View this email in your browser

Hoop Vision Coaching Analytics Newsletter      
Welcome to (unofficially) the first ever coaching analytics newsletter. You don't have to look very hard to find coaching newsletters with X's and O's, motivation, and leadership advice. So a newsletter of the analytics variety seemed only right. Some quick info about me: I am Video Coordinator for New Mexico State University men's basketball. I also have run a college basketball analytics blog, Hoop Vision, since my junior year of high school.

The newsletter was conceived with Division 1 coaches and programs in mind. However, I was surprised and thrilled to get a very diverse group of basketball people signed up after the first release - from fans to coaches to NBA executives. Giving away too many analytics secrets is of course bad for business, so this is designed to really be a conversation starter for coaches. If you are interested in talking further or would like some thoughts on how these ideas can be more specifically applied to your level of basketball, please feel free to email me at

Loyola Chicago and Offensive Balance

Entering the 2017-18 season, Porter Moser had not finished above .500 in conference play since the 2002-03 season. His Loyola Chicago team was picked third in the Valley pre-season poll. The team battled through some injuries in late December, but hit its stride after getting healthy. They won 14 games in a row and became the story of the NCAA tournament with a trip to the Final Four.

The Ramblers biggest strength was on the defensive side of the floor. They finished 17th in the country in adjusted defensive efficiency. However, it's their offensive style that will be the focus of this newsletter. Loyola had a different leading scorer in all five NCAA tournament games. On the season, five guys averaged between 10.5 and 13.2 points per game. Clayton Custer did win the MVC player of the year, but this was a team known for sharing the ball and offensive balance.

In an attempt to put Loyola's offense into perspective, I tried to calculate offensive balance as simply as possible. I took the top five scorers (points per game) for every team in the country during the 17-18 season. I then calculated how much those five values deviate from one another. The result was a "balance" metric.

Loyola ranked #4 in balance out of all 351 teams. The leader of the metric was Gonzaga -- with their top five averaging 13.4, 12.9, 12.7, 12.3, and 11.6. On the flip-side of the leaderboard, it should be no suprise that Trae Young's Oklahoma ranked second-to-last in the country.

The five most and five least balanced offenses are visualized below:

The above graph shows the different ways to be unbalanced. Howard, the most unbalanced team in the country by my metric, did it by getting a combined 44.1 points per game from RJ Cole and Charles Williams. Their third leading scorer, Kyle Foster, averaged just 6.4 points per game.

What's the Relationship Between Balance and Efficiency?
Now that we have a fairly reasonable estimate for offensive balance, the first obvious idea is to look at the relationship between balance and offensive efficiency. In this case though, obvious and informative are two different things. The data shows a small negative correlation between balance and efficiency. In other words, unbalanced offenses were actually slightly better than balanced offenses. This seems counter-intuitive, but there is a pretty reasonable explanation.

The biggest contributor to offensive balance is personnel. Coaching and scheme undoubtedly will play a role, but at the end of the day personnel is going to dictate scoring distribution dramatically. The best example of this is Davidson. Bob McKillop is well know for motion offense that should theoretically produce equal opportunities for all five guys on the court. Yet this season they ranked ninth-to-last in offensive balance. In fact, they've been an unbalanced scoring team for each of the past three seasons. The reason? Peyton Aldridge and Jack Gibbs.

There's a confounding variable that makes unbalanced offenses better than balanced offenses: talent. South Dakota State (Mike Daum), Marshall (Jon Elmore), Oklahoma (Trae Young), and Davidson (Peyton Aldridge) all appear in the list of 10 most unbalanced teams and are big reasons for the small observed correlation.

Because personnel and talent play such a big role in balance, the effects of balance on efficiency is not a particularly interesting question to answer for a coach. If you have South Dakota State's personnel, you're doing your offense a disservice by not featuring Mike Daum. If you have Gonzaga's personnel, you're doing your offense a disservice by not featuring your balance. Neither style is right or wrong, just different.

A More Interesting Question: Are Unbalanced Teams Too Dependent on Their Best Player?

Given Oklahoma's personnel this season, Trae Young was bound to be one of the most ball dominant players in the country. There is probably not a scenario where 2018 Oklahoma wouldn't have ranked at least below average in the balance metric. That's not to say Lon Kruger had no control over the balance. Just how much Oklahoma should play through their star freshman was a constant topic of discussion in the college basketball world this season.

This topic brings up some interesting questions for coaches. The one I want to focus on is: Are unbalanced teams (like Oklahoma) too dependent on getting good games from their best player? 

To answer the question, I eliminated any team in the country that didn't rank as either a top 25 balanced team or top 25 unbalanced team (leaving myself with a 50-team sample size). Then, I went game-by-game for those 50 teams and calculated a metric (z-scores, for the stats nerds) for both how well the best player scored in that game and how well his team scored in that game (relative to the team and player's average).

That methodology might be a little confusing, but the question I'm trying to answer is not. Think about Loyola vs Oklahoma. When Clayton Custer has a bad game, is it less of a problem for Loyola (because of balance) than when Trae Young has a bad game for Oklahoma?

The first graph below takes a look at the Loyola-esque balanced teams and the second graph takes a look at the Oklahoma-esque unbalanced teams:

It's no surprise that in both cases better performances from the best player led to better performances from the team overall. It's also no surprise that the correlation was even stronger for unbalanced (Oklahoma) teams. As we would expect, the Trae Young's of the world have a bigger impact on team success/failure than Clayton Custer.

However, there is somewhat of a surprise if you zoom in on the extremes. Let's look (in table form this time) at just the extremely good and bad games for the best players:

The reason for the stronger correlation for unbalanced teams is actually different than you might think. They actually struggled at about the same rate as balanced teams when their top player had a bad game (40% win percentage for both types of teams). But when their best player had a particularly strong performance, the unbalanced teams dominated (85-39).

If we go back to the four examples I used previously, Davidson/Marshall/South Dakota State/Oklahoma combined to go a dominant 17-4 when Aldridge/Elmore/Daum/Young had their best games of the season.

The 2017-18 data indicates that high volume superstars aren't really playing you out of games on off nights (relative to balanced teams), but they are definitely winning you games when they're clicking. 
Copyright © 2018 Hoop Vision, All rights reserved.

Want to change how you receive these emails?
You can update your preferences or unsubscribe from this list.

Email Marketing Powered by Mailchimp