As some of you know, myself (Chantech) and my professional trash talking buddy Matt Stahlhut (Hundred_percent) do a lot of work related to NBA handicapping and trying to predict outcomes. We've been fairly successful with our predictions and decided to play along with the TrueHoop Stat Geek Smackdown last year and would have unofficially won against a field of top NBA quants. As a result, we've been invited to officially participate this year and you can read about ithere: http://sports.espn.go.com/nba/playoffs/2011/news/story?page=Smackdown-11.
This is what ESPN wrote
"Then there's Matthew Stahlhut. Throughout last year's contest, he emailed me round by round with picks before each series started and before everybody else's picks were published. In a very professional way, he even talked a little trash. And, even though he wasn't really in the contest, he "won" it. He uses sophisticated models that he created (with the help of others) as a consultant for a gambling group."
First and foremost, we're Nuggets fans and wanted to give you guys some insight into what our thought process was and our predictions for the first round. For this column, I provided the commentary for the Western Conference and Matt provided the commentary for the Eastern Conference. Going forward, we'll be doing regular statistics based columns, jointly, under this name: "The Unitary Executives".
Since this match-up is the most interesting to readers of this blog, I'll start with our analysis of the Nuggets / Thunder match-up. If we use data starting at the trade deadline up through the end of the season, we have Denver winning 72.29% of the time - the Nuggets have simply been the best team in the league since the deadline. The percentages break down as follows:
OKC in 4: 1.9%
OKC in 5: 6.4%
OKC in 6: 6.7%
OKC in 7: 12.7%
DEN in 4: 13.9%
DEN in 5: 19.2%
DEN in 6: 25.2%
DEN in 7: 14%
If we use full season numbers, we have Denver winning 55.47% of the time. This isn't that interesting as both teams are obviously very different. The Thunder are obviously much better, as are we.
The question you're probably asking is why are we picking the Thunder in 7 for the Smackdown when our numbers clearly say Nuggets in 6, especially since the data includes the 2 games we played against the Thunder. There are several reasons:
- The Thunder are a tough matchup for us. They nullify two of our important advantages. The first is Nene's ability to create easy buckets in the post and off the pass. The second is our depth and ability to throw wave after wave of starting talent, playing at a fast pace, against any opposition. The Thunder are deep and young - they have the ability to play their starters the types of minutes most playoff teams do without giving up an advantage due to the faster pace at which the Nuggets play.
- Injuries. We can't overstate the importance and impact injuries have had to the Nuggets. Particularly if Ty can't go, AAA can't come back from his hamstring injury (something he hasn't shown he's been able to do in 3 tries), and uncertainties with Gallo and Nene. Ty, Nene, AAA, and Bird are our most important players against the Thunder and we need them at full strength.
- Haralabos mentioned this in the Smackdown, and it's something I've been saying in comments for 2 weeks. The starting lineup Karl played is terrible and can't score against the Thunder. A lot of how we do in this series will depend on Karl's ability to play the right lineups - something he hasn't shown any history of being able to do.
- We've done more work since 3 weeks ago, and have found that Home Court advantage increases in the playoffs - so we're giving the Thunder more than our usual 60% home court advantage.
- It's kind of an emotional hedge against our positions on the Nuggets to win the Western Conference. If we're going to lose that, we might as well win the Smackdown.
The Lakers are once again getting lucky in the matchup department (facing Utah without Okur last year, drawing the Hornets this year, probably getting the Mavericks in round 2). The key to beating the Lakers is twofold - 1. you need a point guard that's fast enough to exploit Derek Fisher's foot speed at this point in his career and 2. you need to rebound defensively. In theory, the Hornets have the point guard and the defensive rebounding to compete, however, that's only half the story. The other half is how a team's offense matches up with the opposing defense and vice-versa. This series is going to boil down to the elite Lakers defense(46.4% eFG%) vs the Hornets average offense (49.9% eFG%) being superior to Hornets' terrible defense (52% eFG%) vs the Lakers' average offense (48.6% eFG%). Additionally, the Lakers have had a propensity to take entire games off this season and have also been running below expectation in terms of shooting performance so we expect a slight improvement on offense.
We have the Lakers winning at 92.1% and winning in 5 being most likely. Here are the numbers:
LAL in 4: 27.1%
LAL in 5: 34.4%
LAL in 6: 17.9%
LAL in 7: 12.7%
The Hornets outcomes are around 3%, so we won't enumerate them.
