I thought it might be kind of interesting for each of us to pick a player not expected to go in the lottery and try to guess the guy that you think will have the best rookie year -- to keep things on the level we'll use PER as the measuring stick and the player needs to log at least 1000 minutes of action during the regular season. I'll bookmark the thread and I'll try to remember to bump it in a year. My pick: Dominique Jones EDIT: changed minutes threshold
If you're using PER as a measuring stick, my guy Pondexter goes away. I think he'll go to a playoff team, show serious value on the defensive end and get a chance to get on the floor b/c of that. But I can't think of a playoff team other than ours where he'd be even a 3rd option on offense, and therefore PER will not be high--unless he goes to MIN with their pick at 23 and they plug him in with Brewer at the 2/3...now that I think about it, Flynn/Pondexter/Brewer/AlJeff/(Cousins or Favors) wouldn't be that bad a lineup to toss out there. But the way MIN thinks, they'd probably take Paul George at 16 if he's available and punt the 23. In which case he'd have a great shot. If Hayward goes to Indy--but that'd be lotto. Great question. I guess if you had to pick before the draft...I'd say George, with Charles Garcia as a sleeper.
PER isn't perfect, but it does provide a generally well recognized metric of efficiency that rewards rebounding, high shooting percentages, assists and low mistakes ... I had to pick some kind of semi-objective measuring stick, sorry.
no worries, I understand needing a metric, and that you're callin it out beforehand. And PER's better than most. It's just going to eliminate guys like Whiteside and Pondexter and Orton and others who aren't as efficient or polished on the O end, but get some serious run due to defense. Where if KP drafted Paul George somehow he might not crack the rotation due to defensive issues
500 minutes? That's like 6 minutes per game. I think the playing time boundary should be a bit higher. Like 1230 minutes. That's 15 minutes per game for someone who plays all the games. It also allows players who are impactful enough to play like 30+ minutes per game to miss time to injuries and still qualify.
I'm going with the French big man, Kevin Seraphin. BTW, that's also who I'd like us to take if we stay at 22.
Win Shares per 48 is good imho, and takes PER into account. Plus, it tracks the potential affect the kid has on the team around him, and how positively he affects the outcome while he's in the game.
I just picked Hollinger's baseline for inclusion on his PER rankings, but you make a good point. I'll adjust it up to 1000 minutes.
Aren't win shares tied to team record though? When we're talking about non-lottery rookies (ie, role players), is it reasonable to assume that the player on the better team actually had the better season? For example, Anderson Varejao got 8.1 WS this year, 0.178 per 48. Brook Lopez got 7.9 WS, 0.125 per 48. But does anyone really think that Anderson Varejao played significantly better this season than Brook Lopez?
Win Shares pretty much adjusts for teammates. No one stat is perfect but looking at how it correlates historically, it is a reasonable stat if not the best imo. Moreso than PER in most instances. PER and WS/48 don't account for durability though. Anderson Varejao was on Hollinger's first team defense.
I'm all about discussing the merits of one advanced stat or another, but that really wasn't the gist of this thread. So far we've got 4 player picks and 13 posts ...
You could divide total WS by team record to get a "Percentage of Win Value"... this makes Brook Lopez's 7.9WS (and consequently, his 0.125 per 48) much more valuable.
That isn't how you're supposed to do it (just sayin'), since superstars have a high amount of win shares even on poor/great teams. If you go to 82games, you'll notice Brook gives up 20 PER on defense. And he had a 110 defensive rating.
