Archx wrote:Doctor MJ wrote:Archx wrote:
You're right, apparently Curry, Lawson, Lively, Exum, DJJ are Mavs top players. At least that's what On/Off tells us. Kyrie and Luka don't matter since they're far behind them.
There's something real here that you really need to acknowledge rather than trying to score points.
Forget about how we rate individuals, what's causing the surprising team results?
We've become accustomed in basketball to having our top players yield outlier On/Off & +/- regression based numbers. So when a player doesn't give us those signals after plenty of sample, we need to ask why not.
What points am i trying to score? I just gave that person an extreme example of high swings of On/Off numbers when it comes to this specific team. Not long ago Luka was at +10 now he's at +0.3. They're still 2pts better when he is actually playing. You could argue that they have a strong bench and those numbers on starters look bad but that is a lineup combination problem then. But you can't convince me that you just remove someone with +0.3 who's averaging 32/8/8 and the team will suddenly become better.
Or you have a better explanation?
To your first question, which I've bolded, your use of sarcasm in the prior post is clear cut rhetoric. Hence my response advocating for sincere analysis.
Re: gave example of extreme swings of On/Off. Right, but others have pointed out in this thread, and myriad other places, that a criticism of noise is one of small-sample size, but that the entire reason why this stuff is significant is because it's been going on for far longer than than that.
If you don't understand what I mean, ask for clarification, but do so humbly with the recognition that this is a well-known thing and has been for far longer than Doncic has been in the NBA.
Re: do I have a better explanation? Well first, you haven't offered an explanation. You explained why you're choosing to ignore +/- numbers, but you haven't explained why the +/- numbers don't match with your expectations given Doncic's profound box score productivity.
Second, the important thing here is not what my explanation is, but the recognition that there's something that has to be reconciled by doing something other than just ignoring the data that doesn't fit with what you already believe.
Now, what's my explanation for things?
I think the focus of Doncic's play style is essentially to win a given offensive possession against heavy defensive pressure. The theory of benefit I'd have for this approach would be one of resilience - not necessarily creating the most effective offensive attack over the course of a season, but more perhaps more unstoppable than any other approach against the very best competition.
In terms of how that looks over the course of a season, we should consider:
a) How good we can expect other approaches to work with typical NBA talent when Doncic is on the bench.
b) How good we can expect the offense to work during possessions where Doncic is on the floor but taking a breather.
c) How problematic Doncic's limited endurance becomes with high usage over the course of games/series/season.
d) How concerning Doncic's defense is in general.
Note that however we decide these things factor in, it's very much possible that Doncic has the capacity to become unstoppable enough to lead his team to playoff series victory after victory...and yet still not display signs of regular season impact like we normally get from top tier stars.
If this is true, as I've noted in many places, his playoff achievement will determine his legacy, and has the possibility to render a lacking regular season impact as a funny quirk rather than anything truly damning. It could then mean that Doncic will never deserve to be a regular season MVP candidate, but could conceivably still end up the greatest player in the history of the sport due to the greater importance of the playoffs.
But as I say all of this, I do need to address your use of the box score production here. We should be clear that every made basket on Doncic's tally - just like anyone else's tally - represents a success. However, if you take that basket, along with anything he assisted on, you're still not going to get anything that's going to make a team win in an of it's own.
If we (very generously) equate each assist to 3 points, and add those to his points scored, and tally that all up, then 32 & 8 represents 56 points. Tiny compared to what it takes to win a game, which means that a team can play in a way that enables a star to produce this and still lose a game easily.
Now, undoubtedly you and others will see this as absurd, because you're used to comparing the players on a team based on data like this, and seeing the guy getting the big numbers as producing more while his teammates produce less. With this thinking, if the team loses, it must be because no one else was producing enough.
In reality though, Doncic plays a style of play that hoards statistics and thus makes the other players on the team look less impressive from a classical box score perspective. You can't, for example, just how effective Doncic's teammates are as passers just by looking at their assists, because when Doncic is running the offense, Doncic is going to be the guy getting much of the assists. Doesn't mean Doncic is doing anything wrong, and doesn't mean Doncic isn't a great passer or anything else, but you'll have a tendency to assume that his teammates are less capable than they actually are if you use those stats to evaluate them.
Consider if we instead look at the Ast% when Doncic is On the court compared to Off.
'23-24: On - 55.9%, Off - 69.2%
'22-23: On - 54.3%, Off - 61.9%
'21-22: On - 57.6%, Off - 62.2%
See the trend of the team being more assist-oriented when Doncic isn't there? Very much a real thing. Doesn't mean the team is a better offense without Doncic, and doesn't mean that any of Doncic's teammates are better passers than Doncic, but it at the very least speaks to how the team offense has functioned differently when Doncic isn't there.
And yeah, were it to turn out that this led to an actual better offense, this wouldn't seem in any way impossible to me. Not because I'm anti-Doncic, but just because I recognize that basketball is a fluid field sport where different schemes can lead to unexpected results.
In practice, I don't think the Mavs's offense ever is better off without Doncic, but I do think that if what you take to be ground truth is the box score, I think the gap with and without him is smaller than what you expect.
For the record, I'd encourage you and anyone interested to look at how that compares with other big-assist stars. Doncic is not the only of them to show trends like this, but it's not the norm.
