Raptors:
https://x.com/bballstrategy/status/2029 ... 62433?s=20
LIneup trends around the league
Re: LIneup trends around the league
5 most used lineups in league:
Timberwolves 710 minutes
Pistons 515
Knicks 417
Hornets
Rockets
Approximate minutes for 5 most used lineups:
Celtics 670
Cavs 330
Nuggets 680
Pistons 780
Rockets 940
Lakers 710
Timberwolves 1070
Knicks 800
Thunder 420
Spurs 680
Timberwolves 710 minutes
Pistons 515
Knicks 417
Hornets
Rockets
Approximate minutes for 5 most used lineups:
Celtics 670
Cavs 330
Nuggets 680
Pistons 780
Rockets 940
Lakers 710
Timberwolves 1070
Knicks 800
Thunder 420
Spurs 680
Re: LIneup trends around the league
Only half of teams have a lineup used over 200 minutes.
200 minutes is a commonly cited threshold for trusting or at least using lineup data.
Teams average about 2.5 over 100 minutes. A bit more than 7 over 50. Results in those sized samples are not that reliable.
Almost all lineups are dink or super-dink sized. Entire season yields very very little meaningful information at 5 man level.
But the pattern keeps getting repeated.
200 minutes is a commonly cited threshold for trusting or at least using lineup data.
Teams average about 2.5 over 100 minutes. A bit more than 7 over 50. Results in those sized samples are not that reliable.
Almost all lineups are dink or super-dink sized. Entire season yields very very little meaningful information at 5 man level.
But the pattern keeps getting repeated.
Re: LIneup trends around the league
Average of a bit over 1 quad per team over 500 minutes. 4 over 300 minutes. 8 over 200 minutes. 29 over 100 minutes
Kerr has used 1561 quads. More than 20 NEW ones every game on average. Just 2 over 200 minutes.
Daigneault, 1209. More than 15 NEW ones every game on average. Just 6 over 200 minutes.
Adelman, 805. More than 10 NEW ones every game on average. 15 over 200 minutes.
Celtics and Spurs close to Nuggets pattern. Knicks half way beyonf to Thunder level.
Carlisle's tank takes team to 1841. An average of less than 2 minutes per quad for the season.
For the season, teams will divide nearly 20,000 quad minutes. 3 have reached 1000 minutes (or 5% of total or 25% of possible) and it is unlikely any others will.
Kerr has used 1561 quads. More than 20 NEW ones every game on average. Just 2 over 200 minutes.
Daigneault, 1209. More than 15 NEW ones every game on average. Just 6 over 200 minutes.
Adelman, 805. More than 10 NEW ones every game on average. 15 over 200 minutes.
Celtics and Spurs close to Nuggets pattern. Knicks half way beyonf to Thunder level.
Carlisle's tank takes team to 1841. An average of less than 2 minutes per quad for the season.
For the season, teams will divide nearly 20,000 quad minutes. 3 have reached 1000 minutes (or 5% of total or 25% of possible) and it is unlikely any others will.
Re: LIneup trends around the league
League average regular season lineups used was 627.
Warriors, Bulls and Pacers over 900.
Rockets were only one below 400 at 389.
To reach 627 lineups, a team could play 3 PGs, 4 SGs, 4 SFs, 4PFs and 3 Cs in every combination... and then make up 10% more lineups.
Warriors, Bulls and Pacers over 900.
Rockets were only one below 400 at 389.
To reach 627 lineups, a team could play 3 PGs, 4 SGs, 4 SFs, 4PFs and 3 Cs in every combination... and then make up 10% more lineups.
Re: LIneup trends around the league
A good analytics shop could sim every own lineup possibility against every commonly used lineup by an opponent. before the game. Someone could get the best simulated lineup to the coach in seconds, if the Coach didn't remember or didn't have a cheat sheet.
Or he could just wing it on his own by his gut & experience.
Testing select lineups against lineup types would give more real world data over the simulations. A Coach should want to know which lineups do best against lineup types. Do they?
Or he could just wing it on his own by his gut & experience.
Testing select lineups against lineup types would give more real world data over the simulations. A Coach should want to know which lineups do best against lineup types. Do they?
Re: LIneup trends around the league
Player performance varies so greatly from game to game, more so with certain players, I think the best coaches are quicker to recognize who is sharp and who is flat, who is playing with confidence, and how the refs are calling the game.
If both coaches are quick to react, a lineup change is cancelled by the opponent changing theirs.
If both are using the same database, it's all on the players to prove who was smarter.
If both coaches are quick to react, a lineup change is cancelled by the opponent changing theirs.
If both are using the same database, it's all on the players to prove who was smarter.
Re: LIneup trends around the league
Maybe Coaches can "tell" sharp / flat in a game, possession by possession, but there is a lot of random variance and they may be finding pattern in noise.
"If both coaches are quick to react, a lineup change is cancelled by the opponent changing theirs." And yet this remains the observed behavior over finding & using the durable better lineups and relying on them far more minutes..
"If both coaches are quick to react, a lineup change is cancelled by the opponent changing theirs." And yet this remains the observed behavior over finding & using the durable better lineups and relying on them far more minutes..
Re: LIneup trends around the league
Pistons 2nd, 3rd and 4th most used playoff lineups were 2 bad, 1 horrendous. Just 1 tested over 50 minutes in regular season.
Only 1 playoff pair in 10 most used reached +5 (Cunningham - Thompson). 8 of 10 meh.
Top Pistons pairs on xrapm begin at about 100th best. That isn't near enough.
Only 1 playoff pair in 10 most used reached +5 (Cunningham - Thompson). 8 of 10 meh.
Top Pistons pairs on xrapm begin at about 100th best. That isn't near enough.