http://www.basketball-reference.com/pla ... rder_by=mp
2/3rds of lineups used over 200 minutes so far are positive. As we hit 50 games this really isn't a lot of usage. Still, barely one per team on average. 8 teams don't have a lineup used much over 5 minutes per game across the season.
Two worst are B Scott constructions featuring Kobe Bryant, both worse than -16 pts per 100 possessions. Bye guys.
The only lineup over 300 minutes better than plus 8 per possessions is the OKC traditional starting lineup at plus 20. Cle and GSW have lineup almost as effective but at just over 200 minutes. Usage level should change in playoffs. Should. Some may fail in part for not playing their best lineup enough. That would be a ridiculous way to go down.
Checking in on Celtics lineup usage (and other teams)
Re: Checking in on Celtics lineup usage (and other teams)
Every Lakers trio used over 165 minutes (not much more than 3 minutes per game) has been negative this season. Almost all bad to horrendously bad. There are 2 barely positive over 150 minutes.
Clarkson-Bryant-Hibbert in about 700 minutes have a bigger negative net points for season than for all other lineups not having all 3 of these on court together that were used in about 1700 combined minutes. -20 pts per 100 possessions. Tanking, ignorance or something else unfavorable.
Clarkson-Bryant-Hibbert in about 700 minutes have a bigger negative net points for season than for all other lineups not having all 3 of these on court together that were used in about 1700 combined minutes. -20 pts per 100 possessions. Tanking, ignorance or something else unfavorable.
Re: Checking in on Celtics lineup usage (and other teams)
Celtics starting lineup is a meh plus 2 per 100 possessions for season. Looking at net offensive minus defensive performance they are shooting and making way more 3s than the opponent. A modern feel. By contrast the Olynyk - Sullinger lineup is very low on 3s. But killing em with 2s and / or defense against opponent 2s. Is Stevens choosing style? The second most used lineup with Smart-Bradley-Turner-Jerebko-Olynyk is killing them net with 2s and 3s. None of the top 3 most used lineups is much more than neutral from the free throw line. Olynyk- Sullinger is actually quite low at ftas too, another style offense compare to the Morey ball ideal; but, near plus 20 pts per 100 possessions is good however you grind it out. It has produced twice the net pts as the starters in one fifth the time. That's style I appreciate. Smart-Bradley-Turner-Jerebko-Olynyk has almost produced as much net pts edge as the other two combined. It is real similar to Olynyk-Sullinger per possession though. Just getting more minutes.
Which lineup plays most / least in playoffs? I know what I expect and what I'd do differently.
Which lineup plays most / least in playoffs? I know what I expect and what I'd do differently.
Re: Checking in on Celtics lineup usage (and other teams)
Is an offensive attack that emphasizes the 3pt game the right choice against starting defenses- on average? Is it the right choice for the Celtics? What does the data say at league level and team level? Of course play calling and what the players do or try to do matters too within lineup choices. Do you have more ways to win against mostly bench defenses? Is the 2pt game especially reading against bench defenses? A few Celtics lineups are an entry point to the questions. Go after understanding of specific and global matchup trends.
Re: Checking in on Celtics lineup usage (and other teams)
Over 11,000 lineups have been put on court this season. Just 36 have played in more than 25 games or just over half the average games played. So 99.7% of all lineups got used less. Deemed necessary some place then ignored. In most cases they probably weren't necessary, the best choice or even good. Only 4.4% of lineups got used in 20% or more of the games. Only 14.4% in more than 10% of the games. Lots and lots of pretty random looking dink lineups out there.
Re: Checking in on Celtics lineup usage (and other teams)
Horrendous -11 showing for Celtics starting lineup last night pulls it below plus 1 per 100 for season. Olynyk got in groove, only to pulled back to playing with new, lesser lineups. Zeller showcasing?
Stick with it or go away from it briefly and then come back to it again despite the meh performance? One of them or both probably.
Stick with it or go away from it briefly and then come back to it again despite the meh performance? One of them or both probably.
Re: Checking in on Celtics lineup usage (and other teams)
Celtics starting lineup negative recently, down to plus 0.1 points per 100 possessions for season. What a choice. And sticking to it?
Re: Checking in on Celtics lineup usage (and other teams)
KO is out for a couple of weeks. With out being able to use him, what do you suggest for a starting lineup?
Re: Checking in on Celtics lineup usage (and other teams)
Yes, Olynyk being out takes away the most tested great option for now, but they still have plenty of options better than 0.1, if inclined. Starters with Sullinger (at plus 7) would be the most obvious next option. With Turner is small but with great results (plus 19 per 100 possessions). With Mickey might be worth consideration and a look, since Johnson (and Sullinger) are not long term certainties. With Zeller is unlikely, but it is plus 37 in 19 minutes.
Missing Olynyk, there is a decent rationale perhaps for not voluntarily introducing more change right now. My previous comment was more update than call for immediate change. I don't care that much who he rolls out first the next few weeks. The real issue will be who he rolls out in playoffs, how they do then and if / when / how he changes if the starters are meh or do badly against a good team as was their habit earlier in the year.
Missing Olynyk, there is a decent rationale perhaps for not voluntarily introducing more change right now. My previous comment was more update than call for immediate change. I don't care that much who he rolls out first the next few weeks. The real issue will be who he rolls out in playoffs, how they do then and if / when / how he changes if the starters are meh or do badly against a good team as was their habit earlier in the year.
