Sochan so far: 15% from 3, free throw 1/3rd of average, meh rebounding, little passing, negative DBPM, -7 BPM overall. 26th on BPM in draft class. (Drafted 9th. 25th on my draft board.)
But in 3 best high minute quads.
DRIP projects him at -0.4. He might just be along for the ride with bad and good units.
More time needed to separate context from player impact.
'22-23 Commentary
Re: '22-23 Commentary
Banchero is scoring at about average efficiency, though only 48% on efg%.
BPM up to +2.6.
Raw +/- on court of -9, 9th best on team. Negative pairs (mostly bad) with all of rotation but 2 bench players.
+6 / 100p though with Bol Bol in almost 50 minutes.
Awful with Carter and Okeke. Just bad with Wagner. Horrendous in short minutes with Bamba.
What frontline will be the future?
BPM up to +2.6.
Raw +/- on court of -9, 9th best on team. Negative pairs (mostly bad) with all of rotation but 2 bench players.
+6 / 100p though with Bol Bol in almost 50 minutes.
Awful with Carter and Okeke. Just bad with Wagner. Horrendous in short minutes with Bamba.
What frontline will be the future?
Re: '22-23 Commentary
Celtics core 3 with near neutral pairs and a slightly negative trio.
Right now things are popping positive when Derrick White is on floor.
Worst pair over 50 minutes is Brogdon - G Williams and then they each have far worse, slightly smaller minute player pairs with others. Team results suck for either with Smart but twice as bad for Brogdon.
Brogdon good with other starters, just not Smart and -22 bad with him. G Williams has ok / near neutral team results with a few but only good results are with D White, with whom he plays with less than 1/2 the time.
Chasing super small sample winners may not be wise. But you either do that or something inconsistent with that which may not be wise either. Eventually you either chose to chase good results or not and it is probably wiser to chase good results than stick with meh to bad results that persist over more than a moderate time.
Right now things are popping positive when Derrick White is on floor.
Worst pair over 50 minutes is Brogdon - G Williams and then they each have far worse, slightly smaller minute player pairs with others. Team results suck for either with Smart but twice as bad for Brogdon.
Brogdon good with other starters, just not Smart and -22 bad with him. G Williams has ok / near neutral team results with a few but only good results are with D White, with whom he plays with less than 1/2 the time.
Chasing super small sample winners may not be wise. But you either do that or something inconsistent with that which may not be wise either. Eventually you either chose to chase good results or not and it is probably wiser to chase good results than stick with meh to bad results that persist over more than a moderate time.
Re: '22-23 Commentary
The only positive Evan Mobley pair over 10 minutes is with Osman. Lots of neutral and bad with Allen. Horrible in short minutes with Garland.
Re: '22-23 Commentary
Doncic- Wood is best Mavs pair over 40 minutes at +38 / 48 minutes.
Dinwiddie is negative with all the other starters and everyone he has played more than 1 minute with.
McGee is Doncic's worst pair and modestly negative.
Dinwiddie is negative with all the other starters and everyone he has played more than 1 minute with.
McGee is Doncic's worst pair and modestly negative.
Re: '22-23 Commentary
The 3 pairs of K Porter, J Green and J Smith are all -16 to -20 and the trio is -20 / 100p. Something, at minimum, needs change and massive improvement. Probably more or much more than 1 thing.
10 most used pairs all negative, all but one bad to terrible.
7 most used trios all negative, all but Porter Gordon Smith terrible. That one just bad. 8th most used was Green Martin Eason at a slight positive. I'd certainly get more data on the best bigger minute trios.
Playing with 189 trios is not very conducive to learning about the most used or best.
10 most used pairs all negative, all but one bad to terrible.
7 most used trios all negative, all but Porter Gordon Smith terrible. That one just bad. 8th most used was Green Martin Eason at a slight positive. I'd certainly get more data on the best bigger minute trios.
Playing with 189 trios is not very conducive to learning about the most used or best.
Re: '22-23 Commentary
Beverley James Davis better than Westbrook Beverley James Davis.
Trio without Beverley is either equal to the trio with Beverley or way worse. Small samples and sites vary in their counts. One or more are imperfect.
Trio without Beverley is either equal to the trio with Beverley or way worse. Small samples and sites vary in their counts. One or more are imperfect.
Re: '22-23 Commentary
Gobert - Towns, 10th most used TWolves pair. +4. 2nd most in 10 most used Gobert- Edwards is slightly better. Not that strong.
