Blazers' rookie guard Damian Lillard has hit the ground running. His strong play combined with early season injuries suffered by #1 pick Anthony Davis finds Lillard on top of NBA.com's Rookie Ladder and a darling of FanPosts here at the BEdge. The Blazers' PG prodigy is the team's first legitimate shot at bringing the Eddie Gottlieb Rookie of the Year (ROY) Trophy to Portland since Brandon Roy marched through a weak field in '06-'07. I thought it would be fun to consider Lillard's chances of upsetting Davis from a statistical point of view.
Although the small sample of games played so far (10 for Lillard, 6 for The Brow) includes the usual caveats, we can at least look to the award's history to understand what to watch in the statistical race to come. I examined the 33 seasons from 1979-80 (when Bird won a tight race with Magic Johnson) to see what winners had in common. Despite Lillard's strong early play, if the season ended tomorrow a Lillard ROY would be (statistically) unprecedented. In fact, even if Davis inexplicably hung them up tomorrow, history says Kidd-Gilchrist would be the statistical favorite with perhaps 7 in 10 odds of winning the Eddie G. As Blazer fans, here are the signs that Dame is in the race as the season goes on:
- He overtakes The Brow's (and MKG's) Player Efficiency Rating (PER)
- if #1 doesn't work out, Davis' PER at least drops below 20
- Davis' nagging injuries limit him to fewer than 70 games
Read on for my reasoning behind stats to watch and some fun ROY facts found along the way.
PER has called ROY correctly over 70% of the time
I started with 1979-80, since I can actually remember players from that era. Looking at the data, two unspoken baseline criteria for ROY became obvious:
- A sixth man has never won ROY (Mike Miller, a dubious recipient in 2000-2001 came off the bench the most among winners at just under 25% of games)
- You have to play to win (Brandon Roy won playing 57 out of 82 games, 70%; three other players have played 70 out of 82, 85%, and won the award)
With these trends in mind, I limited the field of candidates in each season to those who played in at least half of the season's games, and started more than they came off the bench.
Given data and time limitations, I considered four metrics that might predict ROY among each year's candidates: points per game, PER, win shares, and win shares per 48. PER turned out to be the best predictor, correctly calling 25 out of the 35 ROY winners (2 of the 33 seasons had co-winners) for an impressive success rate of 71.4%. Below is a plot of all 35 ROY winners (and co-winners) from Larry Bird (1979-80) through Kyrie Irving (2011-12).
The x-axis shows the PER differential between the ROY winner and the next closest rookie "candidate" (played 50% of games and started 50% of games played). The y-axis shows the ROY's PER on the season (source: basketball-reference.com). Points to the right of the bar are cases where the rookie PER leader won ROY. Points to the left are cases where the ROY was not the PER leader. I labeled a few outliers, mostly for fun. A few things are worth pointing out.
Only two ROYs with negative PER differential won when a competitor had a 20+ PER
Larry Bird (1980, over Magic Johnson) and Amare Stoudemire (2002-2003, over Yao Ming). Bird (20.5) and Magic (20.6) were essentially tied. One could make a good case that Yao was robbed. Poor Mike Miller was probably the most ignominious winner, posting a measly 13.2 PER in a weak class. I'll leave it to Marc Jackson (16.1) fans to make his case.
is the only [edit: Ewing only played in 50 games in 85-86, runner-up Xavier McDaniel] ROY to play fewer than 70 games (or equivalent in lockout years)
Roy had an average ROY campaign based on his PER of 18, but his class was so weak he won the award going away over closest rival Rudy Gay (12.4). It's ironic that if Davis were to win the award with fewer than 70 games played, Roy would be his precedent. It would get interesting, though, since Lillard (and perhaps MKG and others) project to be much stronger candidates than Roy's competition.
Anthony Davis' PER through 6 games puts him in elite company
The Brow's 25.9 PER would put him up top with MJ and The Admiral, an efficiency level he's unlikely to maintain season long. If Lillard keeps up his play, this race is likely to tighten.
If Davis continues at this pace and manages to play at least 70 games, history says he's a lock for ROY. Even if he regresses considerably, history says he almost certainly wins if he maintains a 20+ PER and keeps ahead of the pack. If injuries keep The Brow from cracking 70 games (he's now missed 3 of his team's 9 games), things get interesting. Only Brandon Roy has won ROY playing fewer than that in the past 33 seasons, and that was against a weaker class. If Lillard (or someone else) best Davis' PER, history gives them a better than 70% chance of winning. Emerging early as a legitimate candidate ceratinly can't hurt Lillard's chances should he outperform the competition. But, as it started, this race is still Davis' to lose.