With the weirdness this season has already provided, it can be tough to accurately assess performance on the basketball court with just the eye test. On the one hand, I feel most of us would agree Auburn has played well thus far, especially being down their starting point guard and having his backup transfer out. However, Auburn has lost to two of the three best teams it has played, and four of the six wins have come against teams ranked 200th or worse in KenPom.
So how can we talk about what we’ve seen? Well, if you know me, I’m not one to break down film and grade performances, so I rely on cold, hard numbers. With a natural break in the action at the end of non-conference play, let’s dive in.
As a reminder, the Advanced Box Scores for each game and for the whole season can be found in this Google Sheet. While all stats presented have been generated within the sheet, national rankings and averages are pulled from KenPom.com.
Auburn vs Non-Conference Four Factors Table
|Off Reb Rate
Effective Field Goal % - So far, Auburn has simply shot the ball better than its opponents so far this season. However, it’s not in the fashion you would expect from a Bruce Pearl-at-Auburn team. While the offenses the last few years have relied on sharpshooters from beyond the arc, this Auburn offense is shooting over 58% in the paint, good for 21st nationally. While Auburn still shoots the 10th most threes per field goal attempt in the country, the team is only putting a measly 32.5% of them in the net (182nd in the country). While two of Auburn’s guards (Powell, Flanigan) have put up strong numbers from deep this year, the rest of the team is all shooting below 30%.
Defensively, Auburn has allowed buckets at a slightly above average rate in the paint, but the roster’s absurd length and athleticism has created a stout perimeter defense. Opponents are shooting sub 30% from deep, which helps the team allow a eFG% nearly two points below the national average.
Free Throw Rate - This is another area where Auburn has excelled this year. While converting on the extra free throws is another conversation, it is good to see Auburn getting to the line more often that their opponent. While Auburn lost perhaps the best player in the country at getting to the line in Austin Wiley, each of Auburn’s three tree tops (Cardwell, Akingbola, and Thor) are having better years getting to the line than Auburn’s second best player last year (Okoro).
Turnover Rate - This is pretty clearly Auburn’s biggest need for improvement, at the moment. Turning the ball over on nearly 22% of all possessions is not a recipe for success, but I think this looks a lot better when you consider that the team’s top two point guards in the offseason are not currently on the active roster. Not only that, but maintaining possession of the ball is usually one of the last things to come for a young team. Justin Powell, a freshman playing out of position, is averaging a nearly 2:1 assist ratio, slightly better than J’Von McCormick last season.
On the plus side, the Auburn defense has been active in causing turnovers as well. Jaylin Williams and Justin Powell are each averaging around a steal/game, and Chris Moore has been a revelation defensively off the bench. More than just the steals, Bruce Pearl’s press defense also causes a ton of turnovers, as well. While it doesn’t quite make up for the offense’s mistakes, it does help offset it to a degree.
Offensive Rebounding Rate - Offensive rebounding is where Auburn is really beating up on these overmatched/undersized opponents the last few weeks. Compared to the national average of 28.2%, Auburn is bringing in over 35% of all possible boards on the offensive end, while holding opponents to just 26.3%. A big part of those offensive rebounds come from Auburn’s height (7th tallest team in the country), but Chris Moore has been elite on that as well, nabbing over 16% of possible offensive rebounds on his own.
POST-GAME WIN EXPECTANCY (PGWE)
I’ve been experimenting with a new metric this season that should help us go deeper on Auburn’s actual team performance, and goes deeper than just “wins and losses”. Before I get into it, you may recognize this as a variant of Bill Connelly’s Win Expectancy for college football. Per Connelly:
This communicates how frequently a team would have won a specific game given that game’s primary stats. It is intended to say “Given your success rates, big plays, field position components, turnovers, etc., you could have expected to win this game X% of the time.” It has nothing to do with pre-game projections or opponent adjustments.
This offseason, I did a sort of reverse engineering to be able to apply this to college basketball. Essentially, I swapped out Bill’s Five Factors for KenPom’s Four Factors (discussed above), and generated projected scoring margins for a large dataset of games. These scoring margins actually created a nice normal distribution, allowing me to sort outcomes into “percentiles”, which became the Post-Game Win Expectancy.
In short, I take the Four Factor results from a game, the formula creates an expected scoring margin, and then I can calculate how often Auburn should win that game with projected scoring margin. Make sense? While I’m still adding more games to the data set to generate the formula, I’m pretty confident with the direction for now.
Here’s the breakdown thus far -
Post-Game Win Expectancy
|Second Order Wins
|Second Order Losses
This should all line up about how we would expect. Auburn really only clearly lost one game (Gonzaga), but even then performed a little better than the final score showed. However, the St. Joe’s, UCF, and Memphis games all felt like they could’ve gone differently with a little bit of luck either way, and that’s why we see PGWEs hovering between 35-65%.
