Milwaukee's Brandon Jennings has never fallen from the top spot in my rookie rankings. He's come close--Tyreke Evans nearly passed him a couple of weeks ago--but it's been Jennings all along. Sadly, though, I seem to be an outlier on this topic. The Rookie of the Year race has been increasingly referred to as a two-horse race between Sacramento's Evans and Golden State's Stephen Curry.
The race really comes down to what you value. Unfortunately, many of the media members who will ultimately vote on the award will base their selection on scoring average. This article is primarily aimed at those voters, but also will touch upon the different conclusions reached by those who adhere to more advanced systems.
Player G FG% PTS
Tyreke Evans 64 46.1% 20.3
Stephen Curry 69 46.3% 16.5
Brandon Jennings 70 37.2% 15.8
Reggie Williams 13 54.6% 14.8
Marcus Thornton 64 45.4% 13.8
Jonny Flynn 72 41.2% 13.4
Darren Collison 67 46.0% 11.6
Omri Casspi 69 45.0% 10.6
James Harden 64 39.5% 10.1
Jonas Jerebko 69 48.1% 9.3
Wesley Matthews 72 48.3% 9.1
Ty Lawson 56 51.6% 8.9
Taj Gibson 70 49.6% 8.7
DeMar DeRozan 65 47.6% 8.5
Tyler Hansbrough 29 36.0% 8.5
I included each player's field-goal percentage because for a lot of voters, I'm sure that will be the factor that disqualifies Jennings. If you're a regular reader of Basketball Prospectus, you already know that scoring average is merely a jumping off point in player evaluation, and not a very good one. You also know that raw field-goal percentage only tells part of the story when judging shooters. However, Jennings does have some ground to make up in other areas.
Player TS% eFG% PTS
Tyreke Evans 53.4% 47.6% 20.3
Stephen Curry 56.7% 53.5% 16.5
Brandon Jennings 47.6% 43.1% 15.8
Reggie Williams 66.4% 62.7% 14.8
Marcus Thornton 55.3% 52.3% 13.8
Jonny Flynn 50.3% 44.9% 13.4
Darren Collison 52.8% 48.7% 11.6
Omri Casspi 53.3% 50.6% 10.6
James Harden 54.7% 47.6% 10.1
Jonas Jerebko 54.4% 51.0% 9.3
Wesley Matthews 59.0% 53.8% 9.1
Ty Lawson 60.6% 56.1% 8.9
Taj Gibson 51.9% 49.6% 8.7
DeMar DeRozan 53.7% 48.1% 8.5
Tyler Hansbrough 44.8% 36.0% 8.5
This tells us more about how efficiently our rookies have been getting their points. True Shooting Percentage accounts for the extra points a player gets from the line and from three-point range. Effective field-goal percentage is more a measure of pure shooting ability, giving extra credit for three-pointers made. Jennings comes up short of nearly every other rookie on the list. He's got his work cut out for him. Let's move on to some other categories.
Player USG% AST% TO% ORB%
Tyreke Evans 25.9 25.5 13.4 2.6
Stephen Curry 20.9 23.4 17.0 1.7
Brandon Jennings 26.1 30.3 13.0 1.9
Reggie Williams 20.8 14.1 8.8 3.7
Marcus Thornton 25.5 9.8 7.4 4.5
Jonny Flynn 24.3 24.8 17.6 1.2
Darren Collison 23.2 33.4 19.4 2.2
Omri Casspi 19.3 7.8 11.7 4.4
James Harden 20.8 13.6 13.2 3.4
Jonas Jerebko 16.2 4.0 11.2 10.3
Wesley Matthews 16.6 8.4 11.5 3.2
Ty Lawson 18.1 23.9 16.1 4.0
Taj Gibson 16.9 5.3 15.0 11.4
DeMar DeRozan 18.3 5.0 9.9 5.2
Tyler Hansbrough 25.4 9.1 7.1 12.7
Here's a look at some non-shooting offensive categories. The reasons that Jennings is so highly-valued in my system, NBAPET, are starting to emerge. He's got the highest usage rate of all rookies. A high usage rate can obviously be a double-edged sword. Using and squandering possessions doesn't help a team, but someone's got to use possessions, or an offense will grow stagnant. Jennings has proven that he can get offense when he wants, even a little more so than Evans, which is a surprise. His shooting numbers, which are admittedly poor, are mitigated by how well he passes and takes care of the ball. The gap is closing, but it's still undeniable.
