Was Brandon Roy an Offense-Killing Ball Hog?

We got a host of responses to last Friday's discussion of Brandon Roy's legacy with the Blazers. Among them were assertions that, wins and popularity aside, Roy's lasting legacy will be one of ball-hoggery and tempo-killing. I had suggested in the article that such accusations stem from the post-injury, molasses-kneed play from the Blazers' former franchise player. I wondered if folks had forgotten Roy's pre-injury greatness. Some responded by saying that the suggestion itself was revisionist history, that Roy always did dominate the ball to the detriment of his team and the offense.

This isn't a new debate. The topic comes up every time we talk about Brandon. Before we put this to bed entirely and wish Roy well as a member of the Timberwolves (and whatever other teams he ends up on) I wanted to take a statistical look at the issue.

Since the question involves Roy before his knee issues--pretty much everyone agreeing that Roy's offense was both slower and a shell of its former self after--we're limiting our data to his first three years in the league. We want to determine as best we can whether Roy was significantly more selfish, less productive, more of an offense killer than his peers. Is the "ball hog" label deserved?

Even those who aren't interested in this subject because Roy is no longer with the Blazers player might find the look at statistics interesting and applicable to other projects. Click through to see what we found.

We're going to begin the discussion with a stat called Usage Percentage. It measures the amount of plays a given player "uses" by looking at individual shots taken, free throws, turnovers, and minutes played in relation to the team as a whole.

It's important to understand that having a high Usage Percentage is not in itself a bad thing. Somebody has to take shots or the team scores nothing. Somebody's going to get use out of plays. All major stars have high Usage Percentages because the ball (rightfully) goes through them.

If a player has a high Usage Percentage you want to know two things:

1. What's his role? If he's supposed to be your main scorer, you're good. If he's your distributor or rebounding power forward and not supposed to be a scorer you may have a problem. As a shooting guard and acclaimed scorer, Roy doesn't have to worry about this issue.

2. More importantly, is he efficient? Anybody can have a high Usage Rating. All you have to do is chuck up a bunch of shots (a.k.a. ball hogging). The real question is, do you make those shots? The hallmark of a true star is efficiency remaining high when the Usage Rating goes up.

Of Roy's first three years in the league, I decided to run data from 2008-09. It was his highest-scoring season but it was also the year his Usage Percentage was the highest. If we're going to find flaws, their pinnacle should be here when Roy dominated the offensive play most.

The first thing I did was compare Roy's Usage Percentage to that of the leading scorer on every other team in the league...the collection of his peers. At the time Portland's offense was most Roy-centered, was he using more possessions than other leading scorers? Here's the table:

Among the leading scorers on this list, Roy sits in a tie for 11th as far as number of plays used. Interpretation is a matter of perspective. Those disposed to say he dominated the ball would call him "almost Top Ten". But in a list of 30, tying for 11th puts you closer to the middle of the pack than the top. The raw numbers agree with the latter assessment. Roy's number is closer to the 25th spot on this list than the 6th. His Usage Percentage is high, but not in Kobe Bryant or Carmelo Anthony territory.

But as we said, Usage Percentage alone isn't indicative. It's what you do with it that matters. No single stat can give a complete measure of productivity, so let's look at a few possible measures.

PER (Player Efficiency Rating)

This is John Hollinger's everything-and-the-kitchen-sink stat which measures per-minute production. Roy's PER of 24 in 2008-09 ranked him 6th among these 30 leading scorers. The only players above him: LeBron James, Dwight Howard, Dwyane Wade, Kobe Bryant, and Chris Paul.

True Shooting Percentage

This stat measures shooting efficiency, taking into account the value of three-point shots and free throws. Here Roy tied for 11th at 57.3%.

Offensive Rating

This estimates the points a player produces per 100 possessions. Roy's 123 puts him 2nd among these 30 players, trailing only Chris Paul (by a single point).

Offensive Win Shares

This estimates the number of wins a player contributes to his team via offense. Roy notched a 10.9, good for 3rd behind LeBron and Paul.

Assist Percentage

This estimates the number of his team's field goals a player provided assists for when he was on the floor. Roy's 25.4% ranked him 8th among the 30 team leading scorers. 4 of the 7 players ahead of him were true point guards. LeBron, Joe Johnson, and Stephen Jackson were the only non-point guards who assisted on a greater percentage of their team's shots than did Roy.

As you can see, Roy's weakest comparative standing was his 11th in True Shooting percentage. This was exactly equivalent to his Usage Percentage rank. By every other measure he ran far ahead of his peers in effectiveness compared to his usage rate. The only guys you'd consider above or in the mix with him overall were LeBron James, Chris Paul, Dwyane Wade, and Kobe Bryant. Granted those players, and some others, bent the scale in certain areas which would make them more valuable than Roy. You couldn't argue that the Blazers would be better off with Roy than James. But we're not comparing strict talent to talent here. Rather we're saying that given the rate of his touches, Brandon Roy was more effective with his possessions than almost any leading scorer on any team in the league and was rightly mentioned among the best of the best.

Consider these things also:

• In 2008-09 the Trail Blazers as a team ranked either first or or second in the league in points per 100 possessions, depending on how you measure.
• The Blazers ranked 14th in points per game and 19th in assists overall despite their 29th place finish in pace.
• The Blazers ranked 8th in effective field goal percentage and true shooting percentage overall.
• The Blazers ranked 12th in assist ratio, the percentage of possessions that finished with assists.
• Even as Portland's leading scorer, Roy took only 17 shots per game in that season. LaMarcus Aldridge took 15.5 by comparison.
• 2008-09 also marked the high point in Portland wins since 2000 and still ranks as the 6th highest win total in the 42 year history of the franchise.

In short, looking at the statistics it's difficult to see how Roy was hurting his team, depriving his teammates, hogging the ball, or killing the offense. The "revisionist history" here is remembering his final go-around with the team--limping up the court, eating clock because of immobility, putting up fade-away jumpers--and projecting that back through his entire Portland career. It just wasn't so. Once upon a time Brandon Roy was undeniably, unabashedly great as an individual and for this team. Yes, he got touches, but others did too...including Aldridge. At the end of the day nobody did as much with them, or did so much for the Blazers, as Roy did.

--Dave (blazersub@gmail.com)