2016-17 Celtics

Home for all your discussion of basketball statistical analysis.
shadow
Posts: 252
Joined: Fri May 29, 2015 3:38 am

Re: 2016-17 Celtics

Post by shadow »

Yep, it included all regular season games plus all playoff games through 4/22.

Here are the playoff only results for games through 4/23. Teams are again sorted by their win percentage delta.

Code: Select all

TEAM  WP       Exp WP	Delta
CLE	100.0%	77.74%	22.26%
WAS	66.67%	47.75%	18.91%
MEM	50.00%	32.91%	17.09%
GSW	100.0%	85.46%	14.54%
TOR	50.00%	36.26%	13.74%
CHI	50.00%	43.70%	6.30%
UTA	50.00%	45.22%	4.78%
HOU	75.00%	73.48%	1.52%
OKC	25.00%	26.52%	-1.52%
LAC	50.00%	54.78%	-4.78%
BOS	50.00%	56.30%	-6.30%
MIL	50.00%	63.74%	-13.74%
POR	0.00%	14.54%	 -14.54%
SAS	50.00%	67.09%	-17.09%
ATL	33.33%	52.25%	-18.91%
IND	0.00%	22.26%	 -22.26%
Based on these numbers, these would be the expected series records so far, rounded to the nearest whole game:

CLE/IND: CLE up 3-1
WAS/ATL: ATL up 2-1
SAS/MEM: SAS up 3-1
GSW/POR: GSW up 3-0
TOR/MIL: MIL up 3-1
BOS/CHI: Tied 2-2
OKC/HOU: HOU up 3-1
LAC/UTA: Tied 2-2
Crow
Posts: 10565
Joined: Thu Apr 14, 2011 11:10 pm

Re: 2016-17 Celtics

Post by Crow »

Thanks for the info follow-up.
Crow
Posts: 10565
Joined: Thu Apr 14, 2011 11:10 pm

Re: 2016-17 Celtics

Post by Crow »

Historical footnote seen in an article today:

"Before Game 3, Stevens owned a 2-10 playoff record, the worst in NBA history for any coach with at least 10 decisions."

Surged to 28% win rate for playoff career now.
Rd11490
Posts: 121
Joined: Mon Sep 29, 2014 4:54 am

Re: 2016-17 Celtics

Post by Rd11490 »

2015: 0-4 as a lottery bound team that fell ass-backwards into an 8th seed by 1 game.
This record concerns me about as much as Brett Brown's coaching record concerns me. Looking into wins and losses without the broader context is worthless.
Crow
Posts: 10565
Joined: Thu Apr 14, 2011 11:10 pm

Re: 2016-17 Celtics

Post by Crow »

Context matters but everybody has a context. Last place was last place. Not the last word but a fact and very different than the main young genius coach narrative. His record will improve but when does it get to average and does it go above that? How soon, how far up or above? Time will slowly reveal. Til then it is mainly more super hype I guess... and some checks of evidence that might be different or way different to try to partial offset and balance against the hype that ignores the context of no playoff series wins yet or at least didn't wait for any of them before starting and ramping up to about as much given to any but maybe a few other coaches who earned it.
Rd11490
Posts: 121
Joined: Mon Sep 29, 2014 4:54 am

Re: 2016-17 Celtics

Post by Rd11490 »

I get what you mean, and I agree. I just see these types of stats pushed by talking heads and repeated by fans as if they represent some coaching or organizational failure.
That being said, your analysis in these threads about his erratic line construction has been very helpful to me (and others) in slowing our hype around Stevens and reconsidering how I view him as a coach.
Crow
Posts: 10565
Joined: Thu Apr 14, 2011 11:10 pm

Re: 2016-17 Celtics

Post by Crow »

I might be moderately positive about him if I didn't feel that almost everything said about him was ahead of where it should be so far positive. That is my main point. If there are Boston media outlets that are critical or overall critical of Stevens, let me know where they are. I haven't looked for them but I haven't heard or seen them casually either.

