Razorback Basketball Preview: By the Numbers

“When does basketball season start?”

We’ve all heard that one quite a bit recently. Well, it starts soon, and the Hogs have some reasons to be excited. Mike Anderson’s club went 26-10 (12-6 SEC) last year with a partially-rebuilt roster. The Razorbacks improved down the stretch thanks to a suddenly-stingy shooting defense combined with the development of Jaylen Barford over the final part of the season.

The Hogs lose a few playmakers off that team, but bring back a loaded backcourt and add a solid recruiting class. The Hogs will need to be better, though, as the rest of the SEC may be much improved. If the conference is as strong as the experts say, a 10-8 conference record could get the Hogs into the Dance.

Recapping 2017

I decided to go with a fun, NFL combine-style graph. The higher Arkansas ranked at a stat, the further from the center the point is. Last year’s team was excellent in free throw shooting, avoiding turnovers, and offensive rebounding. The Hogs were bad at defensive rebounding and (surprisingly) forcing turnovers. The Hogs proved to be surprisingly capable at shooting defense.

In football, there are so many different things going on that advanced stats are both difficult to calculate and very informative. In basketball, it’s often the opposite: the stats are much easier to calculate, but may not be as informative. I’m debuting some new numbers this year in the hope that we can quantify the 2017-2018 Razorback hoops team better than ever.

Here’s the general idea of how we track a team’s performance:

Possessions → Scoring Opportunities → Points

The first step in any offense is converting possessions into scoring opportunities. A scoring opportunity is a field goal attempt or a trip to the free throw line. Obviously, the goal of the game is to score points, and you can’t do that if too many of your possessions end without scoring opportunities. Turnovers do this. Offensive rebounds “undo” it, because an offensive rebound essentially “resets” the possession, giving the offense another chance to score. The rate at which possessions are converted into scoring opportunities is called Effective Possession Ratio, or EPR. The formula is simple:

EPR = (Possessions + Offensive Rebounds) / (Possessions + Turnovers)

So if a team has 60 possessions, and they have 8 turnovers and 8 offensive rebounds, that comes out to 1.0 EPR. If they have more turnovers than offensive rebounds, their EPR will be negative: 12 turnovers and 6 offensive rebounds on 60 possessions equals 0.92 EPR. The Division I average is about 0.94 EPR. Mike Anderson’s teams typically have very high Offensive EPRs, since they both avoid turnovers and snag a lot of offensive rebounds. Last year, the Hogs were 54th nationally (0.97) in EPR. That means that for every 100 possessions, Arkansas had 97 scoring opportunities.

You can now see Nolan Richardson’s “40 Minutes of Hell” scheme in context. EPR consists of rebounds and turnovers. So defensively, that means forcing turnovers and securing defensive rebounds. His scheme is a trade-off: he’ll allow bad defensive rebounding as long as he makes up for it with turnovers. The more common strategy is to prefer rebounding to pressure, but Nolan reckoned that there’s an unquantifiable, psychological advantage to being a turnover-forcing pressure defense. Either way, the end result is that your possessions defense should limit scoring opportunities.

Okay, so now that we have a number of scoring opportunities, we move onto the second part: how does the offense convert scoring opportunities into points? The ability to do this is called True Shooting. There’s a simple calculation:

TS = Points / Scoring Opportunities

Here’s the expanded version of the same formula:

TS = (2 * 2FGM + 3 * 3FGM + FTM) / (FGA + 0.475 * FTA)

Free throws are typically shot in groups of two — which is why the FTA denominator is roughly halved — but and-ones and missed one-and-ones mess with that formula, which is why it’s just 0.475.

So if a player is 4 of 10 from the floor with a made 3-pointer and 1 of 2 from the free throw line, his TS is (2 * 3 + 3 * 1 + 1) / (10 + 0.475 * 2) = 10 / 10.95 = 0.91 TS. If you round the denominator, the player scored 10 points on 11 scoring opportunities.

Simple field goal percentage is not a good stat, because 3-pointers are worth more, so you can afford to shoot a lower percentage from beyond the arc and still end up with the same number of points. That’s why I use TS.

So total points can be calculated this way:

Points = EPR * TS * Possessions

But if you want to know efficiency — that is, points per possession — all you have to do is divide both sides by possessions:

Points per possession = EPR * TS

And boom, now we have a simple stat that takes into account almost all of a player’s offensive contribution. Points per possession can be a little misleading because of the difficulty of assigning a “possession” to a single player in basketball, so I call it EV/possession, because it’s essentially a basketball form of the same stat I use in my football analyses: Expected Value.

Let’s run that back for last year’s team.

Trey Thompson was an efficiency machine last year. He had a positive EPR because he totaled more offensive rebounds (30) than turnovers (22). He didn’t shoot often, but he hit 63.6% of his 2-point shots, making him an efficient shooter.

