Posts Tagged ‘Advanced Statistics’

I’m a big fan of On/Off data, which compares a team’s point differential with a player on the court versus when he’s off the court. I’ve referenced it frequently in the past and think it’s one of the more telling reflections of a player’s value to his team in the advanced stat family.

The nice part about On/Off is that it represents what actually happened. The problem with On/Off is it ignores the reasons why it happened. And sometimes, it creates a fuzzy picture because of it.

For example, let’s suppose Kobe Bryant plays the first 40 minutes of the a game and injures his ankle with the score tied at 80. LA wins the game 98-90. The Lakers were dead even when he was in the game, and +8 with him out of the game – Bryant’s on/off would be -8.

In this case, sample size is an issue. But that becomes less of a problem over the course of an entire season. The real concern is the normal variance involved in everyone else’s game. Practically speaking, it takes little outside the norm for Kobe to have played 40 brilliant minutes while his teammates missed a few open shots, and for the opponent to miss a few open shots down the stretch while Kobe’s teammates start hitting them.

The tendency is to look at a result like that and conclude that Kobe hurt his teammates’ shooting and when he left the game it helped their shooting. He very well may have by not creating good looks for them.

Then again, players hit unguarded 3-pointers about 38% of the time. Which means if the average shooter attempts five open 3-pointers, he will miss all five about 10% of the time, simply based on the probabilistic nature of shooting. A fact that has little to do with Kobe or any of the other players on the court.

In our hypothetical situation, all it takes is an 0-5 stretch from the opponent and a 3-5 stretch from LA to produce Kobe’s ugly -8 differential. The great college basketball statistician Ken Pomeroy ran some illuminating experiments on the natural variance in such numbers. His treatise is worth the read, but the gist of it is that his average player — by definition — produced a -57 on/off after 28 games (-5.7 per game) due to standard variance in a basketball game outside of that player’s control. Think about that.

For fun, I just ran the same simulation and my average player posted a +5.6 rating of his college season:

Average Player Simulation

So in two simulations, the average player’s On/Off ranged from -5.7 to +5.6. One guy looks like an All-Star, the other like an NBDL player.

“The Team Fell Apart When Player X Was Injured”

This is a common argument for MVP candidates: Look at how the team fared when he missed a few games and conclude the difference is the actual value a player provides to his team. Only this line of thinking runs into the same problems we saw above with on/off data.

Let’s take Dirk Nowitzki and this year’s Dallas Mavericks. In 62 games with Dirk, Dallas has a +4.9 differential (7.8 standard deviation). In nine games without Dirk, a -5.9 differential (7.5 standard deviation).

Which means, with a basic calculation, we can say with 95% confidence that without Nowitzki, Dallas is somewhere between a -1.0 and -10.8 differential team. Not exactly definitive, but in all likelihood they are much worse without Dirk. OK…but we can’t definitively say how much worse they are.

In a small sample, we just can’t be extremely conclusive. In this case, nine games doesn’t tell us a whole lot. New Orleans started the season 8-0…they aren’t an 82-win team.

We can perform the same thought-experiment with Dirk’s nine games that we did with Kobe’s eight minutes to display how unstable these results are. Let’s say Dallas makes three more open 3’s against Cleveland and the Cavs miss three open 3’s. What would happen to the differential numbers?

  1. That alone would lower the point differential two points per game.
  2. Our 95% confidence interval now becomes -12.1 points to +4.4 points.

That’s from adjusting just six open shots in a nine game sample.

Jason Terry — a player who benefits from playing with Dirk Nowitzki historically — had games of 3-16, 3-15 and 3-14 shooting without Dirk. He shot 39% from the floor in the nine games. By all possible accounts, Terry is better than a 39% shooter without Nowitzki. He shot 26% from 3 in those games. Let’s use his Atlanta averages instead, from when he was younger and probably not as skilled as a shooter: How would that change the way Dallas looks sans-Dirk?

Well, suddenly Terry alone provides an extra 1.7 points per game with his (still) subpar shooting. The team differential is down to -2.2 with a 95% confidence interval of -10.4 to +6.1. Just by gingerly tweaking a variable or two, the picture grows hazier and hazier.

Making Sense of it All

So, what can we say using On/Off data? It’s likely Dallas is a good deal better with Dirk Nowitzki. But, hopefully, we knew that already.

To definitely point to a small sample and say, “well this is how Dallas actually played without Dirk, so that’s his value for this year” ignores normally fluctuating variables — like Jason Terry or an open Cleveland shooter — that have little to do with Dirk Nowitzki’s value. So while such data reinforces how valuable Dirk is, we can’t say that’s how valuable he is.

