The Sacramento Kings are one of the worst teams in the NBA. That would have been the consensus of most NBA observers before the 2009-10 season began. In fact, I asked in October: Will the Kings be Crowned the Worst Team in 2009-10?
Reviewing the Kings potential depth chart before the season began, it’s easy to see why someone (okay, me) would conclude this team would be so bad:
Potential First String
PG: Tyreke Evans [rookie]
SG: Kevin Martin [4.7 Wins Produced, 0.115 WP48]
SF: Andres Nocioni [-0.4 Wins Produced, -0.009 WP48]
PF: Jason Thompson [3.0 Wins Produced, 0.063 WP48]
C: Spencer Hawes [-1.0 Wins Produced, -0.021 WP48]
Potential Second String
PG: Beno Udrih [1.6 Wins Produced, 0.034 WP48]
SG: Desmond Mason [-0.3 Wins Produced, -0.014 WP48]
SF: Francisco Garcia [1.8 Wins Produced, 0.044 WP48]
PF: Sean May [-0.6 Wins Produced, -0.093 WP48]
C: John Brockman [rookie]
As noted in the post detailing this depth chart, the returning veterans produced less than nine wins last year. And only one player –Kevin Martin – was above average last year. These numbers suggest the Kings were going to struggle in 2009-10. After nine games, though, the Kings have already won five contests. Yes, the Kings have a winning record. And although it’s early, it doesn’t look like Sacramento will be the worst team in the NBA in 2009-10. So how did this happen?
The obvious first place to look is the play of Kevin Martin. Thus far Martin is averaging 30.6 points per game and his WP48 [Wins Produced per 48 minutes] of 0.261 is clearly an improvement over what we saw last year (and closer to what we saw before 2008-09). Martin, though, has only played five games and the Kings lost four of these contests. So maybe it ain’t all about Martin.
Well, what about Tyreke Evans, the Kings lottery pick in 2009 (the reward for being so bad last year)? Thus far Evans is second on the team in scoring, averaging 16.8 points per game. Evans, though, is not a very efficient scorer so his WP48 is only 0.058 (average is 0.100).
What about Andres Nocioni, Spencer Hawes, Beno Udrih, Desmond Mason, or Sean May? These players were all below average last year. And this year — as the following list illustrates – all are still below average. In fact, Mason played so poorly that he went from the starting line-up to the unemployment line.
Spencer Hawes [266 minutes, -0.063 WP48]
Desmond Mason [66 minutes, 0.023 WP48]
Sean May [82 minutes, -0.184 WP48]
Andres Nocioni [239 minutes, 0.016 WP48]
Beno Udrih [237 minutes, 0.069 WP48]
Okay, let’s summarize. Martin did return to form (this was expected), but this apparently hasn’t helped much yet. Most veterans are playing as badly as they did last year (this was also expected). And Evans is not very productive yet (this was also somewhat expected). So thus far, most players on the Kings are who we thought they were (quoting Dennis Green). But the Kings as a team are much better. So what gives?
There are three players who are part of Sacramento’s rotation who have not been mentioned. Omri Casspi [170 minutes, 0.126 WP48] and John Brockman [60 minutes, 0.133 WP48] are rookies, and each has been above average.
Although both Casspi and Brockman have helped, neither player is doing enough to transform the Kings. Last year the Kings had an efficiency differential (offensive efficiency minus defensive efficiency) of -8.9. This year the Kings have a -0.5 differential. This difference in these differentials is worth about 22 wins across an 82 game season. To get all these wins, a major producer of wins had to be added to the roster. But none of the players we have reviewed so far is doing enough to produce such a large leap.
Fans of the Kings know there is one name that has been ignored so far. And fans of the Kings probably know who is responsible for Sacramento’s improvement. As a rookie, Jason Thompson was below average [3.0 Wins Produced, 0.063 WP48]. After 310 minutes this year, though, Thompson’s WP48 stands at 0.257. If Thompson continues to produce at this level – and he continues to play 34 minutes per game – when the season ends Thompson will produce about 15 wins. Last year the Kings only won 17 games. So Thompson’s dramatic improvement is enough to significantly change the fortunes of this team.
This story highlights a problem anyone has forecasting the NBA. Relative to what we see in baseball and football, basketball players are far more consistent. Nevertheless, performance can change. Injuries – as we see with Kevin Martin – can cause performance to change dramatically. And age will cause players to get better early in their career (and cause declines later on). In general, the shape of the age profile is a gentle slope (we discuss this in our next book). But in Thompson’s case, we are seeing a dramatic leap; a leap that I don’t think we could have expected given his performance last year.
Hence we see the problem of forecasting. On average, we see gradual changes as people age. But sharp changes are possible (both upwards for young players and downward for the aged). Injuries – and recovery from injuries — are hard to forecast. And finally, the performance of rookies is hard to predict. Given all this, forecasts before the season are difficult. In sum, we can do a pretty good job explaining what happened after the fact. But the future – even with all the great numbers the NBA producers – will still be cloudy.
Of course, this is just a problem for people like me who like to play with numbers. For fans of the Kings, the important story is what has happened to this team. Assuming Jason Thompson is for real (it is early, so maybe he isn’t), the Kings now have two players (Martin & Thompson) who can be significant producer of wins. If Brockman and Casspi get minutes and keep producing, Sacramento is suddenly a team this isn’t horrible. If this story starts to be understood by fans in Sacramento, maybe attendance for this team can rebound. And then, just maybe, Sacramento – primarily because of the play (and rebounding) of Jason Thompson — can keep its NBA team.
The WoW Journal Comments Policy
Our research on the NBA was summarized HERE.
Wins Produced, Win Score, and PAWSmin are also discussed in the following posts:
Finally, A Guide to Evaluating Models contains useful hints on how to interpret and evaluate statistical models.