Another way we can look at this is to sum up the PGWEs into a metric called “Second Order Wins” (also taken from Mr. Connelly). This is a mathematically valid way of saying “based off of Auburn’s performance in each game, they’ve performed more like an X-win team”. Using the eight games played so far, it looks like Auburn may be slightly overperforming, as they have played more like a 5-3 team than a 6-2 team. However, that is splitting hairs, and the difference may be from slight error in my formula.
If Auburn had a second order record of 3.5-4.5 with a real life 6-2 record, you could comfortable say Auburn has gotten lucky and may be due for some negative regression. Meanwhile, if Auburn had a 7.5-0.5 second order record, you could argue that the bounces hadn’t gone Auburn’s way, and that they were playing much better than their record showed.
There are a number of subjective ways to say who the “most valuable” player is in any sport, and anyone who tells you a single statistic can determine it is a fool. However, I’m going to subjectively pick an objective metric to do just that - Adjusted Game Score.
Game score, as formulated by John Hollinger, is a single all-inclusive metric to estimate a player’s productivity, scaled similarly to points scored. Game Score includes points, shots made, shots attempted, rebounds, steals, assists, turnovers, blocks, and fouls. For example, a game score of 10 is decent, while a game score of 30+ is fantastic. Adjusted Game Score (AGS), meanwhile, standardizes each player’s contribution so that the sum of each player’s AGS is equal to the total points scored in a game.
To determine the MVP thus far, I simply chose the player with the highest average AGS over the course of the season.
MVP - Justin Powell
No surprise here, really. The 6’6” freshman has been stuffing box scores all season, between multiple 20+ point performances and leading the team in rebounding AND assists (and of course points). Powell’s been able to rack up the numbers while not even starting the first three games, as well.
As mentioned previously, Powell has maintained a solid 2:1 assist-to-turnover ratio, which still is mind-blowing considering his path to becoming the starting point guard. He also maintains an eFG% of just under 63%, which (over a full season) would be the second best eFG% at Auburn in the Bruce Pearl era. And now I’ve made myself sad about pre-injury Anfernee McLemore, again.
Effective Field Goal %
So after all of that talk hyping up Justin Powell (he’s shooting 51% from three!), it’s time to talk about Allen Flanigan. While Flan was a solid bench player last year who helped fill in when Isaac Okoro went down, he was pretty clearing missing a bunch of pieces in his offensive game. Namely, he had no jump shot, shooting 16.7% from three last year.
This year, Flanigan looks like an entirely different player. The sophomore has shot at a 39% clip from deep, and increased his rate around the basket from 50% to 69%. That’s led to him also having the second best eFG% in the Bruce Pearl era (if maintained for a full year) at 62.9%. The difference between last year and this year is insane, and I think it has been a combination of 1) working with a former All-SEC point guard (his dad), 2) training against the best defensive prospect in the NBA draft (Okoro), and 3) gaining the confidence of being the leader of this team.
This number will drop in conference play. But Flan’s been outstanding so far.
+/- per Minute
So I’m not sure how advanced this metric is, but I wanted to use it to point out a specific player. Chris Moore has posted a +40 so far this year in 92 minutes played (11.5 min/game), giving him a +/- per minute of 0.47, best on the team. To translate, every minute that the freshman wing has been on the court, Auburn has been about half a point better than the opponent. Moore isn’t stuffing the stat sheet by any means, and I’m not an X’s and O’s guy so I can’t tell you what he’s doing right, but it’s clear to me that he’s earned his spot in this rotation.
Defensive Rebounding Rate
Alright, time to circle back to the Golden Child Justin Powell. Some how, some way, Powell is bringing in over 24% of potential defensive rebounds when he’s on the court, best on the team. Think about that for a moment - there are 10 guys on the court at any given time, so you would think each player has an equal 10% chance at a rebound. In practice of course, each defensive player has a better shot at a rebound than each offensive player (about 5.6% for an offensive player and 14.4% for a defender). Meanwhile, your point guard is bringing nearly a quarter of the opportunities. This kid is special.
Points per Shot
PPS is a pretty straightforward calculation. How many points do you score, on average, each time you shoot the ball? While there’s a ton of bias built in, and it’s certainly not a perfect statistic, it’s still fun to look at. Right now, Dylan Cardwell is leading the team on this front. Shooting over 71% for the year (all from two), Cardwell is averaging 1.79 PPS, which is actually just above Austin Wiley’s 1.77 PPS from last year. I fully expect that number to drop in conference play, but it speaks to how physical the freshman is around the basket right now.
Remember, you can find all of my advanced box scores here. This same Google Sheet will be updated after each game!