We can’t ignore randomness and basic variance as part of the story.

Read Full Post »

There are a number of great debates in NBA history. West-Oscar. Magic-Bird. Garnett-Duncan. Kobe-Wade is another. In the spirit of this year’s ridiculously hyped matchup between Miami and Los Angeles, here’s my take on the two best shooting guards since MJ…

First, Kobe does and should always win a career comparison though, as he crushes Wade in longevity. Let’s get that out of the way. After that, it’s a lot more interesting. Who was better at their peak? Who was better during prime years?

The statistics*

*All statistics in this post are per 75 possessions played

Kobe is averaging 26.9 pts 5.6 reb 5.0 ast 1.6 steals 0.5 blocks and 3.1 TOV on +2.8 TS% (relative to league average) for his career.

Wade is averaging 26.9 pts 5.3 reb 6.9 ast 1.9 steals 1.1 blocks and 3.9 TOV on +2.9 TS% (relative to league average) for his career.

Those numbers are eerily similar; Their scoring rates differ by 0.002% for their career. Wade edges Bryant in Win Shares/48 .190 to .188. Bryant edges Wade in career ORtg 112 to 111. So the career averages are telling us it’s really close, and I agree.

But what about their prime years? For the sake of this comparison, let’s do this 2 ways. First, the overall prime of each player (01-10 for Bryant, 05-10 for Wade):

Again, these numbers are quite similar. Kobe played 31% of his games alongside Shaq (01-03), 9% of his games in the 2004 quartet of future Hall of Famers, 37% of his games as the lone assassin (05-08) and another 24% of his games with Pau Gasol.

Wade played 42% of his games with Shaq (05-07) and 58% of his games as the lone assassin (07-10). More shots could be expected without the presence of an all-star big, but in the case of Kobe (and to a lesser degree Wade) it didn’t matter too much. Both players did see an increase in TS% playing alongside O’Neal.

Unfortunately, this comparison includes play that we don’t really want to compare. Kobe was injured in 2005. Wade in 2008. Whether they averaged 20 pts/75 or 4 shouldn’t change the overall quality of these players when healthy, and that’s what we’re interested in. Instead, let’s compare them since 2005 — with the new perimeter rules in place for both players — and remove the injury periods for each player (2005 and calendar 2010 for Kobe, 2008 and post All-Star game 2007 for Wade):

So finally we see some discrepancies. Note that Wade played with Shaq in 2006 but was essentially on his own from then on. When healthy in 2007, he put up a preposterous normalized line of

30.5/5.1/8.3 with 2.2 steals and 1.2 blocks (4.4 TOV) on +5.1 TS%

That’s one of the best lines I’ve ever seen, and given what he did in the 2006 postseason, I’m not sure Dwyane Wade wasn’t the best player in the world for about 12 months between the Spring of 2006 and his shoulder dislocation in 2007.

Bryant’s overall numbers are quite impressive. His 2010 run to start the year was some of his best basketball, before injuries ran him down:

30.6/5.8/4.5 with 2.1 steals and 0.3 blocks (3.1 TOV) on +2.5 TS%

He turns it over less than Wade, but most of that is because of role, as Wade handles the ball more and Kobe simply shoots the ball more. This is also part of the reason Wade’s assists numbers trump Bryant’s; he’s acting as the primary creator with the ball more. And he drives to the basket more – more on this in a moment. Then again, I happen to think Kobe protects the ball better than Wade and Flash creates more than Bryant, so the stats are telling us something that jibes with observation.

Playoff Statistics

Next, there’s the playoff performances. Here’s where the comparison grows extremely interesting for me. Career playoff numbers*:

*Relative TS% here is calculated by taking actual TS% and subtracting it from the weighted average of opponent’s TS% against in the regular season.

Eek. Here we go again. Note Wade’s large increase in relative TS% in the playoffs. He’s now outscoring Kobe and outshooting him. But again, this includes data from pre-prime years. Let’s use the 2005-2010 range again, removing Wade’s injured playoff games in 07:

Here’s where the statistics begin to reflect the differences I see in the players. Kobe and Wade are extremely similar in many regards. But Wade’s increased performance in the playoffs thus far has been fairly astounding. An output of 28 pts/75 on +5.8% relative TS% is really just staggeringly good. Meanwhile, Bryant’s scoring declines while his relative efficiency stays the same. Big edge to Wade here, and we’ve seen displays from him that have generated this edge: 2006 v Detroit, v Dallas, 2010 v Boston, to name a few.

