f4p wrote:lessthanjake wrote:f4p wrote:Steph Curry
Kevin Durant
Giannis Antetokounmpo
Dame Lillard
Steve Nash
John Stockton
actually the fact so many of the high efficiency guys of today are fallers makes me think the hyper focus on efficiency has probably made it more likely for guys to decline as guys in the past were more willing to take (or not smart enough at avoid) difficult shots and then the playoffs weren't so different. now when you try to filter out all the bad shots in the regular season but inevitably have to take some in the playoffs, you get a natural decline.
I could argue (and have argued) about other names you list here, but Steve Nash is just *definitely* out of place. He led some of the greatest playoff offenses in NBA history, while if anything having his box numbers go up (particularly with the Suns, where his scoring ticked up a good bit in the playoffs).
but they don't. i feel like we've had this conversation before but maybe it was someone else. look at my post at the top of this thread. nash is barely ahead of harden way down at the bottom of the list. that's box score drop off that is being measured.
nash's 10 year playoff prime (2001-10) shows:
PER -7.0% (21.5 -> 20.0)
TS% -2.6% (61.3 -> 58.7)
WS48 -26% (0.185 -> 0.137)
BPM -15.4% (3.9 -> 3.3)
harden's prime (2011-2021) shows:
PER -9.8% (25.5 -> 23.0)
TS% -2.5% (61.4 -> 58.9)
WS48 -21% ( 0.232 -> 0.183)
BPM -6.8% (7.4 -> 6.9)
doesn't get much closer, with harden probably a little ahead, especially considering the raw numbers are on another level from nash. and it all comes with a side of:
I don’t think we’ve ever discussed this before.
Please note that, as it relates to the box score stuff, I said “particularly with the Suns,” since that’s where his playoff PPG went up a good bit. I think if you wanted to call Mavs Nash a playoff faller, at least at the box level, then that’d be a more palatable claim.
For the Suns, though, Nash’s PER and BPM are essentially identical for RS and playoffs, so your assertion doesn’t really hold at all. As for TS%, if we look at relative TS% (compared to the actual opponent—an important thing to do here, especially with the Suns playing against some incredible defenses), we find that through to 2010, Nash’s rTS% was +8.9 in the regular season for the Suns and +8.2 in the playoffs. So it was slightly lower, but essentially the same. Finally, you mention win shares, but win shares is a pretty bad stat to measure individual playoff rising or falling with. It’s a stat that ends up distributing win shares to the team in a way that pretty closely tracks the team’s actual number of wins (for various reasons, including that it is tied to defensive rating). So if a player plays equally well in regular season and playoffs, but his team wins a much higher percent of its regular season games than it does in the playoffs, you’d expect his win shares per 48 minutes to go down in the playoffs. That wouldn’t necessarily mean he fell in the playoffs individually—it’d just mean that his team didn’t win as much. And of course Nash’s team winning less in the playoffs than in the regular season is not at all surprising when they’re facing playoff teams (and doing so in a stacked conference, where the average SRS of Nash’s playoff opponents on the Suns was almost 5). In context with his BPM and PER staying essentially the same on the Suns in the playoffs, I think that fall in win shares per 48 mins is basically just capturing the Suns facing really good opponents that they won less against. So yeah, on the Suns, I don’t think there’s a particularly good argument that Nash meaningfully fell in box score numbers.
And that’s before we even get into the fact that these stats are all rate stats (i.e. per-minute or per-possessions) and Nash was a player who increased his MPG quite a lot in the playoffs. For instance, for the Suns through to 2010, Nash averaged 34.3 MPG in the RS and 38.0 MPG in the playoffs. To take your Harden example in the years your specified, Harden played 35.5 MPG in the RS and 36.1 MPG in the playoffs (and the numbers would be pretty similarly close if we took other timespans instead). What Nash did in the playoffs for the Suns was basically put up essentially identical rate stats, while increasing his minutes per game by over 10%. That doesn’t look like a playoff faller, and indeed looks like a bit of a playoff riser, all things considered.
And, of course, it’s also the case that box stats are really not a remotely good gauge of Nash. Indeed, the description of BPM specifically calls this out, saying “Steve Nash was ridiculous on offense, and no, the box score still can’t fully capture that fact.” Box stats aren’t a good gauge of what he does in the regular season, so I don’t really know why we’d say that any changes in those numbers mean a whole lot about his relative performance in the playoffs. If a player is leading the best playoff offenses ever, and those offenses are even better than his GOAT-level regular season offenses, then he’s probably not a playoff faller even if we think his own personal box stats look a bit lower.
