“Oldskool”, the problem with your explanation is that if true, we’d expect to see roughly the same result for the rest of Seattle’s starters. We’d expect that the major minute players would all see negative net plus minus because the garbage time players would swamp them. But that’s not the universal pattern. The Sonics didn’t show the same relative improvement when Wilcox or Watson left the court. It’s a peculiar cummulative garbage time you’re presenting that affected Durant’s relative rating but didn’t produce a similar effect on all starters. Your explanation falls apart in that light. Again, it’s what’s called special pleading where a case applies only to a single explanation but doesn’t seem to hold up to deeper analysis. Sorry.

Let’s be very clear about what was going on in Seattle. They stunk with him. They were not defeating their opponents when he was on the court. They were not winning when he was off the court either, but they weren’t as bad during this time. The same pattern does not hold for all of Seattle’s regular starters. It’s really a matter of the team sucking less when Durant sat on the bench.

It’s always great when someone who seems to lack an elementary understanding of something declares that I’ve failed to prove my point. You failed to make sense. So please, calm down. (And while you’re calm, learning a bit about statistics and the ‘shift’ key on your computer wouldn’t hurt you.)

if 20pts on 43% shooting and 29% from 3pt land is only average. then, I must be wrong into thinking that durant isn’t as good as I’m out to prove. In his first year, learning to play under a new system, a higher level of competition, and new teammates, I’d say he did better than average.

The average NBA shooter shoots about ~45% from the floor and ~35% from beyond the arc. The average points per shot is ~1.24. (Durant, for what it’s worth, returned 1.19 points per shot, again below average) That’s the average for EVERYONE. That average is made up from the shots that stars take and the shots that NBDL 10 dayers take.

Rookies generally don’t do as well as the league average, so perhaps Durant fits in as an average rookie, but he wasn’t particularly good. For what it’s worth, Seattle, as a team shot 44% and 33%. Since Durant was part of that average, it means that he was actually dragging down their average. It’s possible that if he didn’t shoot as much his bad looks would have gone to others who similarly would have seen their numbers fall, but there’s no denying that he was shooting worse than the rest of his team as a whole. He was above their team average in points per shot because he was effective in getting to the FT line and hitting shots there. Overall this suggests strongly that his 20ppg was a large part of taking a bunch of shots, it wasn’t something particularly impressive. It was a midline result on a lousy team, suggesting overall, it wasn’t good by NBA standards and wasn’t worthy of ROY honors.

We don’t know what would have happened if someone else took some of his shots, but actual data suggests that when high volume, low percentage scorers are removed from lineups, scoring doesn’t drop significantly and can improve. This isn’t opinion. This is the result of studying many seasons of data. The high-volume, low percentage shooter can usually be replaced with similar (and sometimes better) production spread out among the rest of the team. Could Seattle have done this? 30% of their on court minutes suggests that they could have. That’s a whole bunch of time to draw conclusions from.

Replacing a 43 and 29% shooter isn’t that hard. The average NBA player does better than this. He took too many shots given his sub-mediocre results. May not be his fault. May be that the coach wanted that. From his history here and the way he utilized high volume low percentage shooters elsewhere, the notion that PJ is a moron cannot and should not be overlooked. May be that other players just allowed it, but it did hurt his team relative not just to what the rest of the league did, but what the rest of the TEAM did. In the 30% of all available minutes (not all of which are “garbage time—it’s not like Durant played the first 35 of every game and then sat for a meaningless final quarter each time out) the team as a whole really was more efficient. His individual plus-minus was pretty constant for all 4 quarters suggesting that it wasn’t really a matter of the matchups that are responsible for this result.

They really did shoot better when he was off the floor. It might not be that they’d achieve the same result for a full game, but data suggests that it’s possible they would. There are a whole lot of variables, but these variables have been studied by people who have mad skills and training studying multivariate statistics. It’s not conclusive, but there’s much to suggest that his contribution could not just have been replaced, but could have been bettered by players on his team.

This doesn’t go to what he could do in a better situation (better teammates, a better coach instituting a better system) but what he did do was below average for the league and didn’t appear to aid his team above the abysmal level they’d have achieved without him.

The illusion of points per game is hard to overcome, but it’s an illusion. It’s probably the most common measure of a player’s worth, but it’s also among the more unreliable ones. Points per game is like batting average in baseball; it’s not meaningless, but it’s not a good predictor of team success or player worth compared to things like slugging percentage, on base percentage, OPS. But there’s historic reasons that people look at batting average. And there’s strong resistance to avoid looking at points per game as a measure of worth in basketball. I suspect that in basketball, part of this is because scoring is conspicuous, charted and the game has many points, even in defensive contests. It’s a hype-able figure and no league seems to have more meaningless hype surrounding it than the NBA. But it’s still not a particularly reliable measure in many, many cases. There are very often high ppg scorers on bad teams. There are seldom high FG%, high ppg scorers on bad teams.

Of course all of this is a distraction from where Durant really hurt the team. The offensive production was similar with him on and off the court (predictable given that he was below average shooting but a bit above average when you include his FT potential). Where they got killed was on defense where they gave up points at a much greater rate with him on the court than when he sat. Again, explaining more than a quarter a game’s worth of data as entirely “garbage time” is a stretch, a real stretch, especially when we don’t see the same sort result for all of Seattle’s starters, which we would if ‘garbage time’ explained things. PER, a stat that ignores most of defense and over-rewards high volume shooters, said he was strictly average. Factoring in these two limitations and there’s no way he was even an average NBA player last year. I don’t think he was close to the best rookie either. Sportwriters still pay too much heed to hype and points per game. Sportwriters are often wrong.

The analogy with football stats has a big problem: football stats show very little consistency from year to year and with increased or decreased playing time. Basketball stats are more consistent and in many, many cases, scale much, much better (again, not opinion, but researched, published and academically reviewed studies indicate this). A 5 rebound in 20 minute player is likely to get 10 if his playing time is doubled. That’s what actually seems to happen more often than not.

I really have nothing against Durant as a person and don’t care one way or another if he succeeds, save that I don’t want it to get in the way of Warrior success. I don’t fear Seattle with him though and won’t until he does something more in the league. I think what he displayed last year doesn’t indicate that he’s going to be the type of player who elevates his team to lofty playoff levels on his own. Few, few players do that. Based on the hype surrounding him when he was drafted, it seems like people expected him to be one of those very few. Early returns suggest otherwise. Most of the players to elevate a team on their own started doing so almost immediately, though most of this was before players entered the league in their teens, so things may now be different. Still, Durant has been 90% hype, 10% baller so far.