With this piece we have brought in a very treasured guest former Head Coach of Donar Groningen Coach Erik Braal, whom notoriously ran the Groningen club as the top Dutch Basketball League club with the best budget and the best analytics program in the Netherlands.
In this article I’d like to reflect on how we used statistics the during the 5 seasons at Donar Groningen while I was head coach. I will start with a short introduction on how I got involved in looking past the boxscore. After that I will give a brief look into our use of analytics during my time at Donar as a head coach and I will finish with some thoughts about how to use different stats in recruiting players.
When I started coaching men’s professional basketball, I also started my search for winning basketball. While starting as a young, idealistic coach coming from the youth academy, I quickly learned that winning was important. I was used to developing players, but now I had to develop players who could contribute to winning.
As a result I looked at winning teams in different leagues and what they did that made them win. I was fortunate enough to be allowed to coach through my mistakes while studying the game and understanding why teams won. The first understanding was that without good players, it’s tough to win. Recruiting good people with talent is the first requirement if you want to be successful. And that the psychology of players and teams, the group dynamics, needs a lot of attention.
But I also felt that winning basketball could be measured. And the easiest tool at hand was using the boxscore. In 2004 Dean Oliver’s book ‘Basketball on Paper’ revealed a lot of answers. The use of points per possession, offensive and defensive efficiency and the ‘big 4’ of winning stats helped me shift my coaching towards winning. By using possessions to calculate efficiency, it became clear why teams are consistent winners. And by looking at the net-rating of teams, it became easier to predict the standings in the league.
In the years that passed at Donar Groningen we won championships and we were competitive in the Fiba Europe Cup. As stated earlier, when you want to win you need good players first. But (good) players need to be directed and put in position to help the team. Being clear about what you want is a first step, and if it makes sense from a winning standpoint, it creates direction towards the goals you set out at the start of the season. ‘How’ the goals are going to be reached is more interesting than the ‘what’ we wanted to achieve.
Analytics can help and be used to put an offense and defense together and to help players play to the best of their abilities. You can make players become aware of what efficient shots are, what helps run an efficient offense. Good shots are indicators of good offense. And with good offense you limit the chance of beating yourself.
So when looking at our use of analytics we first started with a basic understanding of how we did, in comparison to the rest of the league. We used offensive and defensive efficiency to simply determine how we played in comparison to our direct opponents. It created a benchmark for the league. The best professional leagues provide that themselves, but in smaller leagues creating it yourself can give you insight and an advantage over your opponents.
We would use a spreadsheet with all the available data from the teams we played against. We calculated averages in the league to see if, or in what area, we needed to improve. And this cumulated into a dashboard with averages and graphics to easily see what the league looked like in terms of statistics, both offensively and defensively. This also helped in preparing for any opponent. In a glimpse you would be able to see the average pace of a team, their rebounding percentage, balance between inside and outside play and how they played against various opponents.
This started phase 2 of analyzing things further. What areas we would need improvement in if we wanted to finish the season on top of the ranking? Especially in the beginning of the season we found that our opponents were more efficient in offense, while we still won games because of being the best team in the league in terms of rebounding and defensive efficiency. Just looking at scoring, or points against, would not help to understand this. You needed to take into account pace and percentages, and this is why points per possession gives a lot more information then plain scoring numbers.
For example: this past season Donar Groningen was top-5 in defensive efficiency in the Fiba Europe Cup, but Bottom-5 in offensive efficiency. We turned the ball over way too much (19% of our possessions), and had big difficulty in making shots (eFG% of .449). Analyzing the shots we came to an interesting conclusive detail: Uncontested FG% was 12% lower than our contested FG%.
After understanding what areas we needed to improve in, we would follow up with more detailed analysis. We would break down the offense in parts. The parts we broke it down to were multiple levels, such as transition, post ups, pick & rolls & pops, cuts, drives, shots/plays off screens, and playtypes of these kind. We would calculate what each action contributed to our overall efficiency. All actions and numbers would be backed up by video clips to see who used to be involved and what kind of shots we took. That translated into practice where different shot locations got different point values and 3-pointers could only be taken after being created from an inside-out action.
We would also break down shot clock efficiency. In what part of the shot clock, for instance; if in the first 8 seconds or maybe last 6 seconds, were we most efficient was an important question in that regard. It was a tool to understand what was successful and to put the team and players into these situations more. We would do the same for our opponents, so we would have a very good idea what they were good at, and how we could counter their strengths to increase our chances of winning.
All these parts also reflect what effects different set-plays would have. Good shots need to be created. Having good shooters on the team require also to have good shot-creators on the team. Off the ball cutters, who open up space behind them for an open 3-point shot. And this leads me to my last part of this article. A small glimpse of how we used statistical analysis in recruiting new players.
Having the right people on the team is probably the most important part of the season. To be efficient you would have to be very careful about the balance inside the team. You need finishers, but more important in that regard, is the player who can create the shot for a finisher. Yes, players who can create their own shot have value, but from a team perspective you need good roles for each player on the team. A good shot-creator should be a good decision-maker. If a player understands when to attack and shoot, and when to attack and pass and when to pass immediately, you would have an intelligent player. We would try to measure that by looking at the player’s 2FG% (would he be making more than 1 point per shot) and looking at his assist-rate (balance between turnovers and assists). If both would be high, with enough actions made, there’d be a good chance you’d be looking at a smart player.
With all the calculations and analyzing one thing should be clear. As Hubie Brown once said during a basketball clinic I attended: “Players play the game, not you!”
If you like our insights, scouting and breakdowns feel free to contact us at firstname.lastname@example.org for more about any point of interest.