APBRmetrics

Coaching RAPM estimates Coach D adds nothing on offense beyond the players and deserves a modest share on defense.

It rates Donovan 0.8 higher.

Thunder most used 6 lineups average about +20pts / 48 minutes. Overall average is about +12.8. Average for the other 612 lineups is +11.2 or about 45% lower than the 6 most used.

Celtics 6 most used average a bit more than half as much as the Thunder's but play almost 30% more. The rest of lineups net margin drops about 30% lower than the top 6.

Whoever makes finals, it will matter a lot how much 6 most used lineups play.

Thunder 5 most used lineups could be used up to 40 minutes per game if SGA and JDub played 40 minutes. Dort could but does not have to. JDub doesn't have to either, if it is an issue. The other 8 minutes are easy to fill. 6 lineups.

Vary from this if there is good reason but I'd vary to little as necessary or as leads allow.

Anybody in Thunder have this view? Express it to Coach D? Push hard for it? If not right away, in later rounds, if series is tight or behind? I would. That would be part of my analytic value equation. An important part.

The more the Thunder play top lineups, the higher the expected edge. Compared to Celtics, dink vs. dink only offers half the edge.

How much each lineup should be used and when & against whom would be topics for further study. Perfomance in specific matchups would matter some, against similars some and overall some.

How well the top 6 perform against opponent top 6s vs against dinks would matter. And starters vs. starters. Performance of specific dinks against the different classes of opponent lineups would matter too. On and on analysis. How much do teams typically do? How much do Coaches listen and follow? Probably not enough or to the max.

Should Thunder try to emphasize top 6 lineups against dinks above all else and the plan the rotation for that? Maybe. Do the analytics then apply as apparently appropriate by the numbers and judgment.

The starting lineup with Holmgren and Hartenstein does several notable things:

It is really low on 3pt attempts and makes and slightly loses net 3pt fg%. It does win net efg% moderately

It loses net fta moderately.

It barely wins net turnovers in contrast to the overall massive lead on that.

It wins both net rebounding match-ups in contrast to most but not all other lineups.

Overall it great. But it has some issues and is quite different than team norm in important ways.

In first round, SGA's TS% was only 52% and Dort and Wallace were around 45%. Dort and Holmgren with virtually no assists. Hartenstein with 21.5% assist rate but a 20% turnover rate probably makes this a net below average feature.

Dort with other playoff career highs and lows but I'll wait til later to assess.

By season averages, Celtics and Cavs are both far better suited to hold the Thunder's net turnover edge down than league average. Head to head Celtics did, Cavs did not.

Timberwolves, Lakers and Clippers did not. Warriors, Rockets and Nuggets did better.

Huge rebounding losses to Cavs and Clippers. Pretty big against Rockets and Timberwolves. Do not win total rebounding against any of these.

Thunder 3pt shooting was horrible against Warriors. Meh against Nuggets and Cavs. Random variance and possibly signal. Efg% terrible against Warriors. Mostly near league average or low. Not strong against any of these but slightly above average against Celtics and Rockets.

70% of Thunder playoff lineups were used in just 1 of 4 games.

2 years left on SGA's present deal.

How many total current players will be on team in 2027-28? Could be 10+, might be around 5.

Will Hartenstein fulfill contract in OKC? There is reasonable doubt.

Dieng seems likely to be a goner unless accepts a modest extension. At some point it probably becomes him or K Williams.

Doubts as well that Wallace, Joe and Wiggins serve out full contracts.

Topic, TBD.

Flagler, Ducas probably not or not for long.

Jones probably stays at least a year but iffy beyond that.

Carlson, depends on his options.

Mitchell and Jaylin Williams have a pretty good chance to stay but it depends on other money decisions and max budget.

Thunder would be helped by winning big the non-Jokic time and / or time with Jokic and 2 of Porter, Watson and Westbrook.

Someday Thunder might be Topic Joe (or other) J Williams Holmgren... and somebody.

Somebody is likely SGA for awhile but maybe not be a few years from now.

J Williams - Holmgren was great this season. After that is certainly open to different finishes.

Holmgren 2 limited hot and 2 extended cold shooting games against Nuggets in regular season. 3 for 15 from 3.
Something to watch.

How many attempts will he get / take? Make?


Dort strong from 3 against them, including one huge flurry.

Caruso cold from 3 against them. Same for Wallace.

Hartenstein with an awful densive rating against them. Strong offense though.

Nuggets 5 most used lineups are out there about 34 minutes per game in playoffs.

Thunder about 25.

Different series. But what will happen next?

Nuggets use 11 lineups. Top 2 were great for 31 minutes. Bad beyond that.

Thunder use 18 lineups. Starting unit sucked. Second lineup undid the damage. 18 combined minutes. Bench was winning until the Nuggets starters & Westbrook rallied.

Thunder won net turnovers but lost on boards and FT line. Efg% was a tie. Pretty much as expected by full regular season factor data.

Efg% probably will determine winner.

Pace in Game 1 was to Thunder's style.


Nuggets with worst defensive efficiency left in playoffs for all games.

Thunder efficiencies perhaps inflated by weak 1st round opponent.

"Peak" can be now and / or next year. Sure looks like it should be now. 2+ years from now might be very good or great too but might be harder. Maybe not, maybe just different.

Time will eventually tell.

Total playoff lineups is up to 113. 7 with 20+ minutes: 2 strong, 1 good, 2 meh, 2 quite weak. 83 for 5 minutes or less.

Against Timberwolves, it 55 lineups so far. 8 most used are 2 strong, 2 meh, 4 very weak. Only the starters have been used in each game. An average of 18 new for series lineups in each game. 45 used for 5 minutes or less.