Advanced statistics

Finally, there are the advanced statistics. I’m not a PER or Wins Produced guy. So let’s quickly look at Wins Shares, on/off and raw +/- numbers from when they are available.

EDIT: I’ve found the Adjusted Plus-Minus average from Joe Ilardi, 2003-2009:

  • Wade +7.99 (+0.99 on defense)
  • Bryant +7.07 (-0.55 on defense)

Top 5 seasons by Win Shares/48:

  1. Wade 06 0.239
  2. Wade 09 0.232
  3. Kobe 06 0.224
  4. Wade 10 0.224
  5. Wade 07 0.219

82games.com has tracked on/off since 2003. Kobe’s 03-10 on/off is +8.97. Wade’s (04-10) is +8.51. Here are the top 5 seasons by on/off:

  1. Wade 06 15.5
  2. Wade 10 14.1
  3. Wade 09 14.0
  4. Kobe 06 12.4
  5. Kobe 10 12.3

In the playoffs, a lot is determined by team success. We’re also dealing with small samples. Nonetheless, here are the top PS performances by these two based on on/off, with offensive net difference in parentheses:

  1. Wade 06 +22.0 (+22.0)
  2. Kobe 03 +16.5 (+4.5)
  3. Wade 05 +16.8 (+14.5)
  4. Wade 09 +16.8 (+10.9)
  5. Kobe 09 +11.9 (+14.5)

Interestingly, both of their +/- numbers decline in the postseason. I’m not wild about including the small postseason sample sizes, but just for posterity, the raw +/- from the playoffs, 04-10:

  • Kobe +3.3
  • Wade +3.8

Beyond the Numbers

Now that I’m dizzy from that potpourri of numbers, let’s discuss some actual basketball. The most germane element here, from what I gather, is that people don’t understand that Kobe takes harder shots than Dwyane Wade. And that matters. It’s also one of the very reasons Kobe’s so revered.

But I want a player who can generate easier shots — that’s Jordan’s biggest advantage over Bryant despite the stylistic similarities in their games. It doesn’t matter if Earl Boykins can make shots from all over the court in all sorts of compromising positions and all that Shaq can do is dunk…if Shaq can make enough of his attempts dunks. Consider the numbers:

For their careers, they’ve each averaged close to 22 “attempts” (FGA’s + 0.44 coefficient for FTA’s) per game. If Bryant takes five incredibly difficult shots per game, by his own choosing, and he makes two, we might ooh and ahh at the two he makes. It doesn’t change that it’s a low-percentage shot.

Now let’s say Wade takes one difficult per game and never makes it. All of their other attempts are converted at exactly the same rate, let’s say slightly higher than Wade’s career average, or 59% TS%. That would give us the following results:

  • Wade: 24.78 ppg on 56.3 TS%
  • Kobe: 24.06 ppg on 54.7 TS%

That’s not an insignificant difference, and that’s exactly what happens with Kobe Bryant. This is essentially the difference between Wade and Bryant as offensive players. Wade gets to the rim more and takes higher-quality shots.

Incidentally, it’s also why someone like LeBron James scores so efficiently; He draws fouls and scores at the rim abundantly. (Last year, 601 attempts at 71.2% at the rim, according to NBA Hot Spots. In 2006, Bryant took 502 attempts at the rim at 58.4%)

Finally, there’s an intangible difference between these two players. Kobe has been more durable. But he’s also been quite disruptive on more than one occasion. Like Wilt Chamberlain, his attempts to “prove” things at times, like not shooting in Sacramento, is less than desirable to me. Feuding with players and coaches is less than desirable. I don’t mind demanding a trade, but I do mind someone throwing his center under the bus and not shooting in the second half of a game 7 (with accounts of Bryant yelling at his own bench during that half and standing around idly on 27 possessions, by my count).

Instead of his performance dropping in many key situations, Wade’s improves. I saw Wade in 2005 and 2006 enter a mode of extreme concentration at critical moments, often turning games around in the 4th quarter and leading his team to wins with nearly flawless shooting. I’ve never really seen Bryant do that. I’ve seen Wade torch great defenses that befuddled the rest of the league (Detroit and Boston) and I’ve seen Bryant grow frustrated against similar defenses.

And defensively, Wade has been better over this period. The perimeter barrage that he and LeBron have unleashed in Miami this year are like a Jordan-Pippen Lite defensive tandem. At his best, when he was younger, Bryant was a better on-ball defender. But Wade’s been one of the better defensive shooting guards in the league while Bryant has coasted or been assigned to inferior offensive players to rest.

But these players are extremely close at their best. In the end, I’m siding with Wade here. I just trust him more.

Read Full Post »