Page 13 of 21

Re: 2019-2020 season commentary

Posted: Wed Nov 27, 2019 11:44 pm
by Crow
Hawks, awful offense, awful defense. -10 pts / 100 possessions with Trae Young for season and slight worse without. Getting worse. Outscored by 15 points per game over the last 10.

Hunter off to a bad start. Reddish terrible.

Several front court players have decent boxscore stats. But the defense is terrible. The perimeter generally sucks on both ends of the court.

Total mess. Years of development and / or massive changes needed.

Re: 2019-2020 season commentary

Posted: Thu Nov 28, 2019 12:05 am
by Crow
Lakers doing about as good as they could have imagined.

Much more apparent effort from LeBron.

Too much? Long time to playoffs.

Health will be key.

Re: 2019-2020 season commentary

Posted: Thu Nov 28, 2019 5:08 am
by Crow
Bucks are only very strongly successful road in East so far. Celtics, Heat and Raptor are good. In West, TWolves, Nuggets, Lakers, Mavs are strong. Rockets good. Could easily change. Important. Jazz with the weakest road record of the main contenders.

Re: 2019-2020 season commentary

Posted: Tue Dec 03, 2019 6:08 pm
by Crow
Towns with a good 1 yr LARAPM but 89th and not dominating. Almost all TWolves are tiny positive estimates on this. Either LARAPM IS too regressed to mean or this is a team with very broad sharing of credit.

Re: 2019-2020 season commentary

Posted: Tue Dec 03, 2019 6:09 pm
by Crow
Towns with a good 1 yr LARAPM but 89th and not dominating. Almost all TWolves are tiny positive estimates on this. Either LARAPM IS too regressed to mean or this is a team with very broad sharing of credit or coaching / lineup management that produces these results. 4 of 5 most used libeups are strong positives but nothing used even 75 minutes or less than 4 minutes per game. 234 lineups used. Probably 150-200 too many. Chaos that clouds assessment.

Re: 2019-2020 season commentary

Posted: Tue Dec 03, 2019 6:19 pm
by Crow
Daniel House proving to be a very important pickup / retention. Others missed.

Re: 2019-2020 season commentary

Posted: Tue Dec 03, 2019 6:23 pm
by Crow
Bucks with 6 of top 11 on 1 yr LADRAPM. Metric having some trouble separating credit. DiVincenzo #1 in league on this? Hmm...

Re: 2019-2020 season commentary

Posted: Tue Dec 03, 2019 6:29 pm
by Crow
TWolves pretty darn average across 4 factors, off. / def. efficiencies and W-L. Fast paced and pretty strongly money all but they shoot poorly. Playing a lot of guys to get near average on defense.

Re: 2019-2020 season commentary

Posted: Tue Dec 03, 2019 6:44 pm
by Crow
By SRS there are 6 really strong teams and 5 lurking a level or two below.

Re: 2019-2020 season commentary

Posted: Thu Dec 05, 2019 7:17 am
by Crow
There are some familiar names in top 7 on team efg% and some bad team surprises. Among contenders the Jazz are second lowest and Celtics the worst in that group.

There are 6 top teams in the 7 highest ranked by offensive rebounding rate. And 3 in bottom 6. The only top teams slightly further toward the middle are Sixers at 10th and Heat at 21st. Last season the contender split was mainly between middle and low with the Warriors being in the middle group and Raptors strandling middle & low. How much is intention vs. noise? I dunno but it looks like a shift.

On ft / fga, most top teams are high to middle. The weakest contender on this factor is the Lakers.

Contenders take all 8 of the best on opponent efg%. The weakest performances among this group on this factor are the Sixers and the Rockets.

On defensive rebounding about half the contenders are strong, some middling and a few weak. Celtics are 2nd weakest, Raptors awful at 29th.

On opponent ft/fga, most contenders are toward the middle. Bucks best in group, Heat worst.

Adjusted factors should be considered and will be when made easily available.

Re: 2019-2020 season commentary

Posted: Fri Dec 06, 2019 8:43 pm
by shadow
Here's an updated RAPM through games played on 12/5:

https://docs.google.com/spreadsheets/d/ ... ingle=true

Re: 2019-2020 season commentary

Posted: Fri Dec 06, 2019 9:10 pm
by Crow
Your top player value is almost twice as high as Ryan Davis' top estimate. I assume you have been different choices affecting the distributions of estimates. Possibly worth looking about, if inclined.

Re: 2019-2020 season commentary

Posted: Fri Dec 06, 2019 9:13 pm
by Crow
Your top player value is almost twice as high as Ryan Davis' top RAPM estimate. I assume you have been different choices affecting the distributions of estimates. Possibly worth looking about, if inclined. His top LARAPM estimate is almost as high as yours but that is not apples to apples.

His LARAPM pulls Siakam (3.25 on your RAPM and #2) down to +0.86. Tatum gets cut almost in half on LARAPM compared to your RAPM. Giannis rises about 10% on his to capture top spot. The estimates for Harden, Leonard , James and Giannis are pretty close and all very high.

The estimates for Doncic are fairly close and put him in about 60-70th place. The mainstream is not interested in that information though. C Wood about 50% higher on estimate impact rate than Doncic on yours, about 1/3rd lower on LARAPM. But in same basic range, fwiw.

Milsap 9th on yours, 33rd on RD's LARAPM. Gasol 21th on yours, 135th on his. I think I'll try to average them for now but it would be interesting to see how each metric is matching up to team level performances. different run dates are surely affecting things now but that impact should diminish as season progresses. RAPM to RAPM is another comparison but I wanted to see the impact of the luck adjustment or the lack of one.

Re: 2019-2020 season commentary

Posted: Fri Dec 06, 2019 11:55 pm
by Crow
Kevin Love lead the league his first two years on offensive rebounding. Then it started to steadily decline. This season he is 238th. How much is it a change in positioning? Still very strong on defensive rebounding rate.

Re: 2019-2020 season commentary

Posted: Fri Dec 06, 2019 11:57 pm
by shadow
I don't know when his was last updated (I think he initially posted it on 12/1. but not sure if it's been updated since then) or what he's using for his lambda value. By the end of last season the highest overall values in our respective versions were 6.36 (me) and 6.06 (rd) so chances are it's something that will basically resolve itself by the end of the season.