Figured it'd be interesting to compare our four young guys most people think of when "upside" and "tons of young talent" are concerned. I left out Oden because I just get too depressed when I think about him much. I also left out Roy and Aldridge, because it's a lot more clear where those guys will level off. Figured I'd also throw in some numbers from past youngsters we used to put so much hope in at similar ages, just for the sake of comparison. Stats are just for the year at the age mentioned. PER..WinShr..age..mpg..years in NBA..player 18.8 ... 0.7 ... 21 ... 22 ... 2 ... Batum 15.8 ... 2.0 ... 21 ... 19 ... 2 ... Bayless 14.6 ... 1.7 ... 24 ... 23... 2 ... Fernandez 12.8 ... 3.1 ... 23 ... 28 ... 5 ... Webster 15.0 ... 5.3 ... 24 ... 27 ... 6 ... Outlaw (last season) 13.0 ... 0.6 ... 20 ... 24 ... 2 ... Telfair (by far his best before or since) 13.0 ... 3.2 ... 24 ... 27 ... 3 ... Jack 19.6 ... 7.1 ... 22 ... 38 ... 3 ... Randolph (back when he was on a rookie deal and was widely loved) Do I have much of a point to make? Not really. Just thought it'd be interesting to compare past youngsters to the current crop and see how they measure up. I guess what really leaps out is just how vastly superior our current stable is compared to scrubs like Telfair and Jack. You caught the odd glimpse that those two guys might one day be decent, but it involved a fair amount of wishful thinking. This stable is hovering right around "average NBA starter" right now, which is fantastic for guys who are so young. I know Batum's numbers are really only based on 9 games, so it's pretty questionable to cite them. But I've watched those 9 games, and you know what? He didn't look to me like he was doing anything other-worldly. He looked a lot like he did in Europe in the summer. Maybe even a step slower and less involved. I really question if his PER "comes down to earth," or whether this is roughly what we can expect for the rest of the year. Kind of interesting to think about how next year Batum will be the same age and level of experience Randolph was when he absolutely exploded on the league.
I spent 11 minutes putting it into a chart, just like you should have. PER..win shares..age..mpg..years..player 18.8 0.7 21 22 2 Batum 15.8 2.0 21 19 2 Bayless 14.6 1.7 24 ?? 2 Fernandez 12.8 3.1 23 28 5 Webster 15.0 5.3 24 27 6 Outlaw 13.0 0.6 20 24 2 Telfair 13.0 3.2 24 27 3 Jack 19.6 7.1 22 38 3 Randolph
Final version. Then I can actually analyze the numbers. PER..win shares..age..mpg..years..player 18.8 0.7 21 22 2 Batum 15.8 2.0 21 19 2 Bayless 14.6 1.7 24 ?? 2 Fernandez 12.8 3.1 23 28 5 Webster 15.0 5.3 24 27 6 Outlaw (last season) 13.0 0.6 20 24 2 Telfair (by far his best before or since) 13.0 3.2 24 27 3 Jack 19.6 7.1 22 38 3 Randolph (back when he was on a rookie deal and was widely loved)
Win-share accumulates - it makes very little sense to compare it without knowing the minutes played. A better comparison would be Win-share per minute.
Playing on a winning team increases Win Shares and PER. If today's young players had been on Telfair's team, their stats would have been lower, too. The 2 most comparable players on the chart are Outlaw and Webster, with Outlaw definitely looking better.
That just makes it another rate stat. A raw win share comparison factors in the virtues of actually playing more minutes. abeing able to stay on the floor is an important, valuable thing. It's important to have both rate stats and cumulative stats.
It makes absolutely no sense to compare Batum's 7 games win-share to an entire year of ZBo at the same age. Sure - accumulative stats make sense when you compare single season performance of players - but there is nothing to be learned from them without the mention of minutes when you basically do a comparative analysis at a young age, which, by definition, seems to be more about potential.
For a random number of games, like Batum's, I agree. I was just speaking generally, as you seemed to be (to me, anyway). Comparing win scores for players with different MPG has plenty of value...but it should be over a similar number of games and a significant sample. Everything shouldn't be normalized to minutes.
Andalusian is talking about this group of players, not generally. Minutes per game is how to compare players who don't get many minutes.
At the same time, being superior players increases all of those numbers you mention above. Put it this way. I don't think it is an accident we are winning more games.
Well I know a lot of folks will disagree with me, but I don't think PERS is all that. For instance, Sergio Rodriguez has a higher PERS than Derek Rose. But would you take him over Derek Rose on your team?
Win Share, yes. PER? I'm not so sure. Randolph put up a really nice PER on a 41 win team. Seems to me that if you have lousy teammates, you'd have more rebounding and scoring and ball handling opportunities, which if you are halfway decent could lead to a better PER. Provided your FG% doesn't take too much of a hit from being better defended. Anyway, I probably should've added Qyntel Woods, since he was also once a "promising young stud" for this team. *shudder* In his rookie year he had a PER of 11.2 and a WS of .2. It's amazing to me now how giddy we all were over the guy back then. John Nash had pretty thoroughly beaten down our expectations, I guess. On the current roster Woods would likely be sitting behind Dante Cunningham.
PER is good and bad. You have to use it in the proper context. Putting up a high PER in an occasional 6 or 7 minutes of meaningless garbage time against the other team's third stringers isn't the same as posting the same PER when starting and playing 36 MPG against the best players in the league. All it really proves is the first player is the king of garbage time (our 12th man is better than your 12th man, see Randolph, Shavlik 2008-09 PER = 19.4). It's not JUST about sample size, it's also about the role on the team and the level of competition during the minutes played. Starters need to produce on a nightly basis against the other team's best players. End of bench guys get to pad their stats in meanigless minutes against the other team's scrubs who are more interested in padding their own stats than actually playing defense. In a nutshell, PER is most useful for comparing players who play the same position and similar minutes. BNM