Thanks.
Crow
Posts: 10565
Joined: Thu Apr 14, 2011 11:10 pm

Re: 2016-17 Celtics

Post by Crow »

Celtics 5 lineups used 10 plus minutes, +30 combined. 62 dinkier lineups combine for -12. Concentration usually does better. Each got close to 50% of the minutes.
Crow
Posts: 10565
Joined: Thu Apr 14, 2011 11:10 pm

Re: 2016-17 Celtics

Post by Crow »

Strong win streak. We'll see what's next.
Crow
Posts: 10565
Joined: Thu Apr 14, 2011 11:10 pm

Re: 2016-17 Celtics

Post by Crow »

After a long win streak, Stevens & the Celtics only raised their % positive lineups from 30% after 2 games to 31% right now. Apparently the wins are coming from the results of a minority of lineups. I guess finding small-time or even freak positives is a strategy in contrast to playing your best more and more. Sustainable? I dunno. But I still am having a hard time understanding why you wouldn't try to convert your small-time or even freak positives into bigger ones thru heightened use. Experience (reality) is a pretty good guide I would think compared to the alternatives (following / guessing from a mental model / analysis).

BEST 20 lineups, +132. Other 80 used, -103. 237 minutes given to the best 20; 201 to the remaining 80. Looks real close to half I know what I am doing and half not really. Not bad, but not that good. Could someone else do better? We don't know.
Crow
Posts: 10565
Joined: Thu Apr 14, 2011 11:10 pm

Re: 2016-17 Celtics

Post by Crow »

Sometimes data without context doesn't make sense or as much sense it could, so here are % posiitive lineups for the other remaining teams:

Cavs 33%
Spurs 31%
Warriors 35%

Wizards 26%
Raptors 26%
Rockets 31%
Jazz 30%

So you can advance and barely break 30% positive. (Or maybe not, didn't check at series break. some might have been below 30% then.) A bit surprising to me. Do coaches know their % positive lineups? I'd think at least some would less too high. Probably way high.

Only Jazz, amongst teams still alive, have used more lineups than Celtics (109 to 100). Rockets have used less than half as many as the top 2 teams at just 45.

Probably too early to evaluate much but worth checking and thinking about.
Crow
Posts: 10565
Joined: Thu Apr 14, 2011 11:10 pm

Re: 2016-17 Celtics

Post by Crow »

Overall 1212 distinct lineups were used in playoffs to date by all teams. On average a bit less than 29% got cumulatively positive results.

NBA coaches aren't usually getting positive results with their choices. 71% of the time, they didn't gain or lostground with the new coaching lineup choice, based on their experience, judgment, whatever use of the numbers, etc. To gain they either use the positive lineups more or hit on a very productive one, because it was very good for the context or they got lucky or the other guy was not good or was unlucky.
Crow
Posts: 10565
Joined: Thu Apr 14, 2011 11:10 pm

Re: 2016-17 Celtics

Post by Crow »

Stevens has used his 104 lineups twice on average in 10 games. They are the "right choice", but only in 20% of games? His average lineup has been used 4.7 minutes for the whole playoffs or 2.4 per game.

For all the teams still alive, the Spurs and Pop "lead" with 4 lineups used over 20 minutes TOTAL in the playoffs or above a few minutes per game. Everybody else with 2-3. They are dinking and micro-dinking lineups all over the place. Only Cavs and Wizards have a single lineup used much over 10 minutes per game.
Crow
Posts: 10565
Joined: Thu Apr 14, 2011 11:10 pm

Re: 2016-17 Celtics

Post by Crow »

Celtics & Stevens make conference finals, a notable achievement. I think about 40% of current coaches have done that. Stevens' career playoff record moves to 10-11, also probably somewhat above the median. So at This Point, I am fine with calling him an above average accomplishment young coach. "Sorry" for my lag waiting on those accomplishments before assigning that label? Not really. But I do so now because that seems like a reasonable & fair assessment at this time.

Is he "great" or as Danny Ainge ventured, at the time of his HIRE I believe, "the best coach in the country"? I wouldn't go that far, yet. Those labels require still more performance achievements. For me.
Crow
Posts: 10565
Joined: Thu Apr 14, 2011 11:10 pm

Re: 2016-17 Celtics

Post by Crow »

Another week gone, Celtics % of cumulatively positive lineups creeps another percent to 32%. Cavs still at about 33%. Spurs surge to 35%. Warriors inch up to 36%. Cavs and Warriors using the least lineups. Spurs have an edge on Celtics for the perhaps dubious honor of most lineups used.
Post Reply