Et tu, KenPom?

Most of you know by now that I have a minor man-crush on Ken Pomeroy, largely because he has helped create an infrastructure and culture of unbiased basketball statistics that has advanced the game’s statistics well beyond where they were only a few years ago. Unfortunately, Ken has wounded me deeply with his latest article at Basketball Prospectus.

Ken’s assertion is quite similar to earlier comments by fellow BP writer John Gasaway. To sum up: if it can’t get opponents to turn the ball over, the Vol defense is very average. You may or may not remember that I ripped Gasaway pretty hard over his statements. That was easy because John is an Illinois homer, and the Illini can kiss my ass and Bruce’s ass, as well as the collective ass of the BruceBall nation. It was also easy because John gave one simple, weak statistic that hardly justified his position.

What I have to do now is going to be hard, no matter whether I choose to defend my team or my man-crush. You’re breaking my heart, Ken. So what’s it going to be? Is Ken right or wrong? I think the truth is somewhere in the middle.

First, I think it’s important to note that every team has a hard time when it is taken out of its comfort zone by the style of play. In Tennessee’s case it happens two ways: 1) the opponent takes care of the ball and Tennessee cannot run, and 2) the opponent drags the tempo down to create a slow, plodding, halfcourt game. Tennessee is not the only team with this sort of problem. Last I checked, everyone except Memphis has lost a game so far this year, and often when a good team loses it has something to do with exposed weaknesses.

It’s inevitable that every team will play games in which they are taken out of their element. What separates the top teams is their ability to deal with it. Pomeroy suggests we don’t deal with it well, citing our 5 performances in which teams had less than 20% TO rate. The Vols are 3-2 in these games, and 20-0 in the rest. Clearly he has found a pattern to point to.

A few comments here. First, it says something to me that there are only 5 such games that the Vols have had to play. This tells me that we are especially good at forcing teams into our comfort zone. Pomeroy doesn’t really give the Vols any credit for that.

Another comment: 20% is an arbitrary choice. I could just as easily choose the national median turnover rate, 21.1%. This adds two more games to the mix: Ohio State and Arkansas, and in both of these games the Vols played exceptional defense (d-efficiency of 95.8 and 91.7, respectively). It also raises the Vols’ record in these games to a more respectable 5-2. I could also choose the Vols’ median TO% as the benchmark. How does a Vol opponent do if it manages to stay under the Vols’ median? The Vols’ median is 24.9%. 12 games lie below this, and the Vols are 10-2 in those games (can you see what a difference the arbitrary line makes?). It also adds three more games that the Vols held their opponents under 100 PP100.

Obviously the Vols do better when they turn people over. Obviously they struggle in exceptional cases where they don’t. That’s the idea behind the risky pressure man defense. This picture is hardly unique. What happens to Duke if they don’t shoot well? I’ll give you a hint: in five games this season Duke’s eFG% was less than 50%. Their record in these games? 3-2. What happens to UNC if Tyler Hansbrough doesn’t score a lot at the free throw line? In 6 games he’s hit 5 or fewer FTs. UNC is 4-2 in those games. Get the picture?

Ken’s best point in all this is that Tennessee’s opponents seemingly have more control over this weakness than they would against most weaknesses. They simply have to take care of the ball. I’m not sure how true this is, despite the stat he gives. Ask West Virginia, South Carolina, Alabama, Florida, Xavier, and UNC-Asheville about that. All average under Pomeroy’s benchmark 20% TO rate, and all failed to do this against the Vols. Only Alabama (23.9) and UNC-A (21.9) managed to stay under 24%. Why didn’t they just take care of the ball?

Now, I will give Ken that Tennessee plays a high risk, high reward type of defense that can sometimes get burned. Against Texas it certainly did. The Longhorns shot the lights out and are an exceptionally strong team. Against Kentucky it did as well, but the Wildcats hit an exceptional number of 2-point jumpshots in that game, and you can ask Ken how likely that is. It’s not coincidental that the Vols were also well below their standard offensive output in those games. That tends to contribute to losses as well.

Besides, Tennessee has performed quite well in general when taken out of its comfort zone. The Vols are 10-2 when the TO rate is below their median, item #1 on the what-bothers-the-Vols list. They have the same record in low-pace games. That’s astonishingly better than the Vols were a year ago– 10-7 in slow games as well as low-turnover games. It takes good coaching to see that kind of improvement.

