A few recent articles on lineup management. Probably up from past in general.
Are teams doing more or different? I dunno. But probably need still more and perhaps more different.
2019-20 lineup analysis
Re: 2019-20 lineup analysis
Nuggets, 2 of 7 most used lineups positive and only 8 of most used 20. Starting lineup is good, uses a lot and compensates but this is not a good or adequate line-up profile to take to the playoffs.
Most used 20 player pairs has some strong ones. ALL of these negative pairs have Grant. Would they sit him in playoffs? When? Most coaches change too late. -10 on / off. Most of the off would have Milsap but Milsap is neutral on RPM. Grant is only -1.23. A lot of the on must be bad lineups. Nuggets have talent and even good pairs but Malone has found ways to produce lots of negative lineups.
Should look month to month to map changes. The number of dink lineups is comparatively low and the rate of positive performance appears average. The flaw appears to be in a lot of what got elevated into the most used 20. 11 of the negative 12 in this group have Grant.
Most used 20 player pairs has some strong ones. ALL of these negative pairs have Grant. Would they sit him in playoffs? When? Most coaches change too late. -10 on / off. Most of the off would have Milsap but Milsap is neutral on RPM. Grant is only -1.23. A lot of the on must be bad lineups. Nuggets have talent and even good pairs but Malone has found ways to produce lots of negative lineups.
Should look month to month to map changes. The number of dink lineups is comparatively low and the rate of positive performance appears average. The flaw appears to be in a lot of what got elevated into the most used 20. 11 of the negative 12 in this group have Grant.
Re: 2019-20 lineup analysis
The Nuggets either figured some stuff out recently or it broke positive on its own. In last 15 games, the bigger minute Grant lineups have improved to 2-1 positive.
But of late the starting unit turned quite negative. And prominent Craig lineups are negative.
Lots to look at and keep up to date on. One problem perhaps resolved. Several new ones.
But of late the starting unit turned quite negative. And prominent Craig lineups are negative.
Lots to look at and keep up to date on. One problem perhaps resolved. Several new ones.
Re: 2019-20 lineup analysis
Wizards had perhaps the worst performing biggest minute lineup that I can recall (almost -18).
Virtually no big minute lineup or sub-lineups positive without Bertans. They better re-sign him. Or try a very different approach.
Beal, 38 pts / gm over last 10 on great shooting and they still lost his time on the court.
Virtually no big minute lineup or sub-lineups positive without Bertans. They better re-sign him. Or try a very different approach.
Beal, 38 pts / gm over last 10 on great shooting and they still lost his time on the court.
Re: 2019-20 lineup analysis
Less than half of teams with a lineup used 300 minutes. Only a couple surpassed 500 minutes, in part because only a couple played 40 plus of the games. Of the 11 positive lineups, all had a legit center and only half an active 3pt shooting PF. Only 3 had much youth- Heat, Bulls and Griz. Only 2 overall played fast and most were well slower than average (for whatever reasons). 2/3rds shot less 3s than opponent and only half shot them better. But 80% had a better efg%. Modest edge on ftas. Would need to look elsewhere to check shot at rim. The Moreyball story is mixed but not sure which way the balance is tipped overall.
Re: 2019-20 lineup analysis
Sixers talk about giving Brett Brown constant specific lineup advice.
And yet, only 2 Embid lineups have been used over 1 minute per game.
In other news, the 2 best lineups for Sixers out of only 7 total used over 1 minute per game are non-Embid lineups.
Only 2 lineups used over 2 minutes per game and none over 4 minutes / gm.
Most of these things are shocking or just ridiculous depending on your expectations.
On the brighter side 5 of 6 most used lineups are positive, as are 19 of the 20 most used pairs... for the season.
Things might be considered ok or good. But they probably could be better if the well performing lineups and maybe even sub-lineups were played more.
The 3 best performing bigger minute pairs are all between Embid, Richardson and Scott. In raw terms they are twice as good as any other bigger minute pairs or nearly so. The trio is more than twice as good as 85% of the 20 most used trios and yet only plays about 5 minutes per game, which is less than 90% of these trios. More analysis should be done on this trio before making a major change... but has it been done? And was the decision not to elevate it? Why? By the coach and / or the analytic staff?
