APBRmetrics

Hawks really playing the 8 man rotation.

Opponent shooting and rebounding completely offset strong offense.

Have 3 decent reboundersbut they only play about 70 minutes combined per game. Bogdanovic and Hunter are weak rebounders.

Quick check back at GM survey's weakest responses:

20% of GMs had Cade Cunningham as 2nd most likely to be the biggest breakout.

17% had Marcus Smart as most underrated player acquisition, mote than anyone else.

10%, tied for 3rd most, had Pistons as biggest breakout team.

Which rookie was the biggest steal at where he was selected in the Draft?

1. Cam Whitmore (20), Houston – 43%

Uh huh. 31st on BPM in draft class at -4.9. 40th on my board, so no disappointment for me.

They put Kerr and Popovich tied for 3rd for best Coach overall.

Tonight's starting lineup for Pistons worked very well in 27 minutes.

Previous use? 2 minutes for season. Too busy playing with 315 other lineups.

In each conference 12 teams have a chance to at least make play-in status. 2 will come up short of that. Will be major disappointment for those you don't even accomplish that low distinction.

Bottom 6 probably set already.

Nets lineup management:

https://x.com/bballstrategy/status/1743 ... 75781?s=20


Wizards flaws:

https://x.com/bballstrategy/status/1742 ... 23126?s=20

Defensive leaders by LeBron metric:
https://x.com/bballstrategy/status/1741 ... 78407?s=20

Pistons lineup management:

https://x.com/bballstrategy/status/1741 ... 63409?s=20

https://x.com/bballstrategy/status/1735 ... 29004?s=20

Somewhat dated league level lineup management:

https://x.com/bballstrategy/status/1735 ... 64601?s=20

Has this season had fewer than usual injuries to major players?
From this year's minutes leaders, I spotted these examples of stars and near-stars on pace to get at least 1/3 more minutes this season than in recent years.
Showing current minutes and extrapolated to 82 team games; avg minutes over previous 3 and 5 years.
x/yz is '24? divided by avg of other two -- prev3 get more weight than 4-5 years ago.
https://www.basketball-reference.com/le ... _stats::mp
Zion and Porter don't really have 5 years; so it's just x/y

With regard to another metric, maybe Kevin (nbacouchside) can give a current link to DRE? USAToday said they were going to use it during season but I haven't seen it beyond the pre-season projections.

Embid missed games, 10 of 12 on road. Haliburton 8 of 9 on road before last 4 at home. LeBron 4 of 5 on road.

I have this old-fashioned idea that chart or graph producers should usually do a full explanation, especially of anything not completely self-evident.

There seems to be becoming a more general habit for producers of NBA visualizations to say nothing or far too little, as if that spoils the pristineness of the visualization or the experience of it. Does it say something valuable? Say it clearly and fully for everyone, every level. Or at least the core finding. A common phrase on Twitter is "what stands out?" You should tell me and I'll tell you if I agree or have additional thoughts.

Top 20 seeds probably set. Just a modest chance for a change at the 10th seed level.

Pistons at 553 lineups. Grizzlies over 600.

400+ is ridiculous and usually with only a couple of lineups even over 1 minute per game for season but these two take it further.

Almost every team management, analytics and coaching group can be critiqued for allowing / accepting the inefficient chaos and almost entirely insignificant test data. Keep on micro-guessing in near darkness. Only the greater and more inefficient sloppiness of others will somewhat cover over the sloppiness and inefficiency of most of the better teams.

Hawks are probably still the low at 208 lineups used; but having 2 of 3 most used negative and only those 3 over 3 minutes per game limits how much credit can be given to them.

The 10 teams whose on-court age is at least 1 year below the league avg have an aggregate SRS of -4.0
The 10 oldest teams avg 3.15 SRS

OKC is tied with SAS for youngest team at 23.2 years. And their SRS is 7.26
Has a team anywhere near this young been anywhere near this good?

2010-11 conference finalist Thunder had a presumably minutes weighted average age of 23.7. 2011-12 Finalist Thunder went to 25.1 but both were well short of age of youngest to ever win title and didn't.

2014-5 Warriors were near 27. 2016-17 Celtics near 26. Warriors got it, Celtics didnt.

