LIneup trends around the league

Home for all your discussion of basketball statistical analysis.
Crow
Posts: 10533
Joined: Thu Apr 14, 2011 11:10 pm

Re: LIneup trends around the league

Post by Crow »

Lakers:

Starters with Russell positive.

Next 4 most used without Russell and with James presumably at PG all very negative.
Crow
Posts: 10533
Joined: Thu Apr 14, 2011 11:10 pm

Re: LIneup trends around the league

Post by Crow »

Mavs starters with Klay: -15pts / 100p. Next 4 most used with Klay and with and without Doncic positive.
Crow
Posts: 10533
Joined: Thu Apr 14, 2011 11:10 pm

Re: LIneup trends around the league

Post by Crow »

Wizards, 3 most used lineups all with with Carrington - Coulibaly - Sarr, all horrible. Nearly all the horrible bigger minute lineups have at least 2 of them.
Crow
Posts: 10533
Joined: Thu Apr 14, 2011 11:10 pm

Re: LIneup trends around the league

Post by Crow »

Not saying these are the high and low, though they could be:

Knicks 77 lineups used
Griz 290.
Crow
Posts: 10533
Joined: Thu Apr 14, 2011 11:10 pm

Re: LIneup trends around the league

Post by Crow »

TWolves 9 man rotation is being pushed too hard and not getting it done. Dillingham should get more minutes and Shannon and Miller should be given some opportunities. Maybe others as well.
Crow
Posts: 10533
Joined: Thu Apr 14, 2011 11:10 pm

Re: LIneup trends around the league

Post by Crow »

Suns

Beal Booker Durant is only +1 so far this season after being almost +7 last season.

Change of coaches, injuries, about -4.5 pts / 100p change in net margin.
Crow
Posts: 10533
Joined: Thu Apr 14, 2011 11:10 pm

Re: LIneup trends around the league

Post by Crow »

Draft Hawks lineup set:

Y D R J C 14 minutes
Y D H J O 6
Y FD H J C 8
GM KW B J O 6
Y D B R C 2
KW B H N O 4
GM KW H N O 4
Y D R N C 2
Y D H N C 2

Young 34
Mathews 10
K Wallace 14
Daniels 34
Risacher 18
Bogdanovic 12
Hunter 24
Johnson 34
Nance 12
Capela 28
Okongwu 20
Crow
Posts: 10533
Joined: Thu Apr 14, 2011 11:10 pm

Re: LIneup trends around the league

Post by Crow »

Thunder 3 player pairs over 20 minutes per game, 11 over 10 minutes. Celtics, 3 and 16 respectively. Rockets, 10 and 18. Griz 0 and 13. Cavs 2 and 15. Knicks 10 and 15. Mavs 0 and 14. Thunder, lowest amount over 10 minutes per game, Rockets most. Rockets most over 20 with Knicks. Mavs and Griz least. Will they find / use a trio over 10 minutes per game by and in playoffs? Probably. Would be shocking if they didn't. Pretty shocking now.

Nuggets, fwiw, 4 and 18. Bucks 4 and 23.

Which of these are intentional? Which are strategically appropriate? I like the concentration of Knicks and Rockets for testing purposes and future plans. Others might get more concentrated later... or not.

10 pairs in a lineup. Do any have a big minute lineup with all 10 of those pairs over 10 minutes? I'd have to check closer.

Rockets, 1 trio over 20 minutes per game and many close. Knicks 9. Knicks with a quad over 20. Nuggets with a trio over 20. None of these other contenders meet these thresholds.

Only 3 teams have a lineup with 250+ minutes (Knicks, Rockets, Timberwolves).

On average, teams have 14.3 pairs over 10 minutes per game. 2.7 over 20. 6 trios over 10. 0.5 over 20.

Never seen this level of breakdown anywhere. I think there is important information to see, consider and act on.

Looks like about 9,000 trios have been used. Obviously they overlap. About 300 per team.
Wonder how many trio matchups, by same position or otherwise. Probably tens of thousands.

Typically over 100 pairs per team.

For Thunder, 110. 30 over 5 minutes per game and 80 not. 48 under 2 minutes per game. About 30 of those under a half minute per game. For Thunder, about 75% positive, 25% non-positive.