On/Off
Nash -2.7 (+7.3 -> +4.6)
Harde +6.4 (+5.0 -> +11.4)
RAPM (Cheema)
Nash -0.9 (3.1 -> 2.2)
Harden +1.0 (3.1 -> 4.1)
so not only consistent box score underperformance in line with harden, but where harden jumps way up in impact, nash jumps way down. they both start at 3.1 in RAPM in the regular season and then harden almost doubles nash in the postseason. and more than doubles him in on/off. i'm just not seeing this individual rising from nash. it's drop after drop.
Playoff on-off and playoff RAPM are just essentially completely garbage stats unless the player’s playoff sample is abnormally big (which Nash’s is not). Those stats are super dependent on the “off” sample, and the “off” sample is incredibly small (for Nash it’s like a grand total of 1500 minutes, with almost 30% of that being in the late 1990s when he was a bench player), such that the whole thing is basically just noise. And that’s not even mentioning that the adjustments that postseason RAPM does would be based on garbage low-sample data for the other players being adjusted for. Nor is it mentioning that the “off” sample in the playoffs is even more random than it is in the regular season, because stars playing more minutes per game means that a higher percent of the “off” sample is garbage time. These are just not something to base any meaningful point on.
Just for reference, I once ran the numbers on what Nash’s teams rORTG was with Nash on the floor (measured relative to each specific opponent’s regular season DRTG that year). And, with Nash on the floor in the playoffs from 2001-2010 (i.e. all his playoffs with the Mavs and in his main stint with the Suns), Nash’s teams had a +10.65 rORTG with Nash on the floor. And, just with the Suns (where he had more offensive primacy, of course), that number was an astonishing +12.58 rORTG.
and as Colbinii points out, we can't just pretend nash's teams being so great on offense is independent of them being so terrible on defense. his 2 craziest offensive series are the 2005 WCF and 2010 WCF, and they both come with a side of being some of the worst defensive underperformances. "offense is nash's purview" tends to get the defense handwaved away to others, but it can't just be a coincidence. either nash is so terrible himself that he drags his teams down, or his teams lean so hard into offense that they forget they have to actually put some defensive personnel on the court to guard someone. either way, the fact that what you say is far and away the best postseason rORtg didn't lead to any titles speaks to how bad the defense had to be.
Yes, Nash’s team being great on offense isn’t independent of them being bad on defense. They played offensively-slanted lineups. If that weren’t the case, then we’d have to give Nash GOAT consideration for how great his offenses were. But these were still incredibly impressive playoff rORTG numbers, and, quite pertinently for this thread, they were *better* than the regular season rORTG numbers (which were themselves incredibly high). Basically, even if you want to downplay Nash’s rORTG numbers based on team construction, the fact is that the rORTG numbers went up in the playoffs, which is certainly not suggestive of a playoff faller. And I am highly skeptical that anyone else that is generally labeled as a playoff faller would have their teams’ rORTG go *up* a good bit in the playoffs. Indeed, to take one example of someone labeled a playoff faller, I’ve previously calculated that Chris Paul’s offenses from 2013-2022 put up +5.1 rORTG in the playoffs with him on the floor compared to +7.90 in the regular season. Steph’s offenses from 2014-2023 put up +7.41 rORTG in the playoffs with him on the floor and +8.36 in the regular season. I didn’t previously run the numbers for Harden (and this takes a long time to do, so I’m not going to bother doing it just for this discussion), but I suspect it’d be similar. When we are talking about a very offense-centric player, I think his teams’ rORTG going up in the playoffs it’s very strong evidence that he is not a playoff “faller.”
Of course, theoretically it’s possible that Nash is a playoff “riser” on offense (as indicated by his offenses getting a bit better in the playoffs), but is a large playoff “faller” on defense and that outweighs the offense rising. His teams did struggle on defense in the playoffs so it’s not completely implausible. But the argument for that would have to demonstrate that he fell significantly defensively from his already-not-good RS defense level. That seems dubious to me. And, at the very least, the argument about box stats doesn’t get to it.