Consider history or ignore it and try to blast an unprecedented path. In case of Thunder, again and again and again and again in a new era.

Could argue that free agent market dynamics for small market team and this one in particular argues for this general approach but could try avg age of 27 or 26 or 25 instead of under 24. Given league history and specific team history.

Hayward move might be considered something in that direction but it probably won't change things much or enough. Exceptions don't overpower the normal practice.

When do they get to avg of 26, if they will? They got to 26 one season and 27 in another in first era but separated in time and never otherwise; so while not totally unprecedented, the preference is clearly average age below or well below lowest ever for a champion, history be damned / ignored.

Almost no way average age will reach 26 or 27 in next few years with current core and voluminous draft picks and coming cap structure forcing heavy use of first contracts. Maybe they get over 26 eventually. Before SGA is a free agent?

50 minutes qualified players with 25+% usage. Another 50 over 20%. That is out of less than 200 minutes qualified. But it means that about half of an average team's top 7 rotation gets this level of usage and half do not.

About 30% of each group is under 25 years old. So on average teams will have just 1 minutes qualified guy over 20% usage. Lots want that slot. Lots touted and drafted for that slot and wanting even more abd soon. Not the league reality. Higher usage is mainly for the mature. By skill, by stature, by player control, by coaching, by tradition.


For all players, it is about 75 over 25% usage and another 100 above 20%. So a bit less than 1/3rd of all players get 20+% usage. What % of consensus lottery / actual picks do? Full 1st round? Second round? Probably way more. Understandable to a point but worth remembering that many will see usage declines, early in NBA and forever. Affects the future value of players who are predominantly shooters and / or scorers. I'd guess that pre NBA assists and rebounds levels persist on average more strongly than usage and scoring; but if anyone has that research or wants to do it, please share.


Less than 50 minutes qualified with 20+% defensive rebounding rate. About 1/3rd under 25.

Less than 50 minutes qualified with 25+% assist rate. About 1/3rd under 25.

So teams on average have a bit more than 3 20+% minutes qualified usage guys, less than 2 high defensive rebounders and less than 2 high assistmakers. Amount of overlap varies by team. If there was no overlap, then each member of the average 7 man rotation would have one primary role. Only players meeting multiples of these criteria would allow (or force) other players to not meet one of the criteria without the team falling below the average profile.


In Tankathon mock draft, it is 2 currently over 25% usage (Holland and then Dillingham at barely 30%), 7 others over 20% and Sheppard the exceptions at 18%. It is likely that the rekative lack of real high usage players at the top is hurting the regard of this draft class. But should it or this much? Depends what you want and will actually get fron a draft on future usage and future other stuff.


For top 10 of 2023 draft class so far, it 1 guy over 25% usage (wembenyama at 32%), 2 more over 20%, 5 more minutes qualified under to way way under 20% usage and the final 2 not minutes qualified and well under 20% usage.

For top 10 of 2020 draft class a 4th player barely rises over 20% for career average but has dropped below this year. The top 3 picks and one marginal case So the quantity is either the same or barely 1 more than in the 2023 class. In both cases it is is clear majority do not get to 20+% usage early in career and to this level of brief analysis it did not materially increase with early experience.

Pick a high usage guy early and expect high usage? Better have a rotation slot and intention to fulfill than usage level. In most cases the actual experience will be less.


For 2023 top 10, 3 stepped into or up to a 20+% defensive rounding rate role. Just 1 is 25+% assist rate or even 20%. Few top draftees take lead roles immediately on any of these 3 criteria.

For 2020 top 10, 2 met def reb rate criteria. 1 on assist rate for career, q more barely in year 4. If teams want young guys to fulfill lead roles on these criteria early in careers is not happening much. Perhaps because draft selection criteria is emphasizing other things or versatility over elite specialization. Draft what you want / need / value. Don't think to fully get it early in most cases, especially if not giving "it" a high / primary weight in selection consideration.

It can take a long while for lead roles to get passed down. Not sure how many drafted players are still with same team in year 5 but it is probably less than teams originally imagined / planned. Imagine big roles if truly appropriate/ realistic. Probably more realistic to plan for support roles in mist cases and perhaps targeted to 1 or a few criteria.

Mikael Bridges and Cam Thomas with near identical stats across the board over last 10 games.