We hear it all the time. NBA analysts call one point guard “pass-first” and another “shoot-first.” Or they say one guy is “turnover-prone” while another “takes care of the ball.” But really, what makes a player a “pass-first” point guard? How carefree must he be with the ball to be considered “turnover-prone”?

Taking an analytical approach to these questions, I decided to bust out an Excel spreadsheet and try to come up with some answers. Below you’ll see a graph that attempts to classify the top point guards in the league. But first, a little background…

I chose to categorize each player based on two stats. First, to determine if he’s “shoot-first” or “pass-first,” I calculated the shot-to-assist ratio for each player. The bigger the number, the more of a “shoot-first” mentality the player has. Second, to determine whether or not a player is “turnover-prone,” I calculated each player’s assist-to-turnover ratio. I thought about using turnovers per 48 minutes, but I like the idea of including assists so that playmakers are rewarded for the positive as well as the negative. Next, I calculated each player’s Efficiency Per Minute (EPM) to see if there is any correlation between these other statistics and the overall efficiency of the player in question.

The last step was to plot each player based on the two ratios. The graph is a little small, but if you click it, it will take you to a larger version.

Each player is indicated by a separate plot point with his name next to it. (Duh.) The blue diamonds represent the top 10 point guards in terms of EPM. One point guard was chosen per team. Usually, I went with the player that got the most starts at the position.

The further the player is to the right on the (horizontal) x-axis the higher his shot-to-assist ratio and the more of a “shoot-first” mentality he has. Conversely, the further the player is to the left, the more of a “pass-first” point guard he is.

On the (vertical) y-axis, the further the player is towards the top, the better job he does of taking care of the ball (in comparison to the number of assists he posts). Players towards the bottom of the axis have poor assist-to-turnover ratios.

Each axis is set at the average of the 30 players included in the study. The shot-to-assist ratio average is 2.12, while the average assist-to-turnover ratio is 2.66.

A few interesting things to note…

As a whole, the data set takes on a downward slope. Intuitively, this makes sense. “Shoot-first” point guards are generally that way because that is their strength – scoring the ball. They are generally not as adept at being in the playmaker role, so they won’t have a good assist-to-turnover ratio. Conversely, players that are “pass-first” are that way because playmaking is their specialty. They handle the ball well and have good vision.

The sweet spot is in the second (top-left) quadrant. Seven of the top EPM performers at the position – Jose Calderon, Jason Kidd, Chris Paul, Rajon Rondo, Deron Williams, Steve Nash and Chauncey Billups – all have a better-than-average assist-to-turnover ratio and a “pass-first” menality. The other three top 10 EPM performers – Jameer Nelson, Tony Parker and Devin Harris – are reasonably close to that top-left quadrant. The 10 point guards in the top-left quadrant play for teams that had a combined record of 471-349 (.574) in the regular season and seven of those teams made the playoffs. These facts are telling.

Jameer Nelson and Tony Parker aren’t in the ideal quadrant, but they are a good fit for their teams. Think about it – Nelson and Parker play with two of the best big men in the game in Dwight Howard and Tim Duncan. Point guards that play with a great big man will find that assists are a little tougher to come by. Typically, when someone feeds Howard or Duncan, they take their time in the post before they make their move, so the player that passed them the ball doesn’t necessarily get the assist. Plus, anyone can feed the post – it doesn’t have to be the point guard. Think about those seven players in the top-left quadrant for a moment. For the most part, they play with jump shooters (and are therefore more likely to rack up assists).

The Magic are built in the Hakeem Olajuwon-era Rockets mold. They have a dominant big man and the idea is to surround him with great jumpshooters to keep defenses honest. Conversely, Parker oftentimes takes advantage of the amount of attention that defenses pay to Duncan to get to the rim. When he attacks the hole, he’s trying to score and will only pass if the defense forces him to.

Finally, it’s no coincidence that Nelson and Parker are two of the most efficient scorers in this study. In terms of FG%, Parker is first and Nelson is tied for third. Both players shoot better than 50% from the field, so it makes sense that they’d have a “shoot-first” mentality.

Ramon Sessions is a keeper. Luke Ridnour got more starts than Sessions, but I like Sessions more and the Bucks have to decide whether or not to sign the up-and-coming point guard to an extension this summer. Here’s my pitch: He has the #11 EPM (just behind Chauncey Billups) and is on the verge of becoming a very good point guard. His Player Efficiency Rating (PER) of 17.65 is #13 amongst qualified PGs. Throw in the fact that the NBA is becoming increasingly dependent on good point guard play, and the 23-year-old looks like a great (and relatively cheap) prospect. GM John Hammond has to find a way to keep this kid.

Baron Davis isn’t a “shoot-first” point guard?!? Apparently not. His career shot-to-assist ratio is 2.05, so even when we consider his entire body of work, he’s still leans “pass-first.” This is underlined by his 2008-09 campaign, where he was even more “pass-first.”

Derrick Rose turns the ball over too much. I hadn’t yet looked at his numbers, but I could tell this about Rose just watching the Bulls/Celtics series. Every so often he throws a pass that might have made it in college, but isn’t going to make it in the NBA. Most rookie point guards struggle with this, so I’d assume that Rose will start to work his way into that top-left quadrant (or at least the top-right quadrant) in the next year or two. In fact, his assist-to-turnover ratio rose from 2.52 before the All-Star break to 2.56 after. I think he’ll eventually settle in somewhere near Nelson and Parker. He seems like more of a scorer than a pure playmaker, partly do to his incredible ability to finish at the rim.

Where would a healthy Gilbert Arenas fit in? Remember him? Agent Zero has only played 15 games in the last two years, so it’s not really fair to judge him based on those numbers. To get an idea of where he’d land, I calculated his ratios for the 2006-07 season. You’ll find him bunched with Jamal Crawford, Mo Williams, Randy Foye and Aaron Brooks in the bottom-right quadrant, which makes sense because those are all “shoot-first” point guards.

Is Mo Williams’ position on the chart due to his playing with LeBron? After plotting Arenas, it made me wonder about Mo. He was one of the top assist men in the league in his last year in Milwaukee. How does the 2007-08 Mo Williams compare with the 2008-09 version? It turns out that the ’07 version of Mo Williams is a lot like the ’08 Devin Harris, which makes sense considering that both players are very capable scorers and playmakers, but at heart are “shoot-first.”

CONCLUSION

I think the big thing to take away from this is that it’s generally better to have a “pass-first” point guard who takes care of the ball. However, if you have a dominant big man (or a superstar wing like LeBron or Kobe), a “shoot-first” point guard can be just as effective, especially if he can shoot it like Nelson or Williams, or get to the rim like Parker does.

Is there anything that I missed?