For Celtics, 101. 34 over 5 minutes per game and 67 not. 48 under 2 minutes per game. Only 13 of those under a half minute per game. About 66% positive, 34% non-positive.

Some similarities, some differences for those 2. Celtics using less super dink pairs but might want to try to cut them further.

What are the patterns for most successful pairs and trios?

On average, teams have 5 pairs that are +10 per 48 minutes or better and over 5 minutes per game. 17 of the 50 involve a clear-cut center. Only about 6 of 50 involve a probable PG. Only about 13 of 50 involve a very high usage player.

On average, teams also have 5 trios that are +10 per 48 minutes or better and over 5 minutes per game. At trio level, it is close to half involving a center. Only about 30% involve a PG. It appears only about a third involve a high usage guy but I didn't set an exact / solid usage level.

This research could be refined / extended but take it as it is now.

A general point is try to find more strong pairs and trios and play them more. The rate of presence of a center is higher than for the other criteria checked.
Crow
Posts: 10533
Joined: Thu Apr 14, 2011 11:10 pm

Re: LIneup trends around the league

Post by Crow »

For the very best pairs and trios, how do they perform on average on offense and defense and by factor? How are their stints managed by length, max, quarter, opponent quality, with other starters? 3pt frequency & accuracy both ways, etc. What teams already have these answers and who wants them?
Crow
Posts: 10533
Joined: Thu Apr 14, 2011 11:10 pm

Re: LIneup trends around the league

Post by Crow »

Crow
Posts: 10533
Joined: Thu Apr 14, 2011 11:10 pm

Re: LIneup trends around the league

Post by Crow »

Crow
Posts: 10533
Joined: Thu Apr 14, 2011 11:10 pm

Re: LIneup trends around the league

Post by Crow »

Crow
Posts: 10533
Joined: Thu Apr 14, 2011 11:10 pm

Re: LIneup trends around the league

Post by Crow »

Lineups used:

Thunder 298
Celtics 205
Cavs 214
Griz 406
Knicks 132
Mavs 345
Nuggets 167
Bucks 225
Rockets 213
Timberwolves 100
Warriors 331
Hawks 234
Clippers 193
Heat 279
Sixers 327
Spurs 267
Pacers 240
Lakers 203
Pelicans 352

Grizzlies more than 4× lineups of Timberwolves. Median may be around 250.



Last season:

Knicks 321
Timberwolves 362
Nuggets 327
Celtics 391
Thunder 625
Griz 916
Cavs 485


Ought to be able to operate with less than 400 per season. Or 200 probably. Will want to see Wolves final total. Low on purpose or not consciously? That would be interesting to know. For them and everybody. Few signs of conscious concentration. Few if any rationales for lineup counts as high as they generally are imo.

Lower totals suggest more concentrated testing though there are other measures of concentration, especially meaningful concentration at the top. Concentration is imperative for more than a few significant or even reasonably suggestive test results.

Almost no one else has visibly shown their strong interest / concern / priority for more concentrated testing. No one has ever explicitly defended / justified the extreme lineup chaos beyond the implicit coach's discretion to try to win in the moment. The increased lineup concentration in the playoffs is essentially a concession that regular season lineup spread is wider than optimum. How much of that is because of injuries, rest, curiosity to see "everything" even if it will be unlikely to ever be used again or not really caring would take more study & interpretation. If the lineup totals weren't so over the top, the excuses could have more weight. The Griz using 916 lineups last season would almost require using all possible combinations of 4 guys at every position.

Spurs went to 985 in their 2022-23 tank for Wemby. Pretty effective, somewhat covert method of tanking. Or not caring.

Has anyone ever gone beyond 1000? Not to my knowledge but haven't looked exhaustively. 985 is ridiculous enough. Or same for 600.
Crow
Posts: 10533
Joined: Thu Apr 14, 2011 11:10 pm

Re: LIneup trends around the league

Post by Crow »

Herro - Adebayo, no Butler 9 pts / 100pts worse than as a trio.

https://www.pbpstats.com/wowy-combos/nb ... 89,1629639

Herro and Adebayo better alone than just with each other.
Post Reply