I would like to see the Vols continue to get better in eFG% defense, which is the flip-side of the coin that Ken seems so concerned about. Tennessee is very average in this area (131st) but this is well ahead of where it was in Bruce’s first 2 years (302nd and 243rd), and the Vols won a lot of games those years, including a near-miss that would have propelled them to the Elite 8.

Bruce plays a defense predicated on forcing turnovers, so of course it’s not going to be firing on all cylinders when the opponent doesn’t cooperate. My contention is that there’s nothing unusual about this; every team has a weakness that, when exploited, can get the team beat. I’m not sure I buy that Tennessee’s is as easy to exploit as Pomeroy contends. I don’t know that the stats agree with his assertion as he believes they do.

I have a lot of respect for KenPom– that won’t change. But Ken, say something nice about my Vols, k? And if you don’t, I’m going to bring it right back at you . . . k? Hugs and kisses, big guy.

Individual geek stats are now up

Ken Pomeroy now has enough data to publish stats for individual players for the 2007-2008 season, so I thought I’d give a quick rundown of how our players are doing.

Good

• Offensive efficiency is high for six UT players. Tyler Smith (130.9, #59 nationally), JaJuan Smith (127.2, #99), Chris Lofton (117.9, #310), and Jordan Howell (115.4, #403) are all ranked among the nation’s top 500 players. Ryan Childress (117. and Brian Williams (127. would be in this group if they had played enough minutes.
• Brian Williams’s eFG is 87.5%. That’d be wonderful if he took more shots (takes only 7.5% of team’s shots when he’s in the game) and to do that he needs more minutes. Bruce is not yet comfortable with his contributions and I’m not either, but I think this bodes well for the future.
• Wayne Chism (#17 and Tyler Smith (#361) are ranked as offensive rebounders. Chism is also ranked in defensive rebounding (#248).
• Crews is rated higher in offensive rebounding than Smith but doesn’t have enough minutes to be ranked. Childress would also be ranked, likely in both rebounding stats, if he had more minutes.
• Four Vols are ranked in assist rate: Tyler Smith (#220), Ramar Smith (#320), Jordan Howell (#333), and Josh Tabb (#468).
• Four more Vols are ranked in turnover rate: JaJuan Smith (#54), Wayne Chism (#112), Chris Lofton (#137), and Tyler Smith (#345).
• One Vol ranks in blocks (Chism, #164).
• Five Vols rank in steal percentage: JaJuan Smith (#34), Tyler Smith (#103), Josh Tabb (#138), Ramar Smith (#323), and Chris Lofton (#390). Cameron Tatum doesn’t have enough minutes but his steal rate is higher than any of these five.
• Four Vols are above 33% on three-point shooting: Tyler Smith (44.4%), Jordan Howell (41.5%), JaJuan Smith 37.3%), and Chris Lofton (36.4%).

If it seems like I’m mentioning Tyler Smith a lot, it’s because he pops up in so many categories as a positive for this team. Pomeroy has 8 stats that are rankable nationwide: O-rating, eFG%, O-rebounding, D-rebounding, assists, TOs, blocks, steals. Tyler Smith is ranked top 500 in 5 of the 8, and if he played more minutes (60% required and he’s played 59.1%) he’d be in the top 100 in eFG% as well. He’s done a little of everything for UT and done it all remarkably well so far.

Bad

• Chris Lofton needs to be better than 117.9 in O-rating. Last year was his lowest rating in his time at UT at 125.5. He should improve to at least that, I’d suspect, as the season goes along. At least we all hope so.
• Ramar Smith is a ball hog to some degree (180th in possessions used) and has been inefficient (98.2). If he’s going to end a lot of possessions he needs to end them with the ball in the basket more often. His TO% is 24.2– that’s lousy and is worse than he was as a freshman (21.9%).
• Tyler is very unselfish, maybe to a fault. He uses less than 20% of UT’s possessions when he’s in the game, and he’s arguably the best player right now. He needs to drive or shoot more unless it hurts his efficiency too much.
• As exciting as Tatum can be, he’s been the least efficient offensive player on the team (94.0). He definitely needs polish. He and Brian Williams turn the ball over about 25% each, which is probably typical this early for freshmen.
• Only three Vols– JaJuan Smith, Chris Lofton, and Jordan Howell– are above 70% on free throws. Tyler is getting closer after a really bad start, and should eclipse 70% soon. Wayne Chism has been horrible at the line (6-17, 35.3%). This was an area that hurt us last year and at this rate, it will again.

