Checking in on Celtics lineup usage (and other teams)
Re: Checking in on Celtics lineup usage (and other teams)
After 3 games the substitute starting lineup is -15 per 100 possessions in almost twice the minutes as the regular season testing. Stick with it or change? Series on the line.
Re: Checking in on Celtics lineup usage (and other teams)
So it looks like Stevens' substitute starting lineup helped the first time out then was neutral twice and a big negative once. Eh. Not enough.
By contrast Hawks got 35 of their 40 points series advantage from their big minute starting lineup. Its strong playoff performance substantially exceeded a meh regular season average and a slightly negative regular season run against the Celtics. Performance varies. Maybe especially for small sample size lineups.
Bottom-line, Celtics lost. And lost using 88 different lineups in less than 300 minutes. Pretty chaotic, imo. Not quite as chaotic as last season but still heavy on the dink lineups. It appears that when it was Hawks starters vs. Celtics dink lineups, Hawks won handily. Consistent with the general trend that big minute lineups usually beat the dinks. And yet some try to buck or ignore this trend.
By contrast Hawks got 35 of their 40 points series advantage from their big minute starting lineup. Its strong playoff performance substantially exceeded a meh regular season average and a slightly negative regular season run against the Celtics. Performance varies. Maybe especially for small sample size lineups.
Bottom-line, Celtics lost. And lost using 88 different lineups in less than 300 minutes. Pretty chaotic, imo. Not quite as chaotic as last season but still heavy on the dink lineups. It appears that when it was Hawks starters vs. Celtics dink lineups, Hawks won handily. Consistent with the general trend that big minute lineups usually beat the dinks. And yet some try to buck or ignore this trend.
Re: Checking in on Celtics lineup usage (and other teams)
Jackie MacMulan brings you "Isaiah Thomas: Brad Stevens could be one of the greatest coaches who ever lived"
Here is some other information.
Brad Stevens moved up to now rank 22nd amongst coaches active this year on career win% after the season. He ranks no better than 17th on playoff win%. He is tied for last in playoff series wins. Instead of losing by 9 pts per game last playoffs, he cut it to about 7... against an opponent 2 points weaker on SRS than last season's opponent. Progress getting 2 wins but not particularly close on points. On offense his playoff team was below average on all 4 factors and only near to average on one. Overall it was 14th on offensive rating and 4th on defensive rating couldn't keep it close.
The 88 lineups he used got non-positive results 67% of the time. Only one lineup got 6 minutes per game and only 5 got more than 2 minutes per game. Outside these still lightly used top 5, the 83 other dink and superdink lineups averaged barely 2 minutes for the series. Stevens used the 3rd most lineups, 26% above the average.
Take the positive with the non-positive. Or stay with just the hype, if you want.
Here is some other information.
Brad Stevens moved up to now rank 22nd amongst coaches active this year on career win% after the season. He ranks no better than 17th on playoff win%. He is tied for last in playoff series wins. Instead of losing by 9 pts per game last playoffs, he cut it to about 7... against an opponent 2 points weaker on SRS than last season's opponent. Progress getting 2 wins but not particularly close on points. On offense his playoff team was below average on all 4 factors and only near to average on one. Overall it was 14th on offensive rating and 4th on defensive rating couldn't keep it close.
The 88 lineups he used got non-positive results 67% of the time. Only one lineup got 6 minutes per game and only 5 got more than 2 minutes per game. Outside these still lightly used top 5, the 83 other dink and superdink lineups averaged barely 2 minutes for the series. Stevens used the 3rd most lineups, 26% above the average.
Take the positive with the non-positive. Or stay with just the hype, if you want.
Re: Checking in on Celtics lineup usage (and other teams)
The Warriors used 94 lineups in round 1. Injuries contribute to that. It is still bizarre to me though to use that many after a season where presumably you try to sort what to use and what to leave out. But at least they got positive results 41% of the time.
Rick Carlisle used 96 lineups. But only got positive results 21% of the time.
Stan Van Gundy used the least lineups, just 43. 33% were positive.
Use what you want. Got to get a decent share of positive results and probably some big ones. 40% appears to be enough to be the best or close to it.
Spurs, 65 lineups, 38.5%. Thunder 75 lineups, 41% positive. Cavs, 50 lineups, 32% positive.
It could come down to a few lineup calls. Relatively tested or dink. Positive or not.
Rick Carlisle used 96 lineups. But only got positive results 21% of the time.
Stan Van Gundy used the least lineups, just 43. 33% were positive.
Use what you want. Got to get a decent share of positive results and probably some big ones. 40% appears to be enough to be the best or close to it.
Spurs, 65 lineups, 38.5%. Thunder 75 lineups, 41% positive. Cavs, 50 lineups, 32% positive.
It could come down to a few lineup calls. Relatively tested or dink. Positive or not.
Re: Checking in on Celtics lineup usage (and other teams)
If you make 40% positive lineup calls are you a lineup genius? Maybe, comparatively. But worth knowing it is 40% positive and not 50, 60 or 70%. The very best of the inside experts barely hit 40%. Tough. But the achievement level is lower than many might think or are lead by coverage tone to believe.
Re: Checking in on Celtics lineup usage (and other teams)
If you only win about 33% of your lineup calls, you are typical. Or a typical genius if you want to hype it. You lose unless you face another typical (as was the case for Budenholzer over Stevens) or the few who did worse.