The key to this series will be Manu Ginobli's
flopping health. Ginobili is one of the biggest difference makers in the league. With full health we expect the Spurs to flop 72.3% of the time prevail in the series 72.3% of the time, with Spurs in five being the most likely result (23.1% of the time). However, with Ginobili out, The series becomes essentially a coinflip (51.1%) and we expect the Spurs to prevail in 7 (21.6% of the time). San Antonio's offense has been one of the best in the league, mostly because Gregg Popovich has designed the offense to stress defensive rotations. Their offense is proficient at using the Pick and Roll to force defensive rotations on Parker and Ginobli drives. Once the defense rotates, one of their shooters is always open, which they almost always position at the corner 3 locations. Memphis is middle of the league in offense and defense but they're similar to the Lakers in that while their offense isn't elite, they create more opportunities because of their proficiency on the offensive boards.
We have the Spurs winning at 72.3% and winning in 5 being most likely. Here are the numbers:
SAS in 4: 11.5%
SAS in 5: 23.4%
SAS in 6: 17.5%
SAS in 7: 19.8%
MEM in 4: 2.5%
MEM in 5: 5.0%
MEM in 6: 11.2%
MEM in 7: 9.0%
These numbers are assuming Ginobli plays. The entry we sent into the Smackdown (Spurs in 7) was assuming Ginobli misses some time or isn't at full strength.
The Mavericks are a good all around team that's good on both offense and defense. While the Mavericks are very good, they also ran above expectation on wins this year. Their biggest weakness is interior defense and rebounding, particularly offensive rebounding. If the Grizzlies win this series, it will be because of their dominance on the offensive boards - they have a rebound rate of around 30% since the trade deadline, one of the best in the league whereas the Mavericks get about 74.6% of defensive rebounds (above league average).
As a side note, the Maverick's rebounding seems worse because they're not good at offensive rebounding. The Maverick's problems on the offensive boards can almost entirely be explained by their offense's reliance on Dirk jumpers. When Dirk shoots a mid-range jumpshot, he's not in position to get the offensive rebound. The average offensive rebounding rate this year for forwards was ~6%. Dirk's rate was around 2%. Thus, Dirk's offense isn't as good for Dallas as it could be if he was a shooting guard taking those jumpers instead of a forward. You can think of it like this: if you have a shooting guard taking those jumpshots some of the time he makes it and 6% of the time a forward should get the rebound, which leads to some additional points. If your 7 footer is shooting jumpshots and not rebounding, he has to shoot way more efficiently to account for this difference in rebounding. This is an important concept because it helps explain why the Nuggets still play offense at such a high level without Melo. Nene possessions lead to good things - he scores a lot of the time, he passes for easy shots, and when he passes and a guy misses, he gets the rebound and scores a lot of the time.
We have the Mavericks winning at 58.33% and winning in 7 being most likely. We also expect the over/under on Mark Cuban camera shots to be 70. Here are the numbers:
DAL in 4: 7%
DAL in 5: 17.1%
DAL in 6: 14.5%
DAL in 7: 19.8%
POR in 4: 4.7%
POR in 5: 8.5%
POR in 6: 16.4%
POR in 7: 12.0%
Celtics in Five
The Knicks opted to retool their roster by making a blockbuster midseason trade: Danilo Gallinari, Timofey Mozgov,Raymond Felton, Wilson Chandler, etc… for Chauncey Billups and spare parts. Anytime you trade four of your five starters for an aging point guard, there will be an adjustment period, especially when you also add a shoot-first-ask-questions-later small forward with a penchant for horrible defense. But no one expected the level of early difficulty the Knicks had (7-12 in the first 19 games after the trade). They were able to turn it around recently, though, with a nice seven-game winning streak against the likes of New Jersey (twice), Toronto, and Cleveland.
The Celtics also made a move at the deadline, one that was much more surprising: they traded their starting center for a poor-shooting, poor-defending 3/4 in Jeff Green. Most analysts have pointed to this trade as the reason for the Celtics’ struggles without mentioning that Perkins did not play much in the first half of the season and that the change in the Celtics post-trade has been offensive performance, not defense. This decrease in offensive performance is something almost wholly attributable to shooting and something that we expect to return to normal Celtics levels for the playoffs (because shooting numbers are highly volatile).
Luckily for the Celtics, they get to play a Knicks defense that would make Paul Westphal blush. The Knicks’ defense is one of the worst help defenses around the rim I’ve ever seen. My non-statistical prediction is that this series will feature an inordinate number of layups and dunks because the Knicks are so bad at help defense (Nuggets fans probably are aware of Carmelo’s deficiencies in this area).
The Celtics defense is good for three reasons: (1) they are able to play excellent help defense while still maintaining an above-average defensive rebound rate (largely because they rotate so well), (2) they create a lot of turnovers (thank you, Rajon Rondo), and (3) they rarely attack the offensive glass so they are hardly ever caught out of position in transition. Because they play help defense so well, they match up very well with shoot-first players who are poor passers (hello, Carmelo and STAT). Don’t be fooled by the number of points put up by the Knicks (they play at a pretty fast pace), their high-octane offense will be stalled by this Celtics defense.
