Statistical Oddities of Questionable Significance
Posted by Zach Lowe on Dec 8, 2009
Here are a few interesting statistical trends that I don’t feel merit a separate post for whatever reason. Still, they’re interesting, and a few of them might actually be significant going forward. I thought I’d dump them all here and you guys can tell me which are important, which are fluky and why you think they’re happening.
Without further ado, the Statistical Oddities of Questionable Significance (SOOQS):
• The C’s have the best shooting percentage on shots at the rim in the NBA, according to Hoopdata. The team is hitting 67.8 percent of shots in that range, well ahead of the #2 team (Atlanta, at 65.1 percent). Phoenix led the league last season, when they hit 66.4 percent at the rim.
• Nearly 65 percent of Boston’s baskets come off of assists, which places the C’s neck-and-neck with Utah for the highest assisted-on rate in the league. And those two teams are well ahead of the other 28 teams; the Clippers (!) come in third, with an assisted-on rate of 60.1 percent. Interestingly, Orlando (54.6 percent assisted-on) is the only title contender that ranks outside the top 10 in this category.
So this is less a sign of the declining individual ability of the C’s older players (particularly Ray Allen and KG), I think, and more a sign of the emergence of Rajon Rondo and the general good state of the team’s offense. The C’s finished 3rd in this category last season, at 60.5 percent. (Utah led the NBA by a long shot. The champion Lakers were 8th).
• Marquis Daniels has not taken a single shot in what 82games.com defines as “the clutch”—the 4th quarter or OT of a game in which the score is within five points. The Grand Marquis has played just 1 percent of the team’s “clutch” minutes. I have no idea why I find this interesting.
• Rasheed Wallace has taken some shots in clutch time, and every single one of them has been a jump shot, according to 82games.
• The C’s are forcing opponents to turn the ball over on 16.3 percent of opponent possessions, the highest rate in the league, according to Basketball Reference. The last team to have a higher turnover rate over the course of a season: The 2006 Charlotte Bobcats (16.5 percent).
• Here’s one that remains a mystery to me: The C’s have two primary “second units,” and one of them has consistently outperformed the other by a huge margin over last season and the first 20 games of this one. For the life of me, I can’t figure out why.
The “bench units” that get the most minutes typically feature the same four back-ups playing with one starter—either Paul Pierce or Ray Allen. Last year, the Allen unit (Ray plus Big Baby-Powe-House-Tony Allen) played 101 minutes (5th-most of all five-man units the team used), scored 113 points per 100 possessions and allowed about 101 points per 100 possessions, according to 82games.
That is outstanding. To put that in perspective, those numbers, over the course of a season, would come close to leading the NBA on both sides of the floor.
The Pierce unit (Pierce plus the same four back-ups) scored 104 points per 100 possessions and allowed about 100 points per 100 possessions in 145 total minutes. Another solid defensive unit, but the offense sputtered.
This year? The spread is even larger, according to 82games. The Ray Plus Bench unit is killing the league, averaging 127 points per 100 possessions while allowing just 93. Those numbers are so good as to be unsustainable.
The Pierce unit? Not so good. It’s scoring just 101 points per 100 possessions and allowing 116 points per 100 possessions. Again, put that in perspective: Those numbers would be close to the bottom of the league on both offense and defense.
So what’s going on? I still haven’t figured it out, so I’m open to any theory. Does Ray’s better three-point shooting open up the floor more for the other players? Does the Ray-House-Sheed trifecta of spot-up experts (or so-called experts, anyway) present more problems than a line-up with Pierce on the floor? Does Marquis Daniels have a clearer role when Pierce isn’t in the game taking over a lot of the ball-handling duties?
If you’re counting, that’s three questions and we haven’t even addressed the defensive numbers.
But one thing is clear: The second unit guys have played better with Ray over the last 1 1/4 seasons.
• The Celtics are getting many fewer “And-1s” this season, despite the fact that Sheed pleads for one just about every time he shoots the ball. Hoopdata tracks a stat called “And-1 percentage,” which is the ratio of And-1s to total field goal attempts. So if your team earns, say, two And-1s for every 100 field-goal attempts, its And-1 percentage would be (a very bad) 2.0%.
This year, the C’s And-1 percentage is 2.49 percent, placing them 25th in the NBA. (League average And-1 percentage is 2.91).
Last season, the C’s ranked 3rd in the NBA with an And-1 percentage of 3.33 percent.
The big reason for the drop? Brace yourselves: The absence of Tony Allen, who had the second-highest And-1 rate on the team in ’09, behind only Paul Pierce. Leon Powe ranked #5 (I expected him to rank higher), so losing him hurt the C’s here as well.
We’ve written a lot about how the C’s are going to the foul line a lot less often this season. And we all agree that Tony Allen is the most frustrating player on the team. But this is one area where he can help—diversifying the way the 2nd unit can score.
But can he do that without hurting the team in other ways?
That’s the question that will decide whether Tony Allen is a) playing in the post-season; b) riding the bench in the post-season; c) dealt at the trade deadline.
The Ticker