High School Applications of APBRmetrics
High School Applications of APBRmetrics
I was recently hired for my first high school varsity job. Exciting stuff. I'm eager to apply a lot of plus-minus and possession-based analysis to our program. Some of it is old-fashioned charting, and I have my own ideas for ways to incorporate more sophisticated data. Acknowledging the sample size limitations of 4-6 pre-season and 20 regular season games, what types of things would you guys focus on?
Re: High School Applications of APBRmetrics
At high school level I'd probably starting with identifying your 2 best PGs, 3 best wings and 3 best bigs (and probably from a team perspective as much or more so than at the individual talent level). Then use everything you know about them and who they might play well with to formulate say 8-12 lineups from them, then play those lineups for as much of the pre-season as you can to gather data on them consistent with whatever commitment your program has toward giving all players some time for their fair chance, development and experience. Then, if you track the lineup +/-, start the season emphasizing the 3-5 best lineups and maybe a lot and see how they do and adjust as necessary.
I'd also try to do detailed shot-charting and work with the kids to try to get them to clearly recognize the shots they are hitting relatively well and not and try to take more from the areas where they are hitting and maybe work more in practice to improve their next best spots and green hit those shots if they show improvement.
All the players should know what the 10 - 15 best player / locations for shots are on the team in general (other than layups) and in any given lineup according to the charting numbers (and maybe they should be called on regularly in practice to identify them or even asked in huddles during games or on the bench to identify where the ball should go) and the offense should be pretty heavily about getting the ball to those players in those locations. If that list is not a sufficient and balanced offense, then identify the locations and the players you need to add to the mix.
If you want to layout some of your own ideas you might get some feedback on them.
Put together your own plan and re-evaluate it as the season progress and tweak it. Try not to change everything at once after every game or every week. Whatever you start with, you'll probably get ideas along the way to improve things. Pick either the most important potential improvement and address them one by one or perhaps pick one you think the team can successfully implement now and work on what is needed to accomplish the harder ones later after some smaller / easier successful changes and after more time on background work needed to fix them.
I'd also try to do detailed shot-charting and work with the kids to try to get them to clearly recognize the shots they are hitting relatively well and not and try to take more from the areas where they are hitting and maybe work more in practice to improve their next best spots and green hit those shots if they show improvement.
All the players should know what the 10 - 15 best player / locations for shots are on the team in general (other than layups) and in any given lineup according to the charting numbers (and maybe they should be called on regularly in practice to identify them or even asked in huddles during games or on the bench to identify where the ball should go) and the offense should be pretty heavily about getting the ball to those players in those locations. If that list is not a sufficient and balanced offense, then identify the locations and the players you need to add to the mix.
If you want to layout some of your own ideas you might get some feedback on them.
Put together your own plan and re-evaluate it as the season progress and tweak it. Try not to change everything at once after every game or every week. Whatever you start with, you'll probably get ideas along the way to improve things. Pick either the most important potential improvement and address them one by one or perhaps pick one you think the team can successfully implement now and work on what is needed to accomplish the harder ones later after some smaller / easier successful changes and after more time on background work needed to fix them.
Last edited by Crow on Mon Sep 19, 2011 9:50 pm, edited 1 time in total.
-
Bobbofitos
- Posts: 306
- Joined: Sat Apr 16, 2011 7:40 am
- Location: Cambridge, MA
- Contact:
Re: High School Applications of APBRmetrics
Do not agree with this. I think, especially at the HS level, positional sensitivity just isn't important. Playing a "center" just because he's a "center" doesn't really make sense, if he's not one of your 5 best players. Try to identify your best players, and simply play them the most - regardless of position. I also think lineup +/- wont matter at all; I think it's far more useful to use a boxscore related metric. (Like scale PER for HS use, for example)Crow wrote:At high school level I'd probably starting with identifying your 2 best PGs, 3 best wings and 3 best bigs (and probably from a team perspective as much or more so than at the individual talent level). Then use everything you know about them and who they might play well with to formulate say 8-12 lineups from them, then play those lineups for as much of the pre-season as you can to gather data on them consistent with whatever commitment your program has toward giving all players some time for their fair chance, development and experience. Then, if you track the lineup +/-, start the season emphasizing the 3-5 best lineups and maybe a lot and see how they do and adjust as necessary.
Shot charting could be useful, but that's more of an approach during practice. ("Try to get layups", "avoid unnecesary midrange Js")I'd also try to do detailed shot-charting and work with the kids to try to get them to clearly recognize the shots they are hitting relatively well and not and try to take more from the areas where they are hitting and maybe work more in practice to improve their next best spots and green hit those shots if they show improvement.
http://pointsperpossession.com/
@PPPBasketball
@PPPBasketball
Re: High School Applications of APBRmetrics
"I think, especially at the HS level, positional sensitivity just isn't important. Playing a "center" just because he's a "center" doesn't really make sense, if he's not one of your 5 best players. Try to identify your best players, and simply play them the most - regardless of position."
It sort of depends of the quality of the high school team and conference. There are teams and conferences where coaches play their 5 best players and that might be fine. There are conferences where if you don't have a clear PG and at least one clear big or a guy who plays like one (if not two) you are going to get killed on rebounding and defense and not have any post offense.
"I also think lineup +/- wont matter at all; I think it's far more useful to use a boxscore related metric. (Like scale PER for HS use, for example)"
A boxscore metric should be part of the decision-making (I didn't mention it, but I assumed it would be used) but I'd use +/- too, especially for role players but I'd want to see it for everyone, even individual "stars", who might be freezing others out and dominating the ball and hurting the offense and perhaps not playing much D either. High school is probably the prime location for this type of player, emulating certain NBA stars who play this style of ball.
It sort of depends of the quality of the high school team and conference. There are teams and conferences where coaches play their 5 best players and that might be fine. There are conferences where if you don't have a clear PG and at least one clear big or a guy who plays like one (if not two) you are going to get killed on rebounding and defense and not have any post offense.
"I also think lineup +/- wont matter at all; I think it's far more useful to use a boxscore related metric. (Like scale PER for HS use, for example)"
A boxscore metric should be part of the decision-making (I didn't mention it, but I assumed it would be used) but I'd use +/- too, especially for role players but I'd want to see it for everyone, even individual "stars", who might be freezing others out and dominating the ball and hurting the offense and perhaps not playing much D either. High school is probably the prime location for this type of player, emulating certain NBA stars who play this style of ball.
Re: High School Applications of APBRmetrics
When charting +/- I would only chart your own substitutions and not those of the opponent. You can adjust for opponent strength later by computing SRS on the entire league and just giving every opponent 5-man-unit a rating that corresponds to the team's overall SRS rating. That way, a lineup of yours that went -1 vs the best team in the league will come out better than a lineup that went -1 vs the worst