Here are the numbers:
BOS in 4: 18.4%
BOS in 5: 29.9%
BOS in 6: 18.9%
BOS in 7: 17.1%
NYK in 4: 1.1%
NYK in 5: 2.5%
NYK in 6: 6.5%
NYK in 7: 5.7%
Heat in five
The Miami Heat are the best team in the league this year based on regular season numbers.
There. Somebody had to say it.
Now that doesn’t mean they are a lock to win it all (they don’t have home court and there are other very good teams). But they are good. Very good.
Defensively, the Heat are designed contrary to standard basketball wisdom. Their defense is stacked with two of the best wing defenders in the game (LeBron and Wade), and they lack your standard interior defenders. This has been a successful strategy because both players are so fast that they can help down low without leaving their men open. This strategy has only really been exposed once, by San Antonio, of course, with a fast ball-movement offense that takes advantage of over-helping (I’m talking to you, Dwyane Wade).
The Sixers are an interesting story. About 20 games into the season, they were struggling. I checked the numbers and we had them rated as an above .500 team. That seemed crazy, but they went on a run that nearly lead them to the sixth seed in the East. Not bad for a team that started 5-14. They are above average on defense because they create turnovers and hold opponents to below average shooting. Their main positive on offense is that they don’t turn the ball over.
There is nothing all that interesting about this series, so I’ll just present the numbers:
MIA in 4: 29.1%
MIA in 5: 35.0%
MIA in 6: 17.4%
MIA in 7: 11.7%
PHI in 4: 0.4%
PHI in 5: 0.9%
PHI in 6: 2.8%
PHI in 7: 2.7%
The Bulls are amazing. And they are a good example of what good coaching can do to a team. They are a near duplicate, from a strategy standpoint, of the 2007-2008 Celtics on defense. And they perform almost as well despite employing Carlos Boozer, who is no one’s idea of a good defender. Their bench is even better on the defensive end as Asik and Thomas (yes, that Kurt Thomas) have been two of the better defensive big men in the league in limited minutes. The only chink in the Bulls’ defensive armor is Derrick Rose, who is a pretty good isolation defender (mostly because he has great help defense), but tends to get lost in pick-and-roll situations.
Offensively, the Bulls are Rose or bust. Rose uses an incredible amount of possessions (a 32.2% usage rate). He is a great one-on-one player who has morphed into an acceptable shooter (although I think his 3 point percentage is above his true talent level). His one downfall may be that he is not a very accurate passer at times and that is why we’ve seen less of an impact from his teammates as you normally would when playing with a guy who can beat his man consistently.
I never thought I would live in a world where Josh McRoberts is a good NBA player (he is). The problem with the Pacers is that he is one of their best players, and you will have difficulty beating an elite NBA team in a seven game series with a group of solid players, but no really good players. They don’t shoot the ball particularly well and turn it over too much on offense. On defense, they don’t create enough turnovers, but are otherwise very solid. Unfortunately, for them, this is a recipe for a quick exit against a defense like Chicago’s.
CHI in 4: 34.5%
CHI in 5: 35.5%
CHI in 6: 16.2%
CHI in 7: 9.3%
IND in 4: 0.2%
IND in 5: 0.6%
IND in 6: 1.8%
IND in 7: 1.9%
Let me clear something up from the onset: I NEVER said that the Magic were the best team in the East. What I said was that we had them rated higher than most experts had them rated. And we have the Hawks rated worse than most experts.
The Magic are good at defense for two primary reasons: (1) Dwight Howard, and (2) Stan Van Gundy. To be honest, I’m not even sure that’s the correct order. The Magic actually play surprisingly good defense when their big man is on the bench, especially considering their personnel. Both the Magic’s offense and its defense are good because of shooting; they hold their opponents to low efficiency shooting while they shoot very efficiently themselves.
The Hawks are nothing special, they are essentially the same team that the Magic wiped the floor with last season. The only difference is that Joe Johnson has gotten worse and Al Horford has gotten much better and has even turned into one heck of a defender. They plan to play one-on-one defense with Howard (especially with Jason Collins), which is not a bad strategy given that so much of the Magic offense comes from the attention Howard draws. But look for Howard to put up huge numbers with this strategy.
Even though the Magic lost game one, nothing really changes. The Hawks shot very well and, given the Magic’s defensive history, I do not expect this to continue through out the series. The numbers may now predict a seven game series, but I do not forsee a Hawks upset.
Here are the numbers:
ORL in 4: 28.5%
ORL in 5: 34.8%
ORL in 6: 17.6%
ORL in 7: 17.6%
ATL in 4: 0.4%
ATL in 5: 1.0%
ATL in 6: 2.9%
ATL in 7: 2.8%