Embid - Horford is the worst performing pair in the most used 20 and slightly negative. So what to think / say about the massive contract signing decision, the analytic advice that preceded that, the coaching that followed and the players themselves' interaction and performance? Well, let's first check how it has done recently. Any improvement or, in lieu of that, minute reduction? Absolutely the opposite. Huge minute increase in last 15 games (to almost 2.5 times the usage level that preceded it) and a dramatic decline to about -9 pts per 100 possessions. This is a total debacle. Coach, analytics, management. I guess the pair was in fact about neutral before it went to hell but they did not stop it from going to hell or simply limit the minute exposure. Will they fix it? I dunno. They better fix it one way or the other for playoffs, if there are playoffs.
The disaster appears possession related. Turnovers, rebounds, fouls in some manner. Would need to look at another source beyond B-Ref. Maybe later.
There are other recent problems beyond this pair. In fact over last 15 games, the 4 most used are now all slightly negative as are 45% of the 20 most used. I could look further but this is probably enough for the moment.
All these things found and said in 45 minutes.
And yet, only 2 Embid lineups have been used over 1 minute per game.
In other news, the 2 best lineups for Sixers out of only 7 total used over 1 minute per game are non-Embid lineups.
Only 2 lineups used over 2 minutes per game and none over 4 minutes / gm.
Most of these things are shocking or just ridiculous depending on your expectations.
On the brighter side 5 of 6 most used lineups are positive, as are 19 of the 20 most used pairs... for the season.
Things might be considered ok or good. But they probably could be better if the well performing lineups and maybe even sub-lineups were played more.
The 3 best performing bigger minute pairs are all between Embid, Richardson and Scott. In raw terms they are twice as good as any other bigger minute pairs or nearly so. The trio is more than twice as good as 85% of the 20 most used trios and yet only plays about 5 minutes per game, which is less than 90% of these trios. More analysis should be done on this trio before making a major change... but has it been done? And was the decision not to elevate it? Why? By the coach and / or the analytic staff?
Embid - Horford is the worst performing pair in the most used 20 and slightly negative. So what to think / say about the massive contract signing decision, the analytic advice that preceded that, the coaching that followed and the players themselves' interaction and performance? Well, let's first check how it has done recently. Any improvement or, in lieu of that, minute reduction? Absolutely the opposite. Huge minute increase in last 15 games (to almost 2.5 times the usage level that preceded it) and a dramatic decline to about -9 pts per 100 possessions. This is a total debacle. Coach, analytics, management. I guess the pair was in fact about neutral before it went to hell but they did not stop it from going to hell or simply limit the minute exposure. Will they fix it? I dunno. They better fix it one way or the other for playoffs, if there are playoffs.
The disaster appears possession related. Turnovers, rebounds, fouls in some manner. Would need to look at another source beyond B-Ref. Maybe later.
There are other recent problems beyond this pair. In fact over last 15 games, the 4 most used are now all slightly negative as are 45% of the 20 most used. I could look further but this is probably enough for the moment.
All these things found and said in 45 minutes.
Re: 2019-20 lineup analysis
Someone else wrote about Embid - Horford here https://thesixersense.com/2020/02/21/ph ... -together/ and later.
And another writer earlier https://www.inquirer.com/sixers/sixers- ... 00104.html
But the Sixers let the problem explode in size & severity anyways.
And another writer earlier https://www.inquirer.com/sixers/sixers- ... 00104.html
But the Sixers let the problem explode in size & severity anyways.
Re: 2019-20 lineup analysis
Warriors have no idea how to use Paschall. Every pair in his 10 most used was very bad.
There was one bench lineup with Wiggins that did good in 40 minutes but that isn't much time or likely to be a future mainstay. It also depended on fouls and turnovers and may be a fluke.
The only pair in the Warriors 20 most used that is still possible and not significantly negative is Green - Lee and it was only neutral. Everything else- pretty much stop doing it. Or fix it with changes (Curry, Thompson, better strategy).
Most Wiggins pairs are negative, many badly. With Mulder and Green mildy. The one game with Curry was disastrous. A few deep bench pairs were ok in tiny minutes. But overall not encouraging.
There was one bench lineup with Wiggins that did good in 40 minutes but that isn't much time or likely to be a future mainstay. It also depended on fouls and turnovers and may be a fluke.
The only pair in the Warriors 20 most used that is still possible and not significantly negative is Green - Lee and it was only neutral. Everything else- pretty much stop doing it. Or fix it with changes (Curry, Thompson, better strategy).
Most Wiggins pairs are negative, many badly. With Mulder and Green mildy. The one game with Curry was disastrous. A few deep bench pairs were ok in tiny minutes. But overall not encouraging.
Re: 2019-20 lineup analysis
The Suns have of 5 of 6 most used lineups positive and 14 of top 20 positive... but they end up -1 pts / 100 possessions overall by losing by more on the other 413 dink lineups used. Half a victory negated.
