Frank: Held out as long as I could. Feels good. =)

Third – I like Faried a lot, but something tells me there is something about him that might not show up in WP48. Like, for instance, that he was the 370th ranked defensive player in the league by Synergy (math math math 30 teams x 12 active players = 360 players). And Harrington was (somehow!?!?) the 46th ranked player in the league.
The spread probably isn’t really that wide, but that’s probably more than…

When you see Blair and Faried dominating PAWS40 and WP48 while being considered only decent players (neither plays a ton of minutes, and Poppovich at least knows how to set his rotation better than anyone here) isn’t it time to start looking at why that discrepancy exists? Like the fact that they are defensive sieves that can’t score without a teammate’s help? Or that neither can space the floor, an increasingly important offensive skill set in today’s NBA? Their rebounding is amazing, and that certainly has value, but the hall of fame rebounders are all there because they were also elite defenders, grabbing rebounds isn’t as important as preventing your opponent from getting inside the paint

Fourth – and I can’t emphasize this enough even after having written this 5x in the past — a team’s draft list is the culmination of scouting/stats/intuition. Some guys are better at it than others. Even the best GMs have swung badly and missed. Some of that is just being a bad judge of talent, and probably a lot of it is just bad luck/circumstances.

The most blatant bust of the 2009 draft was obviously Hasheem Thabeet – a pick made by Chris Wallace who was also the architect of one of the most laughed-at trades in NBA history (Gasol trade). But in retrospect, Wallace doesn’t look all that dumb now for that Gasol trade does he, now that Marc Gasol has turned into a top-flight center? Anyway, let’s say that Wallace’s draft rankings for 2009 were like this (I’m pretty sure they weren’t, but just for argument’s sake):
1) Griffin
2) Thabeet
3) Harden
4) Rubio
5) Lawson
6) Curry
7) Taj Gibson
8) Blair
9) Jrue Holiday
10) Darren Collison

MAN – 9 out of the top 10 guys on his list are really spot on. Unfortunately, the one mistake he made came at the spot he picked. That’s just bad luck. If he had the #3 pick, maybe whoever has the #2 pick takes Thabeet and Wallace picks Harden and looks like a genius. JUST like Presti picking Durant – if he has the #1 pick rather than the Blazers, maybe he takes Oden and HE’S the unlucky one.

GMs only have 1 chance to look good, and if they miss, they look like idiots. Guys who publish draft rankings look great if they somehow ranked a guy a little higher than he was taken, and that guy turned out to be good. They can be wrong 60% of the time, but if those 40% were really good, WOW!

And re: Isiah – we really should just move on re: his trade/FA history. Comparing WP48 to the worst GM in recent memory is like shooting fish in a barrel. Try picking on someone actually good.

Basically, you can’t discount “basketball experts” because some of the people being paid to make basketball decisions are terrible at it. What you do is look at what the GOOD basketball people are doing. Any one of Poppovich’s actions is more indicative of proper basketball analysis than Vinny Del Negro’s entire career.

What you also seem to purposely miss is that I’m sure half this board would actually agree with you that guys like Bradley and Austin Rivers shouldn’t be high picks. I don’t think they are trash players, but I have tremendous doubts about Rivers’ ability to produce in the NBA. Flashy scorers like Jamal Crawford do get a dumb amount of attention, especially from the media. The problem is that most of the “busts” frank mentioned don’t lack flashy scoring, they lack any ability to play defense. Seriously, out of the top 5 for 2009 AND 2010, not one of those players would even be considered an average defender at this point in this career, nor will they ever be close to elite, no matter how much they practice. So even assuming that high offensive efficiency and rebounding numbers are the primary ways to derive value from the box score, you totally miss out on half of what a player will be required to do in the NBA. And impact cannot be boiled down to a context free count of shots made vs. shots missed, or total rebounds gathered. So there’s that

ruruland:
Bam. Frank’s back

Held out as long as I could. Feels good. =)

OK here goes:

The Honorable Cock Jowles: You’re assigning binary values to players who haven’t even played yet

The Honorable Cock Jowles: …basing decisions that are naively based on PAWS would lead to a better team than choosing players (like Austin Rivers) who have terrible stats but “good” attributes like a “sick crossover” and “extreme [confidence],

Anyone else see the contradiction here?

