In which I might complain a bit about shooting variance.

Every so often I’ll venture into the comments on game threads. And one of my main takeaways is I feel like sometimes people don’t read what we write.

As this game was unfolding, it was clear Missouri was the better team. They were dominant on both sides of the ball. They were crushing Howard at the rim, turned them over consistently, and made their lives miserable on offense. Howard struggled to do nearly everything on offense all game long. Well, except one thing.

Meanwhile, on the offense end, the Tigers offense wasn’t crisp but they were able to attack the rim, run efficient secondary offense, and really had their way for the most part. Oh… except for one thing.

Have I, or anyone else at this blog, ever uttered the phrase “did you make your threes?” I have, right? Many times.

Here’s the thing about variance, most of the time you play within your averages. Every so often you have a shooting variance game, sometimes it’s good and sometimes it’s not. What’s rare though, is when both teams have a shooting variance night but one team is really really bad and the other team is really really good. That, well it happened last night. Fortunately for Missouri they were that much better than Howard, and the shooting variance didn’t really hurt them past making the game closer than it should have been.

But I stress to you and I hope you believe me when I say this: Three point shooting defense is, by and large, randomized results. The only reason anyone got upset about this game is because Howard made over half their threes, and Mizzou had their worst shooting night since going 3 for 28 against Alabama two years ago. All things being equal, this game was much closer to a 30 point win than it was what the score reflected.

Team Stats

Really, the Tigers shot 13.6% on three made 3FGA all from Tamar Bates. 22 attempts. They could have doubled that number and made 6, still shot poorly and won by 24, while still allowing Howard to shoot over 50%. There are few things more frustrating than when you watch a shooting variance game, but we did!

• Here’s what 3FG variance gets you: Mizzou shot 72.4% from inside the arc to Howard’s 33.3%. Howard attempted more inside the arc than the Tigers did, which hints that Missouri was effective in running Howard off the 3-point line. If you’re hitting over half your threes you might want to try and get off more than 19 shots, but that’s all they had.
• So if you take everything other than 3-point shooting away: this was dominance. You don’t get to do that part though, so that’s where things stand. The bones of this team are very good. You can see the blueprint.

Sure there are a few things I’d like to see them do better, fewer turnovers against a team like Howard. More assists (although assists largely come from good jump shooting). I just don’t have many complaints about how things went other than the frustration of having to watch a shooting variance night.

Player Stats

Your Trifecta: Tamar Bates, Mark Mitchell, Anthony Robinson II

On the season: Tamar Bates 5, Anthony Robinson II 4, Mark Mitchell 2, Aidan Shaw 1

Matt Harris told you to buy Ant Robinson stock and I hope you did, because the early returns are promising. While Tamar Bates has been as good as expected, Ant has probably been the teams best player in the first two games. Maybe if you’re Dennis Gates that’s not what you wanted. Not as much because of Ant, and more because you hope your higher priced players might be carrying more weight, but it’s promising. He’s played 50 minutes, scored 29 points, 7/8 from 2FG and just ¹⁄₄ from 3FG, but 12/15 from the FT line and a 147 Offensive rating so far this year. That’s good.

It was also nice to see Gates feed Mitchell early and often. Plenty of rim opportunities, some easy chances, and lots of fouls. Perfect for getting Mitchell going.

The offense looked a little more like you would want it to. You can’t have everyone having a good day, and Tony Perkins and Caleb Grill certainly didn’t have good nights, nor did Jacob Crews.

Josh Gray needs to cut down on the turnovers. But 8 players with a Floor Rate over 40% is what you want. If lineup data is your thing, RockM+ is the place for you. I’ve always like Matt Harris’s collection of the lineup data, I think it’s an undervalued analysis approach. Mizzou didn’t have a lot of weaknesses again Howard, but they did struggle to really put the Bison away when they had them on the ropes. But ah the three point shooting.

Any time Mizzou extended their lead, Howard hit a three. Up 13? Don’t worry, Howard made a 3. Go up 14! Howard makes a 3. The Bison were pesky, to the point where a three point make cut the lead to four points.