No lineup used over 4 minutes per game. 5 over 1 min / gm. Should have concentrated far more.
No big minute pairs more than mildly negative, but Saric tends to be in the weaker / raw neutral ones.
Rubio Booker Oubre is good and Rubio with one without the other is good. But Booker Oubre without Rubio must be bad. Rubio Booker and /or Oubre with Bridges and Ayton or Baynes are very strong / best lineups.
Their rotation bench is generally ok. The deep bench performs very poorly.
Okobo's RPM estimate is -2. Not horrible but a bit less than you hope for. His "big minute" lineups go from terrible (the biggest minutes) all the way thru ok to great. But none even get 1 minute per game. Pick a couple. Just not the current top coaching choice. At pair level, the message appears to be play with starters but not Booker. Horrible with a number of subs. I guess they probably should go into Okobo year 3 but I'd be eager to trade if they could get someone half-decent back. Things are complicated by Jerome sucking out of the gate though.
In general, they probably won't make much roster change. Backup PG is most pressing but anywhere they can get better shooting (especially 3 PT shooting) or shot defense is worth pursuing. And the deep bench can / probably should be disposed of as needed or able.
Monty Williams did some things right but he still underperformed expected wins by four and they were below average against good teams Good teams in general and especially good on defense. And also very bad against average offenses. Will have to do some things better to improve. Lineup management, shot selection, matchup strategy, development, etc.
No lineup used over 4 minutes per game. 5 over 1 min / gm. Should have concentrated far more.
No big minute pairs more than mildly negative, but Saric tends to be in the weaker / raw neutral ones.
Rubio Booker Oubre is good and Rubio with one without the other is good. But Booker Oubre without Rubio must be bad. Rubio Booker and /or Oubre with Bridges and Ayton or Baynes are very strong / best lineups.
Their rotation bench is generally ok. The deep bench performs very poorly.
Okobo's RPM estimate is -2. Not horrible but a bit less than you hope for. His "big minute" lineups go from terrible (the biggest minutes) all the way thru ok to great. But none even get 1 minute per game. Pick a couple. Just not the current top coaching choice. At pair level, the message appears to be play with starters but not Booker. Horrible with a number of subs. I guess they probably should go into Okobo year 3 but I'd be eager to trade if they could get someone half-decent back. Things are complicated by Jerome sucking out of the gate though.
In general, they probably won't make much roster change. Backup PG is most pressing but anywhere they can get better shooting (especially 3 PT shooting) or shot defense is worth pursuing. And the deep bench can / probably should be disposed of as needed or able.
Monty Williams did some things right but he still underperformed expected wins by four and they were below average against good teams Good teams in general and especially good on defense. And also very bad against average offenses. Will have to do some things better to improve. Lineup management, shot selection, matchup strategy, development, etc.
Re: 2019-20 lineup analysis
Bucks two most lineups are great but then 8 of the next 18 are negative. Room to play the top 2 more. Reason to consider dropping some of the next tier. Hill in 5 of the negative lineups, 4 of which are very bad. Interestingly they include the lineup with 4 other starters and an all bench lineup.
Re: 2019-20 lineup analysis
15 of 20 most used Celtics lineups are positive. That is impressive.
Of the 5 negatives, 4 are bad to terrible. All 5 of these have Smart and 4 have Brown. Overall that pair is good, though on the lower end for the 20 most used lineups.
Of the 5 negatives, 4 are bad to terrible. All 5 of these have Smart and 4 have Brown. Overall that pair is good, though on the lower end for the 20 most used lineups.
Re: 2019-20 lineup analysis
5 of 7 most used Sabonis - Turner lineups are positive. Overall that isn't a problem. More detailed review against types & qualities of lineups would be appropriate though.
Re: 2019-20 lineup analysis
85% of the Lakers most used 20 lineups are positive and only one is bad. At pair level, almost all the big minute pairs are good or better. Romeo - KCP is the least strong but still barely positive.
Re: 2019-20 lineup analysis
Put aside 2 three PG lineups and the other top 6 lineups for Thunder are All negative. Every positive lineup in the 20 most used have 2 or 3 of the PGs. Will they keep all 3? If they don't, lineups with 2 will probably get less common.
Re: 2019-20 lineup analysis
Griffin - Drummond is gone but it was pretty good in general and very good to great in starting units the past 2 seasons that were not used near enough.
Team often throw away stuff that they can't replace, at least immediately.
Team often throw away stuff that they can't replace, at least immediately.