clock menu more-arrow no yes

Filed under:

They’ve Got Numbers Part 1: Tony Allen

Why do advanced stats show Tony Allen is an elite offensive rebounder?

NBA: Memphis Grizzlies at Denver Nuggets Isaiah J. Downing-USA TODAY Sports

The NBA has become a trailblazer, along with the MLB, in getting fans to care about and be aware of advanced statistics. These aren’t the most commonplace - people aren’t getting in bar fights over a player’s PER - but points, rebounds, and assists per game just aren’t good enough on their own anymore.

A good example is Andrew Wiggins. He puts up pretty good traditional numbers; he averages 23.0 PPG as of this writing on .452/.346/.759 (FG/3P/FT) shooting splits. 22 year olds don’t average 23 a night easily.

But according to some advanced stats, Wiggins might not be as good as those prior numbers suggest. Andrew Wiggins is 185th in ESPN’s Real Plus-Minus (RPM), a stat that takes the simple plus-minus stat and adjusts for caliber of teammates, and 429th(!) in Defensive RPM, absolutely horrendous numbers. PER is a more standardized version of a player’s overall statistical contribution. PER was created (by Grizzlies VP of Basketball Operations, John Hollinger, no less) so that 15 would be the league average every year, 20-25 would be All-Star level, and 25-30 would be MVP-level of play. Andrew Wiggins’ career high is 16.5, and this year that number dropped to 16.2.

Do those advanced numbers spell doom for Andrew Wiggins? Do they mean that his per-game numbers are actually inflated and worthless? Or do they mean the advanced stats are worthless instead, not truly encapsulating how good Andrew Wiggins really is?

No to both.

There’s no one statistic that can compare players or teams to one another with 100% confidence. You can use traditional or advanced stats, or you can adjust numbers per-100 possessions or per-36 minutes, and there still wouldn’t be a single number that works all the time. No single stat is going to prove how good or bad you think Wiggins really is as an NBA player.

Why? Because numbers for numbers’ sake mean nothing. If you’re comparing players number-for-number, you’re eventually going to contradict yourself because different players have their contributions show up in different ways.

For example, Michael Jordan isn’t the greatest of all time because he averaged more points-per-game than everyone else, but it definitely helps. However, if you started ranking all-time players based on just points-per-game, then Bill Russell wouldn’t even crack the top 200 players ever, which is blasphemy.

Then, we could rank all-time players based on PER. That way, we use a more well-rounded statistic that is scaled for all production. Do you want to do that? Cause then we end up with Greg Monroe as the 60th best player in basketball history. Yes, that Greg Monroe. That’s not a world I agree with, nor is it one I want be a part of.

That’s not the point of advanced statistics. PER is the most standardized and useful numerical representation of an NBA player’s production, and it STILL isn’t enough to accurately represent that player’s skill, talent, and value.

This especially becomes true on the defensive side. Blocks and steals are the only defensive stats a player can accumulate, which PER includes, but PER can’t calculate how well a player dodges screens or alters shots at the rim. PER can’t calculate denying your defensive match-up the ball every time down the court, or making them think twice about shooting when they do get the ball, a la Kawhi Leonard.

The point is that statistics aren’t the end of the discussion of who is better, and that’s not how they should be used. All stats, advanced or traditional, are there to provide more information, not to make the decision for you. Statistics are a part of the discussion, but that discussion without context and the eye test is a non-starter.

And this is not me trying to disparage stats or analytics. Not only are they personally interesting, but it’s clear that they help tell a more complete story. It’s also clear that the teams that invest in analytics reap the benefits, because having more information will never be a bad thing.

But every stat has a story.

Marc Gasol is averaging a career-low in both rebounds-per-game and total rebounding percentage this season. Without context, it just looks like regression for a 32-year old center who was never a great rebounder anyways. But with a huge uptick in Gasol’s 3-point attempts, he now plays more on the perimeter instead of banging down low, which provides him with far fewer opportunities to rebound.

Sometimes statistical anomalies can explained as easily as that. Because the NBA is weird, some statistics will be a little harder to find explanations for, just like Wiggins. Is he really only the 429th best defender in the league? No, because RPM just measures how your team does with you versus without you, not how well you force your opponent to pass out of a shot or make a decision. Stats need context.

But that’s exactly what I intend to do with this piece. I want to take statistics and bring them to life with real in-game examples so that the numbers aren’t just digits in a spreadsheet. Encouraging stats, discouraging stats, anomalies, and trends all have very real basketball causes and impacts, and the Grizzlies feel them like every other team.

Let’s dissect some particularly noteworthy or eye-popping numbers in Memphis and determine how exactly they’re coming to be.

Tony Allen’s Offensive Rebounding

Tony Allen has the best Offensive Rebounding Percentage (ORB%) in the NBA among guards at 9.4%.

He’s also leading the league among guards in ORB-per-game at 2.3 a night. But ORB% calculates the percentage of available offensive rebounds that a player recovers. It takes into account the amount of defensive rebounds the other team gets because there are only so many rebounds to recover.

But TA isn’t just good for a guard; he’s one of the best on the Grizzlies. He’s tied for 2nd on the team in offensive rebounds per game and 3rd in ORB%, behind post players Brandan Wright and Zach Randolph.

How does Tony Allen get so many offensive rebounds, and how is he always in position? Well, you can thank his bad shooting. Since he’s not an effective 3-point shooter, a lot of the offense he does create is based on cuts to the basket.

Tony Allen plays near the basket quite often. Most of the time he’s down there while Marc Gasol plays on the perimeter in a positional role reversal.

Here TA sets a screen for Mike Conley and rolls to the basket. He doesn’t get the ball, as Conley passes to Gasol to attempt a 3. Tony Allen has good position near the basket, but when Eric Bledsoe gets in front of him, he tugs him away and gets the offensive rebound. He even tries to put it back up and draws free throws.

Here’s another good example. Since TA isn’t a great shooter, Juan Hernangomez is fine giving him space on the 3-point line. But Hernangomez fades away and gives Allen an open lane to cut to the basket. Again, TA doesn’t get the pass, but he has great position to clean up the miss by Conley.

A lot of rebounding is positioning, and Tony Allen is really good at getting that position as shown in the videos above. Here again, Tony Allen has good position way under the basket and he’s able to keep Klay Thompson at bay to get the offensive rebound.

Tony Allen shows you don’t have to be a big man to be a good rebounder. Size and verticality help, but sometimes opportune position and active hands are all you need.

That’s how Tony Allen has been one of the best offensive rebounders for the Grizzlies. Part 2 will take the same process and analyze how Mike Conley plays with and without Chandler Parsons.

Statistics courtesy of basketball-reference.com, nbawowy.com, and nba.com/stats

Follow @sbnGrizzlies