses below average quickness and agility for a modern NBA power forward (especially the new breed coming into the league recently), he rises up the athleticism charts when compared with NBA centers. This shift helps mask his faults on pick and rolls since lumbering bigs do not have the ability to get that extra step on him that opponents have exploited so often in recent seasons.

Furthermore, changing that athleticism weakness into a strength by changing who guards him transforms Lee into a much more dangerous offensive player since he is smart enough on the court to use it to his advantage. The shocking lack of true back to the basket scorers in the NBA today means that there are less players built to use Lee’s size deficiency at center against him and the team has Andrew Bogut (or Dwight Howard…) to defend those guys.

Fortunately, we also are not flying blind when it comes to understanding how this would work since Lee has played quite a bit of center during his eight-year career. The indispensable separates out how a player and his opponent do when he shifts between positions, creating an incredibly useful tool to analyze the situation.

Rookie season: Did not play any C- got a majority of his minutes at Small Forward (oops)

Second season: Played a majority of his minutes at PF but had a better PER and net PER (meaning the difference between how he did and how the person he guarded performed) at center.

Third season: Finally stopped playing SF. Better personal and opponent PER against centers than power forwards.

Fourth season: Plays vast majority of minutes at center for the first time. Has better personal and opponent PER against centers than power forwards, with a positive net PER at C and a negative one at PF.

Fifth season: Again spends most of his time at center. Fueled by a meaningful improvement in defense with the position shift, Lee has a better net PER at center than power forward.

Sixth season (first with Golden State): Returns to playing mostly power forward. Has a nearly equal PER at PF and C, but has a nearly four point decline in opponent production at center, creating another big difference in net PER.

Seventh season: Again plays more than twice as many minutes at PF than C. Actually has a better net PER at power forward, though he produces a remarkable 24.2 PER at center (this would have been tied with Dwight Howard for sixth in the entire NBA at any position).

Eighth season (the one that just finished): Closer split in minutes because of the Bogut injury. Better net PER at C again, fueled by Lee being almost three points better at center.

In nearly every season where he played both positions, David Lee did a better job at center than power forward. It cannot be dismissed as a small sample size since many of his Knicks' seasons had him logging more minutes at C than PF. Furthermore, the massive roster turnover that both New York and Golden State had during his tenures means that the results has less bias because of playing with or without certain teammates that could distort the results.

Simply put, David Lee plays better basketball when he plays center than power forward whether you use the eye test or advanced statistics.

What this means moving forward:

Having a legitimate starting center (whether that be Bogut or Howard) makes going with a more novelty center like David Lee a useful luxury. Practically applied, that means that the Warriors should be open to moving their All-Star even though the market for him might not be perfect due to his high contract and the fact that so many teams are angling for space in the big 2014 free agent class.

Either way, holding onto Mr. Third Team All-NBA has its advantages. I would make his primary role one that very few teams have ever employed for a big man: backup center who also frequently initiates the offense. While he would still play the full slate of minutes (thus rolling in with the starters for long stretches over the course of the game), my main task of Lee would be to be the lynchpin of the team when Curry and Bogut sit.

There are two basic philosophies behind bench play in the NBA: rotations and line changes. In a rotation system, ideally close to interchangeable parts swap in and out when players are tired keeping the system largely intact. Unfortunately, the Warriors cannot and will not find a player who can do what Stephen Curry does when he sits. As such, having a player on the roster who can create for others and generate his own offense would keep other teams on their toes during the times when Golden State was most vulnerable this season, particularly in the playoffs when opponents get to face them and figure out how to exploit those times. 

This shift in primary responsibility makes Lee the anchor of a second unit that can help accentuate his strengths and mask his weaknesses. A bouncy, athletic power forward can handle some of the more dangerous 4 and 5 men while shooters (like Klay Thompson and Brandon Rush) and versatile defenders (like Draymond Green and Brandon Rush) already have places on this team. It also enables Golden State to do what they do best with the starting lineup: a more dangerous and creative version of the “4 Out” offense that Orlando utilized a few years ago with a more capable and potent player running the show. Plus, it allows the Warriors to ask a little less of their backup PG even though having one that can take over for Curry would certainly be a good idea if available.

Making this the main directive for a leader like Lee would also allow him to make an imprint on a group that sorely needs it. Bogut and Curry should be the faces and key cogs of the starting lineup no matter what and they represent the ceiling of this team. We learned in the playoffs this year that Curry should be initiating when he is on the floor and players who take away from that ball control should either manage that part of their game or take on other roles. Moving David Lee’s minutes helps open up the floor with a PF who can make threes and could actually even lead to some interesting lineups where he replaces Bogut and makes a “4.5 Out” system to keeps opponents on their heels.

Some may ask what happens to Festus Ezeli if this shift occurs. The answer would be that he acts as the safety valve for when either Bogut or Lee gets injured. Considering their histories, this means there will be plenty of minutes for the developing big man.

Creativity and a system built on maximizing the distinct strengths of this team stands as the best and only likely way for the franchise to step up another level in the Western Conference. Building on the idea that their better is better than other teams can handle can make them the toughest out in the playoffs with the right moves and development.