Player G MP AST PTS P+A +/-
Tyreke Evans 64 2389 359 1296 2014 -3.7
Stephen Curry 69 2470 388 1138 1914 +5.2
Brandon Jennings 70 2307 417 1105 1939 +0.9
Reggie Williams 13 328 29 193 251 --
Marcus Thornton 64 1547 88 886 1062 -0.1
Jonny Flynn 72 2082 319 968 1606 -8.2
Darren Collison 67 1799 375 779 1529 -3.2
Omri Casspi 69 1778 88 732 908 +2.0
James Harden 64 1472 127 648 902 +8.5
Jonas Jerebko 69 1877 45 639 729 +1.5
Wesley Matthews 72 1695 96 652 844 +0.5
Ty Lawson 56 1194 183 498 864 -6.9
Taj Gibson 70 1814 61 606 728 -1.8
DeMar DeRozan 65 1418 46 550 642 -5.4
Tyler Hansbrough 29 511 28 246 302 --
Efficiency stats and tempo-free metrics are golden for helping us to evaluate players on an apples-to-apples basis. They aid us in assessing concepts like roster-building, fair market value and future production. However, when you're looking backwards, those stats have to be balanced with raw production. A 70% True Shooting Percentage mark is great, but if it comes with 14 percent usage for a player that played a quarter of the available minutes at his position, that efficiency is mitigated. Staying healthy enough to play and getting your coach to have enough confidence in you so that you see the court are things that are just part of the gig. If nothing else, Evans, Curry and Jennings stand out because they've played so much.
In addition, because those players have played so much, they've accumulated some impressive counting stats. The P+A column denotes points plus (assists*2). This crude method of measuring points created is far from the best way to split up the responsibility for a team's points, but it does give us a broad outline of productivity. Here we see that given Jennings' ability to stay healthy and to play well enough to get consistent minutes, as well as his ability to set up teammates and get his own shots, means that the raw accumulated production between Evans, Curry and Jennings is very close. Jennings loses on the efficiency front (again, not as much as you might think given his ability to take care of the ball as a full-time point guard), but he's created more points for his team than Curry and not many fewer than Evans.
The plus-minus numbers you see here are the two-year, adjusted figures from BasketballValue.com. I present them just to acknowledge the on-court/off-court metrics of our group of rookies. I'm not big on plus-minus even in this advanced form other than for the evaluation of lineups and other player groupings. However, I know these numbers mean a great deal to many students of the game, so I present them here without comment and allow you to draw your own conclusions.
We've left out a big chunk of the equation: Defense. As always, these numbers have to be presented with the caveat that measuring defense is a much more inexact science than measuring offense.
Player DRB% STL% BLK% +/- dMULT TEAM
Tyreke Evans 13.3 2.0 0.8 +0.5 .869 20
Stephen Curry 11.8 2.5 0.5 -1.3 1.025 29
Brandon Jennings 10.6 2.0 0.6 +1.3 .888 6
Reggie Williams 12.9 1.0 0.2 -9.8 1.010 29
Marcus Thornton 9.0 1.6 0.5 -0.4 1.035 21
Jonny Flynn 8.2 1.9 0.1 -8.2 1.089 28
Darren Collison 8.8 2.0 0.2 -3.2 1.008 21
Omri Casspi 16.2 1.5 0.7 +3.8 .978 20
James Harden 12.4 2.4 0.9 +2.3 .986 5
Jonas Jerebko 16.0 1.7 1.3 -2.1 .952 26
Wesley Matthews 7.6 1.5 0.6 -1.8 .955 10
Ty Lawson 7.0 1.8 0.1 -0.9 .984 17
Taj Gibson 19.1 1.2 3.5 +0.1 1.044 11
DeMar DeRozan 10.6 1.4 0.9 -3.4 1.136 30
Tyler Hansbrough 16.7 1.6 1.1 +1.3 .993 14
The winners here are clearly Evans and Jennings and it's primarily because of defense that I think Curry should be slotted as third in the rookie race. The +/- figures here are again from BasketballValue.com and reference the one-year, unadjusted on-court/off-court difference in defensive rating of each player. I've taken each figure times minus 1, as BV expresses it the opposite way that I prefer. A positive rating is good on this chart. dMULT is my metric, found on our player cards, that measures the production of a player's box score counterparts against their season norms, with 1.000 being average. Evans and Curry come out very close in dMULT, with both figures qualifying as outstanding. However, Jennings' Bucks are 1.3 points better in Defensive Rating when he plays, a 0.8 point advantage over Evans' showing with the Kings. Also, it's no small detail that Jennings' team ranks sixth in Defensive Rating, while the other top-scoring rookies all play on teams that rank 20th or worse. As a player that's logged over 2,300 minutes already this season, Jennings deserves a good deal of credit for Milwaukee's defensive success.