Zion Williamson, great team results with starters. Completely awful with all of bench.
Harden - Embid, modestly positive. 2nd best in 10 most used but that list is underwhelming.
Zion Williamson, great team results with starters. Completely awful with all of bench.
Harden - Embid, modestly positive. 2nd best in 10 most used but that list is underwhelming.
Re: '22-23 Commentary
Average of 16.3 lineups used per team last night.
David Locke recently expressed amazement that Will Hardy used 19 lineups in a game a few days ago and wondered if it was a record or near record.
It is not.
The norm and modestly above the norm are largely unknown as are the extremes.
I have tried to bring light to the lineup reality over the last few years with little feedback.
It is good to see a few other brief efforts in the market these days.
Acknowledging the reality is step 1. Step 2 is to question its sanity.
Locke marveled at Hardy's ability to get good results that way. A few will, some of the time or on average against similar practitioners. The best or the luckiest. But much / most of the chaos fails and appears inferior to greater discipline and concentration from what the data shows.
Does what NBA coaches do and teams allow for lineup management in regular season qualify as acceptable or intelligent? Not to me and nowhere close. That the playoffs are pretty different does not excuse the regular season chaos or fully redeem the actors in my view. The regular season chaos is beyond excuse or tolerance. The playoffs are better but not as optimized as they should / could be.
I'd really like to know exactly what analytic shops are doing on lineup management and how much is being used and why more isn't.
Instead of 500-800 lineups used in regular season, it could be 100-200 or less or far less. What would be lost by playing 300-600 dink to super dink lineups less or not at all? I'd argue very little. Most get left on shelf after 1-2 uses and come playoffs, so they don't appear that important even to those who throw them out.
Far more important to further test the most used, presumably the best ideas from coach and organization.
If you are satisfied with what the coaches and analytic shops are doing... then carry the hell on. You aren't listening. I dunno if I'll keep commenting on it. It is so obvious and major departure from rational and yet no other leading voice of the "analytics movement" has ever made it a public cause. Unfathomable to me but thus so. Unfathomable on the issue. But it appears Coaches don't like challenges to their practices and it appears few organizations are willing to push such challenges. So fathomable in terms of getting / holding jobs or holding them without major contests and risks. Tap in or actually tap forward.
Is lineup management "better" than 5 - 10 years ago. Not enough... but maybe some? Probably should look into that. I know of no one who has done that publicly.
David Locke recently expressed amazement that Will Hardy used 19 lineups in a game a few days ago and wondered if it was a record or near record.
It is not.
The norm and modestly above the norm are largely unknown as are the extremes.
I have tried to bring light to the lineup reality over the last few years with little feedback.
It is good to see a few other brief efforts in the market these days.
Acknowledging the reality is step 1. Step 2 is to question its sanity.
Locke marveled at Hardy's ability to get good results that way. A few will, some of the time or on average against similar practitioners. The best or the luckiest. But much / most of the chaos fails and appears inferior to greater discipline and concentration from what the data shows.
Does what NBA coaches do and teams allow for lineup management in regular season qualify as acceptable or intelligent? Not to me and nowhere close. That the playoffs are pretty different does not excuse the regular season chaos or fully redeem the actors in my view. The regular season chaos is beyond excuse or tolerance. The playoffs are better but not as optimized as they should / could be.
I'd really like to know exactly what analytic shops are doing on lineup management and how much is being used and why more isn't.
Instead of 500-800 lineups used in regular season, it could be 100-200 or less or far less. What would be lost by playing 300-600 dink to super dink lineups less or not at all? I'd argue very little. Most get left on shelf after 1-2 uses and come playoffs, so they don't appear that important even to those who throw them out.
Far more important to further test the most used, presumably the best ideas from coach and organization.
If you are satisfied with what the coaches and analytic shops are doing... then carry the hell on. You aren't listening. I dunno if I'll keep commenting on it. It is so obvious and major departure from rational and yet no other leading voice of the "analytics movement" has ever made it a public cause. Unfathomable to me but thus so. Unfathomable on the issue. But it appears Coaches don't like challenges to their practices and it appears few organizations are willing to push such challenges. So fathomable in terms of getting / holding jobs or holding them without major contests and risks. Tap in or actually tap forward.
Is lineup management "better" than 5 - 10 years ago. Not enough... but maybe some? Probably should look into that. I know of no one who has done that publicly.
