Blazers best team in NBA?

Welcome to our community

Be a part of something great, join today!

Register Log in
Rating Systems.
To my knowledge, there are basically two kinds on NBA player-rating approaches. 1. Box Score. 2. Adjusted Plus Minus. Box Score ratings are low variance but do not capture everything. Offensively, box scores capture most things and so those ratings are quite good. Defensively, box scores miss too much so you can't trust them to overturn what you think you see. In contrast, Adjusted Plus Minus captures almost everything but unfortunately is high variance. Multiple seasons are needed to begin to stabilize.

Raptor from FiveThirtyEight
Some ratings systems like Raptor combine both approaches. They still suffer the same problems but to a lesser degree. These systems also have different ways of combining box scores with plus minus and thus get different answers from each other. So even those have to be taken with a grain of salt.

Nurk vs KAT - Raptor
KAT - Raptor 538.PNG Nurk - Raptor 538.PNG
 
wizenheimer said:
well, let's see...

* KAT's defensive rating is 2.1 points better than his team this season, and he's 3rd in DBPM
* last season, his rating was 2.2 points better than his team and he was 4th in DBPM
* the year before that, his rating was 4.9 points better than his team, and again was 4th on the team in DBPM
* the year before the year before, his rating was 4.1 points better than his team and he was 2nd on the team in DBPM

defensive stats are noisy, so those don't necessarily mean he's not bad at defense. But, they do indicate that calling him "historically" bad is probably over-stating the badness by quite a bit. He's likely better than Kanter

and, I kind of question how much Nurkic's defense actually means for Portland. I'm not saying it's not a factor, but just how big a factor is debatable in my view. I saw this yesterday in Portland's lineup page at bbref:

View attachment 38460

that would suggest that the step up defensively from Kanter to Nurkic, and from Jones to Powell, is only worth o.5 points. And we know Powell is a massive upgrade offensively from Jones, so maybe, there was no real positive impact from the offset of Nurk's defense over Kanter's offense. And, the minutes are nearly identical so there is no sample size skew

now, if you expand that list a little:

View attachment 38461

things get a little more complicated. To start with, the 3rd ranked lineup in minutes suggests that the addition of Powell's offense, not Nurkic's defense, was worth 8.2 points. Enough of a gap to tend to confirm that Powell over Jones has a lot more to do with differentials than Nurkic over Kanter

the 4th ranked lineup there seems to contradict a lot of the other assumptions. But we're getting into much smaller sample sizes so skew and noise are almost certain. It is kind of funny that the only change in #2 to #4 is CJ for Trent and that made Portland -25.0 points worse. I think CJ is overrated around here but he's not that bad

maybe the biggest factor for #4 is that is the period of time when CJ came back, before Nurkic, and Dame was playing about the worst BB of his career

so, a hypothetical upgrade from Nurkic to Kanter may not have the negative impact you imagine. It won't happen obviously and Portland would have to pay a much heavier price than just Nurkic or CJ
Click to expand...

I think the sample sizes are just too small to draw conclusions from the 5-man lineups and we don't know how strong were the opposing lineups. For example, lineup #2 vs #4 the only difference is McCollum for Trent. It would suggest that Trent is 25 points better than CJ. We know that's not true.

Edit: Oh I missed that you caught that 25 point swing already!! So given that it makes it impossible to say from these lineups whether Nurks D is worth a “mere” 5 points or not.
 
Last edited:
bobf said:
I think the sample sizes are just too small to draw conclusions from the 5-man lineups and we don't know how strong were the opposing lineups. For example, lineup #2 vs #4 the only difference is McCollum for Trent. It would suggest that Trent is 25 points better than CJ. We know that's not true.
Click to expand...

yeah, I talked about that in my post

I was mostly responding to all the hype about the current starting lineup and how good their differential has been. The previous starting lineup of Dame-Trent-Jones-RoCo-Kanter had a differential almost as good as the current one, and the sample sizes are the same

I'd tend to agree that even then, the sample sizes are small enough to potentially generate plenty of noise. Blazers are on a hot streak and that's going to skew numbers, just like a losing streak will

Bottom line is, on defense, Portland hasn't shifted it's numbers much. They are still the team that will finish 29th in the league in defense. And even in the best case scenario their defense was still not in the top half of the league. Their offense is potent though, just as long as Dame is playing really well
 
wizenheimer said:
yeah, I talked about that in my post

I was mostly responding to all the hype about the current starting lineup and how good their differential has been. The previous starting lineup of Dame-Trent-Jones-RoCo-Kanter had a differential almost as good as the current one, and the sample sizes are the same

I'd tend to agree that even then, the sample sizes are small enough to potentially generate plenty of noise. Blazers are on a hot streak and that's going to skew numbers, just like a losing streak will

Bottom line is, on defense, Portland hasn't shifted it's numbers much. They are still the team that will finish 29th in the league in defense. And even in the best case scenario their defense was still not in the top half of the league. Their offense is potent though, just as long as Dame is playing really well
Click to expand...

We are still 29'th overall, but in the 21 games with Nurk post-injury we are 6.5 points better on D and 11th in the NBA.
 
We are now the hottest team in the NBA. We have the longest win streak and we are the only team at 9-1 in the last 10. We remain the 2nd best team on the road.
 
By the way don’t look now but Suns might fall to 3rd. #2/#7 might be Clippers/Lakers.
 
Yes, the title is misleading, but in my mind, the original stats point to an important conclusion:

If our bench can play equal to our opponents bench, we have a VERY good chance to win.

That was the case in the first playoff game against the Nuggets. Bench positive +/-, and a Blazer win.
 
You must log in or register to reply here.

Users who are viewing this thread

Total: 1 (members: 0, guests: 1)
Back
Top