First, at some point a reasonable person would accept that if a player can’t even get into a game for an undeniably GOOD coach, that that player might not be that good. For every 1000 of these players, there might be 1 Jeremy Lin. But we’re talking about statistics here, not the 1 extreme outlier.

Second, the fact that someone dominates at the college level (or even NBDL level) does not mean he will dominate at a higher level. I currently dominate my 2 year old in bball, does that mean that will hold as he gets wiser/stronger/faster (like what happens to competition when players go from college to NBA). For some players yes, for other players no. For Nick Fazekas, no. (cue THCJ reporting to me Fazekas’s WS/48 or WP48 = a tautology).

Third – I like Faried a lot, but something tells me there is something about him that might not show up in WP48. Like, for instance, that he was the 370th ranked defensive player in the league by Synergy (math math math 30 teams x 12 active players = 360 players). And Harrington was (somehow!?!?) the 46th ranked player in the league.
The spread probably isn’t really that wide, but that’s probably more than…

“PAWS40 had a miss rate of 60% in the top 5. Also in 2010.” You are saying absolutely nothing. My opinion has been that PAWS is a fuckload better than whatever analysis NBA scouts are using (cough cough points per game), and that basing decisions that are naively based on PAWS would lead to a better team than choosing players (like Austin Rivers) who have terrible stats but “good” attributes like a “sick crossover” and “extreme [confidence],” which is how some Yahoo or ESPN.com writer described Rivers’s upside.

And again, Nick Fazekas is a dominant player on the level he’s been allowed to play at. Do you really have that much faith in the decision makers at the NBA level? You know that the Sixth Man of the Year award goes, each and every year, to the bench player who scores the most points per game? You know that Juwan Howard and Antawn Jamison still get playing time? That a GM recently traded the farm for Stephon Marbury, Eddy Curry, Jamal Crawford, and Steve Francis, then signed Jerome James to a 5 year, 8-digit contract? Do you really think that “experts” in sports are legitimately “experts,” with some high degree of cognitive and visual efficiency that makes them better than predictive statistical models, even if the models are not ideal?

1) Player X is great because my stat system says so

2) Player Y would be great except that his coach won’t play him. It’s the coach’s fault (even when it’s Greg Popovich) that he can’t see the genius of PAWS40/WP48 etc. It certainly couldn’t be that some players generate “false positives” for being a good player

3) Player Z that isn’t predicted to be good by PAWS40/WP48 isn’t a good player no matter what anyone else says because he’s not good by PAWS40/WP48

This is like the Wall Street bailout — heads THCJ wins, and tails every other stat system loses.

Look, in 2010, PAWS40 had a 60% MISS rate on the the top 5 players (I would categorize Zoubek, Aldrich, and Damion James as misses).

In 2009 the top 5 PAWS40 were Dejuan Blair, Blake Griffin, Ty Lawson, Ahmad Nivins, and Lester Hudson. I know you love Blair, but Blair can’t even be trusted to play more than 22 min/game in the reg season and more than 4.5 min/game in the WCFs for probably a top 5 coach in NBA history in Pop. For the #1 ranked guy in your scoring system, that qualifies as a MISS. The #1 ranked guy should be an All-Star caliber guy, not a bench guy that can’t even get off said bench in the biggest series of the season.

So that means that in 2009 PAWS40 had a 60% miss rate.

How is this a good system again? GMs are fired for making one or two bad high lottery picks.

I’ve said it 100 times and I’ll say it again – just because a system can predict one or two Ty Lawsons doesn’t mean that it should be forgiven for predicting greatness for Nick Fazekas, Ahmad Nivins, Lester Hudson, Brian Zoubek, Hassan Whiteside, etc. Like Bill Clinton just said, even a broken clock is right 2x/day.

And re: Bradley – he’s a GREAT defender even if he gets away with fouling constantly (like Shump). And he’s turned into a very reliable 3 point shooter (41% for year). IMHO the Celts win the ECFs if Bradley is playing.