Sunday, 16 March 2014

Art Rondeau on the Psyche of Shooting: Part One

At the Sloan Sports Analytics Conference, I had the pleasure of meeting Art Rondeau, a Trainer of Neuro-Linguistic Programming (NLP), and Peak Performance Coach. Art has worked extensively with professional athletes in a number of different realms, and his role is part of what makes his perspective both diverse and unique. Art has compiled a number of works for Hickory High, operates his own blog (Game Time at the Garden of Good and Evil), and has comprehensive list of testimonials listed here. I was fascinated by Art's work and his approach, reached out to him for a few insights on possible misperceptions and issues with the game, and the NBA, and he kindly obliged. You can find more on Art at, and follow him on Twitter @ArtRondeau. The following is part one of our two part exchange.

Angus Crawford: Recently, you put together a guest post for Hickory High discussing the standard practices of free-throw shooting, and how certain aspects of it add unnecessary length to a game's duration, and clog the overall flow. Aesthetically, how much of a concern do you think the time it takes the average fan to watch/follow an individual game is for the NBA? Which elementary components of the game might need remodelling in order to better serve the league (and its product) as a whole?

Art Rondeau: There are parts of many games that are brutal to watch (“Hack-a-Shaq” and the numerous timeouts late in the 4th quarter to name two). Eliminating or improving these situations (and I'm working on articles suggesting fixes to both of them) would help a lot. When new NBA Commissioner Adam Silver was interviewed at the Sloan Sports Analytics Conference, he mentioned that studies show that the average time fans watch games on television has dropped from 50 minutes to 30 minutes. So, in addition to some over-the-top complaining by some NBA analysts who seem like they'd like the NBA to play a five game season and just 20 minutes per game, there is evidence that people are watching less for some reason(s). I think it's in everyone's best interest to find ways to make the game more watchable but to do so in ways that don't change the game too much. Dramatic changes probably aren't going to happen and if enough simple changes were made, dramatic changes would not even be necessary.

The guest post that you referred to suggests not allowing defenders to line up for free throws until the last one is ready to be taken (since it doesn't matter who rebounds a missed first free throw when the shooter's at the line for two). That would save between 30 seconds and a minute for each free throw, saving 10-20 minutes per game. The other suggestion was to wait to shoot technical free throws awarded during the first 3 quarters of games and shoot them right before the 4th quarter. In some games, there'd be no saving at all. In others, that might save 5-10 minutes. Eliminating “Hack-a-Shaq” would save twenty minutes in some instances and save no time in most. Changing how timeouts are handled late in the game would save time and make the games more exciting. Another area for improvement is inbounds plays; saving five minutes here and ten minutes there would benefit all concerned.

What must be avoided is changing the game so much that you can't compare players from different eras. There has been an idea floated to eliminate free throws all together (I'll pause here for "Airplane" fans to all say "eliminate free throws") and I've written an article to show why that would be a bad idea based on historical comparisons. One idea that has been floated is to cut the game from 48 minutes to 40 minutes. But if you reduce the game by almost 17%, you ensure that no one coming into the league will ever break onto the top ten all-time scoring list. As I mentioned in my "saving time" post, if you think player introductions take too long, the answer isn't having the teams play 2 on 2.

AC: One of the heavily-discussed research paper presentations at Sloan covered the idea of the "Hot Hand," and its validity. What have your studies and your work produced on the subject of the psychology behind players netting consecutive shots? Where do you stand on the so-called "Hot Hand" mindset?

AR: Contrary to most of the research, the Sloan paper suggests that a Hot Hand exists. I'm on board with that. In trying to disprove that the Hot Hand exists over the years, there are a number of flaws that were made in the studies due to lack of knowledge at the time of the study and some due to other errors. Without spending too much time on the flaws, it's important to list a few of them because shaking some of the belief in the "No Hot Hand Exists" studies is important in proving that the Hot Hand does actually exist. In what is probably the most well known study, Gilovich, Tversky, and Vallone looked at a year's worth of shooting data from the Philadelphia 76ers to determine if there was evidence that anything but random series of made shots existed. They also defined the Hot Hand as the likelihood that a player's next shot has a greater chance of going in if he hit his prior shot. They used free throws in the study and, as you see above, there are too many issues causing players to miss free throws to use them to prove or disprove anything like this. Finally, for here, they discount players' beliefs in Hot Hands.

All of this causes problems. It's very possible for a player to have a Hot Hand (what I'll define here as an out-of-the-ordinary game shooting well above his average for reasons other than blind chance) without hitting a lot of consecutive shots. If X is a make and O is a miss, we could see a normally 40% shooter have a game where he shoots X-X-O-X-O-X-X-O-X-X (70%) while only hitting consecutive shots 3 times. If he shoots something similar 3 games in a row (maybe missing his first shot in the 2nd game to break up the Xs), would we say he didn't have a Hot Hand because he didn't have many consecutive makes? Some would but I think most people would realise that something special was going on. A Hot Hand doesn't mean perfection, but holding it up as perfection helps when trying to prove that it doesn't exist.

Defining making a shot or missing a shot as a random act like flipping a fair coin is not accurate. Coins don't get tired. The defense doesn't double-team coins after a couple of heads in a row, making it harder for the coin to come up heads a 3rd time. The coach can order the player to shoot on a given play - whether the shooter wants to shoot or not. The coin doesn't care.

In looking back at the 76ers' data or at our example of X-X-O-X-O-X-X-O-X-X, why were the shots missed? How many times did the shooter miss a shot that he only took because the shot clock was about to expire and he chucked it up just to give his team a chance at the rebound? How many times did he take a shot that would have gone in but it got blocked (something that goes into the box score as a miss)? How many times did the player take a shot he knew probably wouldn't go in because he'd been fouled and wanted that foul to be in the act of shooting so he could go to the free throw line? We don't know and neither did the people who examined the 76ers' data. None of those three examples of why a player might miss is proof that the player didn't have a Hot Hand at the time that he missed the shot. They're only proof that the player missed. Without knowing why he missed, we can't legitimately disprove the Hot Hand.

Discounting beliefs is a mistake because a person's belief in his/her ability to achieve an outcome can have a significant impact on what the person actually achieves. Almost invariably, when I work with an elite athlete, one of the things that we'll do is to identify any negative beliefs about their ability to win and replace them with positive beliefs. There are studies that are helping to quantify the power of beliefs -- for example, identifying changes in blood chemistry when a belief changes -- but in my own work I have many examples of negative beliefs being the cause of an athlete's poor performance and installing a positive belief being part of the program that breaks the athlete out of a slump and/or helps him/her to excel during the next game.

In working with basketball players, I'm able to provide better examples of a relationship between the athlete doing a customised series of mental exercises and a superior shooting performance. In a very small sample, former Providence College player Maris Laksa shot 50% and, in the next game, 62.5% in his 2 games on my program after having been in a 6-week slump. In a larger sample, during the 1999-2000 NBA regular season, New York Knicks shooting guard Allan Houston shot 60% or better in just 5 of 52 games when he was off my program and shot 60% or better in 15 of 30 games when he was on it. Allan shot 50% or better in just 10 of 52 games when he was off my program and shot 50% or better in 27 of 30 games when he was on it. When an NBA player who is considered an excellent shooter can only shoot 50% or better in one out of every five games on his own but manages to shoot 50% or better in 9 of every 10 games after doing customised mental exercises, I think there's enough proof that how well a player shoots isn't random. More data is needed, but I'm off to a good start.

You can follow up on Art's work with Allan Houston here (inc. detailed statistics), and read this New York Post feature on Art teaming up with the former Knicks shooting guard.

No comments:

Post a Comment