APBRmetrics

Looking to help improve efficiency is the goal for our team on both sides of the ball (obviously). Looking for ideas in going about calculating it - not so much focused on the results per say, but what increases our chances of success. I.e. not looking to just eFG%, PPP, or TS% for our defensive metrics, but more so tracking what percent of shots attempted were contested and things like that.

Looking for any type of input similar to the above on both sides of the ball, any ideas are greatly appreciated.

Thanks guys!

CoachO wrote:Looking to help improve efficiency is the goal for our team on both sides of the ball (obviously). Looking for ideas in going about calculating it ...

It seems like you're looking for additional 'primary metrics'. Stuff like 'points off fast breaks', 'shots contested', or 'passes forced' that could - in principle - be added to a player's box score line.

In general, what you calculate will depend on the resources you have. If you have suitable data and computation resources, you could use something analogous to the model in this paper:
http://www.sloansportsconference.com/wp ... l-Time.pdf

If you have training performance data, you may be able to customize training load (and rest) for optimal match day fitness, and player rotation for peak performance.

If it's a guy in the stands taking notes, then I'm not sure how far you can get from the current box score stats.

Something to keep in mind is that everybody knows that 'efficiency is good', but you can't go out on the court and 'be more efficient'. That's not actionable. It may make more sense to start with a specific decision and do analysis that speaks to that decision in particular. For information-driven decision making, it's also important to understand what the expected outcome is, and how to measure it.

Thank you for the reply. We can have a few people tracking data in games (and even practices if need be) similar to what I mentioned.

We do track training performance throughout the year and adjust accordingly as the year goes on.

I don't think we are looking to just change the results based on what we find. But, use them as a tool to backup what we see. If we say player x is not a good defender, what else besides player line-up's are we really looking at? How well they fight screens, play the ball, contest shots, etc. Just trying to find ways to measure these and see what we have. Just looking for ideas and suggestions.

I don't think we are looking to just change the results based on what we find. But, use them as a tool to backup what we see. ...

This may seem like a sarcastic question, but why go to the trouble of measuring things if you've already made up your mind? Are you looking to track player improvement or to set up a more objective player evaluation scheme going forward? Are you having the second string players track the first string so that they learn to make better decisions?

Regardless, it's a matter of making a list of things you want the player to do (offense or defense) and working out ways to keep track of each of them. That can vary by assignment. Basically, every time you tell a player to do something on the court, there should be a way for an observer to tell whether the player is doing it.

I don't really understand basketball that well, but it's easy to come up with obvious things to look for:

Getting back in transition and establishing the defense quickly.
Dribble penetrations allowed
Pass penetrations allowed
Player penetration allowed
Uncontested shots allowed
Good shots and bad shots allowed
Man marking distance
Box out rate and distance from the hoop
Box out coordination with teammates
Keeping opponents out of the low post
Covering opponents in the corners
Smart fouls / not-so-smart fouls

Players love to look at numbers and it can be used as a tool to get them better understand their responsibilities. If we have a number that we are shooting for (i.e. 80% of shots contested) and then we can look to see who ranked where and how the other team scored it provides us with more information. It is only a tool, but if it can help the players then it will help our result as coaches in what we are after.

We will have a team of people to track it at our level.

Offensive/defensive efficiency is the standard, but not everything that goes into efficiency is easily controlled, like 3PT% or FT%.

For instance, there are a number of factors a team has more influence over on defense: proportion of shots at the rim, proportion of shots behind the 3PT line, number of FT's, DRB%, and forcing turnovers through steals, unforced error (out of bounds), or offensive fouls drawn.

No question. We aren't necessarily looking to control those things. Just looking for something where we can basically monitor activity/effort on the defensive end. If we contest 90% of 3pt shots and they shoot 45% from behind the arc, we will live with a team having a good shooting night. But, contesting only 60% of them would be a problem regardless of end result, in our opinion. Just looking to gain ideas on what else might be worth looking into.