Re: '22-23 Commentary
First comparison:
Steve Kerr's Warriors lineup management
2014-15: Most used lineup, 810 minutes. 3 lineups over the equivalent of 2 minutes per game for season, 7 over 1 minute.
2021-22: Most used, just 346 minutes. 1 lineup over the equivalent of 2 minutes per game for season, 3 over 1 minute.
Far less concentration.
Should look at all years in between and other coaches.
Someone with a good / full database could look at every lineup matchup, expected vs actual results. If it the lineup chess that matters more than the lineup itself, show the results.
Steve Kerr's Warriors lineup management
2014-15: Most used lineup, 810 minutes. 3 lineups over the equivalent of 2 minutes per game for season, 7 over 1 minute.
2021-22: Most used, just 346 minutes. 1 lineup over the equivalent of 2 minutes per game for season, 3 over 1 minute.
Far less concentration.
Should look at all years in between and other coaches.
Someone with a good / full database could look at every lineup matchup, expected vs actual results. If it the lineup chess that matters more than the lineup itself, show the results.
Re: '22-23 Commentary
The Warriors most used lineup minutes were steadily down from 2014-15 to 2019-2020. Rise since the absurd 123 minutes in 2019-2020.
Number of lineups over 2 minutes per game never increased from 4 and toggled to 0 or 1 three times including last season. Number over 1 minutes has also steadily fallen by over 75% since start of Kerr. Number of player pairs over 1000 minutes cut by more than half.
Not that random. Explicitly intentional? I dunno. Made it work? I guess. Wise? I want to say no but more study may be appropriate. Need to see league trends and matchup results. Talent is likely the biggest factor, bigger than lineup management.
Number of lineups over 2 minutes per game never increased from 4 and toggled to 0 or 1 three times including last season. Number over 1 minutes has also steadily fallen by over 75% since start of Kerr. Number of player pairs over 1000 minutes cut by more than half.
Not that random. Explicitly intentional? I dunno. Made it work? I guess. Wise? I want to say no but more study may be appropriate. Need to see league trends and matchup results. Talent is likely the biggest factor, bigger than lineup management.
Re: '22-23 Commentary
Between 2014-15 and last season, the number of lineups in league used for 600 or more minutes has ranged from 2 to 6. 3 in 19-20, 2 last season. So 94% of teams last season didn't have a lineup much used over 7 minutes per game or about 15% of time.
Total lineups used in league is a fact no longer accessible on NBA.com (with a 2000 lineup display limit); but when it was possible a few years ago, I found it to be about 20,000. I'd like to track the trend, but will need some other resource to find.
Total lineups used in league is a fact no longer accessible on NBA.com (with a 2000 lineup display limit); but when it was possible a few years ago, I found it to be about 20,000. I'd like to track the trend, but will need some other resource to find.
Re: '22-23 Commentary
There are 17 lineups over 50 minutes right now. The number over 10 or even 7 minutes per game probably fades far. But I'll watch to see if the revolution has started or it is just a small sample phase.
Of course injuries and roster changes affect the long run but concentration assessments will still come up for consideration / discussion.
Of course injuries and roster changes affect the long run but concentration assessments will still come up for consideration / discussion.
Re: '22-23 Commentary
There are currently 55 player pairs over 150 minutes. The teams with the most and presumably the conscious belief in the value of doing so are Pistons, Timberwolves, Sixers and Blazers. All with at least 5. 4 teams with 40% of the total, almost half the league have none. The middle third have about an average amount on average. Different ways to roll.
Booker - Bridges way way ahead of all qualifiers, D Murray - J Collins second and only a handful of others I'd call "good". Selection for big minutes does not appear very successful. Less than half of the qualifiers are positive. Would be wise if the selection were better. There are candidates which might be successful if not for current design / thinking / complacency with not successful big minute pairs.
Booker - Bridges way way ahead of all qualifiers, D Murray - J Collins second and only a handful of others I'd call "good". Selection for big minutes does not appear very successful. Less than half of the qualifiers are positive. Would be wise if the selection were better. There are candidates which might be successful if not for current design / thinking / complacency with not successful big minute pairs.
Re: '22-23 Commentary
Doncic, currently 10th on team for team +/- on court (+4), 11th on / off (-9). Despite the big individual stats.
Projection:
14th best projection of DRIP in league, 18th on DARKO. DARKO has him treading water short of +4 for 3 plus years.
Projection:
14th best projection of DRIP in league, 18th on DARKO. DARKO has him treading water short of +4 for 3 plus years.