It’s clear from watching and from looking at stats that Tennessee has a deep, talented team, but it’s going to have to find better defensive consistency, better post play, and better free throw shooting if it’s going to seriously challenge for a Final Four run. With only 7 games under our belt, it’s hard to know too much, but we have a good idea what to keep an eye on as the season progresses.

Intro to a geeky stat

The regular season is now 20% over (6 games vs. D-I played, 24 to go). I figured that this was as good a time as any to round up the relevant stats and take a peek at how we’re doing, as well as how we stack up against our competition and the the nation.

Adjusted for schedule, the Vols rank 19th in the nation offensively (117.5 PP100) and 23rd defensively (83.6 PP100) . Not bad. Now I will introduce another fun (and geeky) stat: Pythagorean expected win percentage.

Pythagorean expected win percentage (or “Pythag” as I describe it verbally) has its origins in the statistical analysis of baseball teams. Essentially, it was designed to calculate how many games a team should have won, based on how many runs it scored and how many runs it allowed. Obviously, most teams aren’t going to match this expectation exactly– they may win one game 15-2 and then lose the next 8-7. Then they have scored a bunch more runs than they allowed, but are still just 0.500. Get the picture? Anyway, the basic formula is

E(W%) = runs scored^2 / (runs scored^2+runs allowed^2)

For this particular team, E(W%) = 11.5^2 / (11.5^2+4.5^2) = 0.867. You can see that they are perhaps playing better than their record would indicate. If it weren’t just 2 data points (games) we might like to use this information to predict that they will win more games than they will lose, and to date have just been “unlucky” to achieve only a 0.500 record.

A similar formula can be used for basketball, using points scored and points allowed– but we have a problem. The scales between these two sports (total scores per game) are very different, so we can’t expect one formula to work for both. In fact, the formula used for basketball can be modified to more closely match outcomes. The end result is that instead of squaring all terms, a higher degree is used for the exponent– anywhere between 11 and 17, typically. Ken Pomeroy has found that 11.5 is the best exponent to use for NCAA basketball, making the formula

Pythag = points scored^11.5 / (points scored^11.5+points allowed^11.5)

Problem #2– as we’ve already discussed before, tempo matters in basketball. In baseball, there’s no such thing– everyone gets 9 innings, 3 outs per inning. But in basketball, possessions vary by tempo. So we will use our handy dandy tempo-free measures, offensive efficiency and defensive efficiency, in points-per-100-possessions.

If we plug-and-chug, as a math professor used to say, we get a Pythag rating for Tennessee of 0.980. That means Tennessee has played well enough, according to the formula, to expect to win 98% of its games. That’s good enough for #13 in the country, and for what it’s worth, West Virginia and Texas are 1-2 in that stat. Now, the Pythag is a bit inflated at this point in the season. 98%, by the end of the year, will only have been achieved by the top 3 or 4 teams in the nation, and #13 will be more in the 94-96% range.

And of course, we must qualify this by saying most teams have only played 5-7 games or so. These ratings will fluctuate considerably in the next month, and late in the season they will be much closer to being fixed (and accurate).

Now, the way I use this stat is primarily to rank and compare teams– the actual number, as an estimate for win %, is not terribly meaningful to me. But it does represent, in my mind, the most accurate portrayal of how well a team is actually playing. Note that wins and losses play no part in the calculation, so this is strictly quality of play, not outcome. Some teams will play very well and not have a record that reflects it (Kentucky was #14 in Pythag last year but lost 12 games), and other teams will win more games than their Pythag would suggest (Tennessee was #31 last year). What Pythag allows you to do is to look before last year’s UK-UT game and say that both teams play good basketball, and that Kentucky had a good chance of winning (and did in fact win 1 of 2). Poll rankings and RPI do not give you a very reliable way to make this assessment.

This will be a stat I mention frequently, so I thought it was important to give you at least a working knowledge of what it means. Tennessee is off to a good start in Pythag this year, and hopefully it will remain high as we plow into the meat of our schedule.