Yeah, I think it is as nonsensical a model as you can get as it's based, for the most part, on ESPN's garbage RPM stat
Vehemence, or just rhetoric?
No, not garbage. Can't tell what your objection to it is - perhaps valid, but it hardly seems likely that it would justify dismissing the stat entirely, as you seem to be doing.
Its big virtue is that it controls for who else is on the floor - that's a big step up from ortg and drtg, and from the raw +/- that you get in the box scores nowadays.
My own objection to it is that using box score priors biases it in favor of players who get more of the countable stuff in the box score.
Like any statistic, it isn't the whole truth about a player - what you might call the Siren Song of the Single Statistic.
What I like about it most is that for certain players it suggests surprising or provocative things. It doesn't like Klay's defense very much, for example, despite conventional wisdom. Then you can ask yourself, well why not? And look for yourself.
The big problem with all the different types of statistics, including the various flavors of plus/minus, is that the sample sizes are seldom big enough, while the sample sizes for the different players vary so widely.
(which could be Pelton's stat, IDK),
It's not.
which I always have hated, even more so because they refuse to give the formula for the stat.
Yes, what's up with that.
Having said that, I don't think that I could evaluate it anyway.
I do trust that the plus/minus module of it, controlling for who else is on the court, has been done accurately.
I don't know how to evaluate how much weight they've given to box score priors, even if I knew what it was or how they've translated them.
I also wish that they'd make their web page sortable in several dimensions, like by team, and let you search by individual player.
Anyway, I find it to be a very useful stat that is looking at the big picture; that's absolutely essential, and I don't know a better place to get what it gives you.
RPM is a stat that projects future performance. An entire part of the stat estimates an impact a player has on his individual team's past net +/- to predict what the player will do going forward.
Year in and year out a look at the RPM standings shows some of the goofiest leaderboards in stats. Danny Young is apparently one of the best RPM players last year, 13th in the league, better than Siakim and Leonard, much better overall and impactful players that was on Green's team. In 2019 Chris Paul had the highest RPM in the league and Robert Covington was 8th. In 2017 Jae Crowder and Amir Johnson had the 20th and 21st highest RPM in the league. That was IT's unreal year where he was almost MVP level. He ended up 59th in RPM that year.
"Danny Young" - I think that you mean Danny Green?
First, you've made a couple of mistakes:
In 2019, it was Paul George who had the highest RPM in the league, not Chris Paul. Similar names?
In 2019, Covington was 8th among small forwards, not overall.
We don't need to say much about IT's defense; as I think you surely noticed, it sucked. If you just look at RPM for offense for him, he was indeed, as you say, "almost MVP level", at 6th overall. But RPM is not measuring just the offensive contribution, nor is it being used to pick MVP candidates.
It doesn't look worth it to do a whole big thing here. I'll just return briefly to a point I was making earlier: it's more useful to ask why the stat is giving you what it's giving you than to just take the conventional wisdom and see if the stat fits it.
You can't win without big-time shot creators, and those are the players that are commonly thought of as the best players - AND you can't win without them. But if you've got them (like Toronto did last season), then the complementary players like Danny Green don't have to do what they don't do well.
It happened that Green was entirely healthy as well, and was playing D like he used to a couple of seasons ago. More than that, though: the gravity that Green had (he shot a tremendous .455 from 3 on a big sample size) helped create that phenomenal ball movement they had last year, and the single coverage and spacing that Leonard and Lowry and Siakam enjoyed (and Norman Powell and Fred, too...).
So while Green didn't have to create shots, and in fact it was better if he wasn't creating shots, all the same he was key to them creating shots, and their effectiveness as a team in doing that bumps up his RPM.
Enough. I don't need to spend more time on this one.