Daryl Macon was Arkansas’ top high-use offensive player, while Barford was the worst. Point guards that aren’t elite shooters are typically near the bottom, since they are the most likely to turn the ball over and the least likely to secure offensive rebounds. Barford is a volume player, so his numbers aren’t that concerning, although he’ll need to get that number up a little bit this year.

C.J. Jones was a fantastic offensive player in limited minutes. His defense is apparently lacking, but Arkansas really needs a pull-up shooting guard (more on that later), so I would not be surprised if he plays a bunch of minutes.

Here’s the Red-White Game:

A monster performance by true freshman Daniel Gafford is the highlight here. But the strong games from Jones and Adrio Bailey shouldn’t be discounted, either. Bailey is competing with the active-but-turnover-prone Dustin Thomas, the pedestrian-and-currently-suspended Arlando Cook, and some newcomers for the starting job at the 4, a position Arkansas desperately needs to see improvement out of. Bailey always seemed like the best scoring option there. Maybe he can get more minutes.

Another way to get more context is to standardize all stats by minutes played. For example, if a guy scores 8 points in 10 minutes and another scores 15 points in 21 minutes, who scored at a higher rate? By calculating per 40 stats, we can see how players would do if they played the whole game, considering their stat trends.

Thompson is a complex player, and he, along with Gafford, will be a key to Arkansas’ success in the post this year. He averaged just 14 minutes per game last year, but based on these numbers, he deserves much more.

“But if Thompson gets more minutes, won’t that cut into Gafford’s?”

Maybe not. I wonder if Anderson is considering experimenting with some Thompson-Gafford lineups where Thompson plays the 5 and Gafford plays the 4. Thompson could draw a post defender up to the top of the key, clearing things out for Gafford. Thompson is a great passer, rebounder, man defender, and shot blocker, so he could complement Gafford’s scoring. Just 8 minutes a game with the Thompson-Gafford lineup could allow Gafford to play 28 minutes a game and Thompson 20.

Another competition to watch is at 4. Thomas is a great offensive rebounder and a surprisingly-good passer. Bailey, Cook, and two newcomers will be left scrapping for minutes. Bailey’s ability to get steals and blocks gives him a leg up. Tinkering with potential lineups:

• BALANCED: Gafford at 5, Bailey/Hall at 4
• BIG: Thompson at 5, Gafford at 4
• OFFENSE: Gafford at 5, Thomas/Osabuohein at 4
• DEFENSE: Thompson at 5, Bailey/Hall at 4

Guards are a little more straightforward. Barford will be a “volume” player, so his efficiency won’t be great but the Hogs will be comfortable putting the ball in his hands. If he can’t improve on last year’s 26.6% 3-point percentage then he needs to defer more shots to Macon, Jones, and Beard. Otherwise, he’s arguably Arkansas’ top player.

Macon and Jones should shoot around 40% from beyond the arc, while Beard needs to match last year’s 36%. Khalil Garland will figure in at the 3, assuming he’s cleared to play. I wonder if Anderson’s “big” lineups could include a forward at the 3. Several SEC teams are bigger this year and the Hogs may need the size.

Keys to the season

1. Knock down the 3-ball. The NBA and college basketball are moving towards more 3-point shooting. The 1994 national championship team was 3rd in the nation with 22.8 3-pointers attempted per game. That would rank 97th last year. Shooting 3-pointers helps spread the defense and can help a team get back after a missed shot. With Hannahs gone, the Hogs need C.J. Jones and Daryl Macon to hit around 40%, and Anton Beard to hit around 35%.
2. Find a 4. The power forward spot has been a consistent weakness for a decade or so now. The Hogs don’t need a star here, but they do need a rebounder and defensive specialist, if not an occasional scorer. There are 5 guys battling for this slot (6 if one of the 5s is played here) and someone needs to emerge.
3. Rediscover that turnover mojo. My theory is that Mike Anderson has improved his team’s halfcourt game year-over-year each season he’s been in Fayetteville. He was criticized in his first couple of years because his teams were only good at fast-breaking. But as the halfcourt game has improved, the Hogs have looked less and less like a “40 Minutes of Hell” team and more and more like a “conventional” team. In some ways, that’s a good thing, since Arkansas typically has more talent than its SEC brethren. In other ways, it isn’t, because in an improved SEC, the Hogs may not be able to simply out-talent everyone on the schedule. The Hogs played a lot of halfcourt defense last year, and fell to an Anderson-worst 176th nationally in forced turnover rate (18.1%). Since EPR consists of rebounding and turnovers, and Arkansas doesn’t rebound well, then they have to force turnovers to make up for it.