So we've covered a lot of ground and looked at the performances of these players from a number of different perspectives. At this point, it wouldn't surprise me if you haven't budged on your Evans-Curry-Jennings ballot, though I'd hope that the defense chart would at least convince you to drop Curry a notch. I have to admit, at this point, even I'd vote for Evans. Also, this would be a good time to remind you that I picked Evans to win the rookie award before the season. If I wanted to construct an argument in favor of Evans in order to pat myself on the back, it'd be easy to do. However, that's not what I believe. And we're not quite done with our comparisons.
Player PER WS WARP WP82
Tyreke Evans 18.3 5.1 6.4 9.8
Stephen Curry 15.5 3.6 4.2 4.6
Brandon Jennings 14.8 3.7 4.6 12.1
Reggie Williams 21.0 1.2 0.9 1.2
Marcus Thornton 17.7 3.4 2.3 4.1
Jonny Flynn 13.0 0.0 0.0 2.5
Darren Collison 15.7 2.0 3.1 4.7
Omri Casspi 13.3 2.5 1.3 3.0
James Harden 14.4 4.0 2.1 2.4
Jonas Jerebko 14.0 3.8 2.0 1.6
Wesley Matthews 12.5 4.0 -0.2 2.6
Ty Lawson 16.6 3.4 3.0 2.3
Taj Gibson 13.5 3.6 2.2 1.9
DeMar DeRozan 12.0 1.6 -0.6 0.7
Tyler Hansbrough 14.6 0.7 0.8 0.2
PER and Win Shares were taken from Basketball-Reference.com (as were a lot of the numbers in these tables.) WARP and WP82 came from our player cards at Basketball Prospectus. WARP is Kevin Pelton's core metric; WP82 is mine.
I'm not going to address the first two columns. PER is a very popular bottom-line metric and I'm not about to try to talk you out of using it. Win Shares is based on the Bill James concept, but I have to admit I've never really studied how the brilliant folks at Basketball-Reference calculate them. WARP is a excellent player-value metric that I use a lot. Kevin's system differs from mine in that it doesn't as aggressively estimate individual defense, but the system does have a strong defensive component, while incorporating the indispensable concept of replacement value.
WP82 aims to be exactly what you'd think it would be. It estimates the wins produced by a player over the course of a full season. (The '82' part of the name signifies that the metric is prorated for 82 games, but at the end of the season, the metric is simply 'Wins Produced.) I make this estimate in a few steps: 1. Estimate a player's raw points created; 2. Adjust this figure based on how efficiently a player used his offensive possessions; 3. Adjust this figure a second time based on how much a player has held his counterparts above or below their expected production; 4. Square the result of each player's adjusted points created figure; 5. Calculated the percentage of each player's squared points created of the team total; 6. Multiply the percentage calculated in Step 5 times team wins. Thus, a player's WP82 total is hard-wired into the team total. Does this give an advantage to players on good teams and hinder the Brook Lopezes of the world? You bet it does.
Those that have spoken out in favor of Brandon Jennings for the Rookie of the Year Award have cited his contribution to helping his team to a breakout season. This is the sort of intangible-based media-speak that makes statheads cringe. However, WP82 is telling with numbers the same story those people are telling with anecdotes.
All through our comparison of the top rookies, we've seen that the top three players everyone is pointing to--Evans, Curry and Jennings--deserve to be the top vote-getters, though Darren Collison might have entered into the discussion if Chris Paul hadn't returned from injury and Marcus Thornton is coming on fast. I think we've seen that once you account for defensive contribution, Jennings and Evans stand above Curry. After that ... it's all about what you value when you mash the numbers together. To me, the fact that Jennings is close to Evans in production, a little behind in efficiency and a little ahead on defense makes this a close race. However, the fact that Jennings plays on a team on pace to win 46 games, while Evans plays on a team on target to win 27 is the tiebreaker. Brandon Jennings is the Rookie of the Year.
You can pick apart this argument and get future Twitter updates twitter.com/@bbdoolittle.
Bradford Doolittle is an author of Basketball Prospectus.
You can contact Bradford by clicking here or click here to see Bradford's other articles.