But then the Tigers went on an 11 point run keyed mostly off of defense. Maybe it’s disappointing to see a team struggle to put an inferior opponent away, but for me the frustrating was watching a shooting variance night happen and knowing it’s a shooting variance night as it’s occurring.

Mizzou was 30 points better. And on 8 out of 10 nights it’s a 26-30 point win. On this night it was a 15 point win, the other night it’s a 40 point blowout.

---

True Shooting Percentage (TS%): Quite simply, this calculates a player’s shooting percentage while taking into account 2FG%, 3FG%, and FT%. The formula is Total Points / 2 * (FGA + (0.475+FTA)). The 0.475 is a Free Throw modifier. KenPomeroy and other College Basketball sites typically use 0.475, while the NBA typically uses 0.44. That’s basically what TS% is. A measure of scoring efficiency based on the number of points scored over the number of possessions in which they attempted to score, more here.

Effective Field Goal Percentage (eFG%): This is similar to TS%, but takes 3-point shooting more into account. The formula is FGM + (0.5 * 3PM) / FGA

So think of TS% as scoring efficiency, and eFG% as shooting efficiency, more here.

Expected Offensive Rebounds: Measured based on the average rebounds a college basketball team gets on both the defensive and offensive end. This takes the overall number of missed shots (or shots available to be rebounded) and divides them by the number of offensive rebounds and compares them with the statistical average.

AdjGS: A take-off of the Game Score metric (definition here) accepted by a lot of basketball stat nerds. It takes points, assists, rebounds (offensive & defensive), steals, blocks, turnovers and fouls into account to determine an individual’s “score” for a given game. The “adjustment” in Adjusted Game Score is simply matching the total game scores to the total points scored in the game, thereby redistributing the game’s points scored to those who had the biggest impact on the game itself, instead of just how many balls a player put through a basket.

%Min: This is easy, it’s the percentage of minutes a player played that were available to them. That would be 40 minutes, or 45 if the game goes to overtime.

Usage%: This “estimates the % of team possessions a player consumes while on the floor” (via sports-reference.com/cbb). The usage of those possessions is determined via a formula using field goal and free throw attempts, offensive rebounds, assists and turnovers. The higher the number, the more prevalent a player is (good or bad) in a team’s offensive outcome.

Offensive Rating (ORtg): Similar to Adjusted game score, but this looks at how many points per possession a player would score if they were averaged over 100 possessions. This combined with Usage Rate gives you a sense of impact on the floor.

IndPoss: This approximates how many possessions an individual is responsible for within the team’s calculated possessions.

ShotRate%: This is the percentage of a team’s shots a player takes while on the floor.

AstRate%: Attempts to estimate the number of assists a player has on teammates made field goals when he is on the floor. The formula is basically AST / (((MinutesPlayed / (Team MP / 5)) * Team FGM) – FGM).

TORate%: Attempts to estimate the number of turnovers a player commits in their individual possessions. The formula is simple: TO / IndPoss

Floor%: Via sports-reference.com/cbb: Floor % answers the question, “When a Player uses a possession, what is the probability that his team scores at least 1 point?”. The higher the Floor%, the more frequently the team probably scores when the given player is involved.

Touches/Possession: Using field goal attempts, free throw attempts, assists and turnovers, touches attempt to estimate, “the number of times a player touched the ball in an attacking position on the floor.” Take the estimated touches and divide it by the estimated number of possessions for which a player was on the court, and you get a rough idea of how many times a player touched the ball in a given possession. For point guards, you’ll see the number in the 3-4 range. For shooting guards and wings, 2-3. For an offensively limited center, 1.30. You get the idea.

Anyway, using the Touches figure, we can estimate the percentage of time a player “in an attacking position” passes, shoots, turns the ball over, or gets fouled.

In attempting to update Study Hall, I’m moving away from Touches/Possession and moving into the Rates a little more. This is a little experimental so if there’s something you’d like to see let me know and I’ll see if there’s an easy visual way to present it.