Re: Checking in on Celtics lineup usage (and other teams)
Personally I'd like to see him roll out Micky and see what he can do, but at this point i'm just hoping that all these dink lineups are just to get everyone playing time before he tightens the lineups in the playoffs (no way he plays more than 7 or 8 right?).
Do you scrape lineup data from stats.nba or do you scrape play by play data and generate your own lineup data?
Do you scrape lineup data from stats.nba or do you scrape play by play data and generate your own lineup data?
Re: Checking in on Celtics lineup usage (and other teams)
He played ten guys for an average of at least 10 minutes per game last playoffs. I assume he will use at least 9 that much this time. Not sure which guy gets 10th spot but I think it is likely a 10th will meet or be close to this level of use. Last time 96 different lineups in 192 minutes. This time?
I mostly use the lineup data that basketball-reference offers.
I mostly use the lineup data that basketball-reference offers.
Re: Checking in on Celtics lineup usage (and other teams)
New, by far leader on net rating among lineups with 200 plus minutes. Toronto lineup with Lowry and 4 bench guys at plus 31.8 per 100 possessions, besting OKC's starters. http://bkref.com/tiny/7QYaV
Their mainline starting unit is slightly negative. Some coaches seemly don't especially want an immediate lead. Psychologically this might pay off in more focus, effort. It is a theory anyways. Coach upshifts into a better lineup... and makes a case for being a positive force. Seen it in past with Georg Karl, Casey's former boss and others. Celtics starters in 26th place here, still barely positive.
Their mainline starting unit is slightly negative. Some coaches seemly don't especially want an immediate lead. Psychologically this might pay off in more focus, effort. It is a theory anyways. Coach upshifts into a better lineup... and makes a case for being a positive force. Seen it in past with Georg Karl, Casey's former boss and others. Celtics starters in 26th place here, still barely positive.
Re: Checking in on Celtics lineup usage (and other teams)
Celtics likely to end up using 30plus % fewer total regular season lineups than last season. But 85% of lineups still have gotten 10 minutes or less use for season as of a few days ago.
What will Stevens do in the playoffs? As scattered as last time or more concentrated?
In early December: "Net plus minus of best 5 performing lineups is about plus 90, or 90% of total team net points for season. Best 20 lineups are about plus 200, while the remaining 229 lineups are about minus 100. Dinking is on average a losing proposition." Today the best 5 lineups represent only about 70% of total net points for season and the rest of lineups after the top 20 have had a milder negative impact. So Stevens has gotten a bit better at using lineups 6-20 and a bit better at the dink lineups. But the top 5 lineups are contributing less percentage-wise and the after top 20 dink lineups are still hurting overall, not helping. The latter is troublesome and the former might be worrisome as well.
What will Stevens do in the playoffs? As scattered as last time or more concentrated?
In early December: "Net plus minus of best 5 performing lineups is about plus 90, or 90% of total team net points for season. Best 20 lineups are about plus 200, while the remaining 229 lineups are about minus 100. Dinking is on average a losing proposition." Today the best 5 lineups represent only about 70% of total net points for season and the rest of lineups after the top 20 have had a milder negative impact. So Stevens has gotten a bit better at using lineups 6-20 and a bit better at the dink lineups. But the top 5 lineups are contributing less percentage-wise and the after top 20 dink lineups are still hurting overall, not helping. The latter is troublesome and the former might be worrisome as well.
Re: Checking in on Celtics lineup usage (and other teams)
Brett Brown's lineup management:
1 lineup over 100 minutes for season. Just 6 over 50 minutes. So no lineup used over 2 minutes per game for season and just 3 over 1 minute. Really studying lineups and allowing players to get to know how to use them. Not. Just 3 lineups used in more than 20 games. 36 positive, 699 non-positive; so less than 5% of these lineup choices were positive. Way to tank and not learn much. -5 on actual - expected wins. The only team worse than the Celtics on that.
1 lineup over 100 minutes for season. Just 6 over 50 minutes. So no lineup used over 2 minutes per game for season and just 3 over 1 minute. Really studying lineups and allowing players to get to know how to use them. Not. Just 3 lineups used in more than 20 games. 36 positive, 699 non-positive; so less than 5% of these lineup choices were positive. Way to tank and not learn much. -5 on actual - expected wins. The only team worse than the Celtics on that.
Re: Checking in on Celtics lineup usage (and other teams)
Brett Brown's 76ers comparison of this season vs. last:
Last season in bottom 20% on 4 factors out of 8 (offense & defense). This season, 6. A worse SRS. A win% not that much more than half of last season's debacle. He retains the lowest winning % in NBA history for coaches who got more than 82 games, including all those coaches of bottom tier, low budget teams before the maximum and minimum team salary rules. But great job on the happy face, we're working hard tank, have a 2 yr extension.
Last season in bottom 20% on 4 factors out of 8 (offense & defense). This season, 6. A worse SRS. A win% not that much more than half of last season's debacle. He retains the lowest winning % in NBA history for coaches who got more than 82 games, including all those coaches of bottom tier, low budget teams before the maximum and minimum team salary rules. But great job on the happy face, we're working hard tank, have a 2 yr extension.