Stumbling on Wins
by Eric in Madison on Mar 20, 2010 12:16 PM CDT
Email
PrintDavid Berri and Martin Schmidt of Wages of Wins fame have a new book out soon called Stumbling on Wins that was recently reviewed on True Hoop by Henry Abbott. It's quite interesting, and I wanted to point it out for anyone interested. Some ruminations after the jump.
Stats analysts, right or wrong, generally get ignored for one of two reasons: either they aren't good writers, and hence do a poor job of communicating their ideas, or they lack humility and hence piss people off before they have a chance to really consider the ideas. Take, for example, Bill James. He's always been a pretty good writer (though he's gotten a lot better), but early in his career he was so blunt that he turned a lot of people off. It wasn't until he softened his approach by admitting there is much that he doesn't know that he has been embraced (relatively) by the baseball establishment. Abbott makes the point that Berri and Schmidt are too blunt in their assertions given how much we still don't know. However, they present certain ideas that are worth considering:
1. Teams radically overvalue players who do well in the NCAA tournament. Something we should especially be aware of as the Wolves evaluate draft prospects. Berri writes that appearing in a final 4 can improve draft position by 12 spots. Yikes. I wonder if we didn't see a species of this last season with Jonny Flynn. His amazing game in the Big East Tournament, plus getting to the Sweet 16 could not have hurt his draft stock; and in fact he began to shoot up draft boards around this time. Beware the late riser. Another draft note they make is that, despite rebounding ability being one of the most transferable skills from college to the pros, rebounding ability does not really effect draft position. Points, efficiency, size, and assists are among the stats that do effect draft position.
2. NBA teams radically overpay for scoring. Undoubtedly true, and perhaps at the center of our site's debate over Al Jefferson. I used to fall into this trap, and probably still do on occasion. I used to defend Allen Iverson, arguing that a) getting 25 shots off in an NBA game is a skill, and b) that even at low percentages, his teams were often better off with him hoisting because that's all they had. We have a slightly different argument about Jefferson, it seems to me, which is that you need someone who can put the ball in the basket, and that trading away Jefferson will create such a dearth of inside scoring that to trade him away would be a big mistake. But he's also the highest paid player on the team, and he's so highly paid because he scores points. If his points are overvalued by his contract, then trading him does make sense.
There are other interesting notes in the review, about peak age, defense, and more. You should check it out.
30 comments
|
0 recs |
Do you like this story?
Comments
As it relates to point #1
I really hope David Kahn is not watching the NCAA’s….Seriously, just don’t go there and get all emotionally attached to certain players (let us fans do that!). Instead, at the end of the season, he should have video footage across the entire season of top prospects along with synergy sports stats and any other type of advanced data. The player’s entire body of work needs to be analyzed and poured over, with the most weight applied to conference play, where teams scout each other and know each other very well. He should have his draft board 90% set by the time the combine, workouts, and interviews take place. Those should only be used for tweaking. Had he taken this approach last year, Jonny Flynn would have never been drafted by us.
Well put
Though apparently he and Rambis are stopping on the way home today to see Kentucky play, then going to Milwaukee Sunday to watch Ohio State-Georgia Tech.
I am scared of the player who gets hot late catching their eye. On the other hand, if they get a top 4 pick, those spots are more or less in stone. The problems might come with the Charlotte pick or if he trades into the lottery for a 2nd pick again.
by Eric in Madison on Mar 20, 2010 12:43 PM CDT up reply actions
NCAA's
I agree, to an extent, that NCAA Tourney performances are overrated for draft purposes.
But, with the “1 and Done” nature of the superstar’s college career, it’s an important time to see what the 19-year olds can do after a full season of college ball.
Derrick Rose was much-better by the time he was playing in the tournament than when he first arrived at Memphis. It sounds like the same could be said for Derrick Favors, this year. Wall and Cousins have been consistent throughout, so there probably isn’t a lot to be learned from them in this tournament.
So, if we’re talking about Derrick Favors, then I disagree with Rascall’s advice to Kahn. I think he’s a player to watch closely, right now. But if Kentucky wins it all, that doesn’t really change much re: Wall/Cousins — at least not in my mind. We can see what they can and cannot do, and whether they cut down the nets or lose to Wake Forest tonight won’t really change that.
I agree to an extent
that when there is so little info on these freshmen, you have to use all of it, and seeing if there has been improvement matters.
On the other hand, it’s not guys like Favors or Cousins or Wall that this applies to. They are going top 5 no matter what happens. It’s the guy who was a 2nd rounder or projected in the 20s that has the huge tournament and moves into the lottery that scares me. It’s like what happened to Hassan Whiteside, except it happened to early for him. All of a sudden, he went from nowhere to being a top-8 guy. Now there’s been time to reconsider, and he probably won’t go that high. We’ll see.
One guy who could move up on a big tournament that scares me is Quincey Pondexter from Washington. One guy that could be hurt by a relatively poor performance is your guy, Greg Monroe. It’s these guys that relying on the tournament is dangerous for.
by Eric in Madison on Mar 20, 2010 12:56 PM CDT up reply actions
Good points
Although Wall, etc. are definite top 5’s, there still may need to be a decision made. For example, Wall vs. Turner at #1 or Favors vs. Cousins at #3. Relying too much on the tournament games could result in making the wrong call. I’m actually kind of hoping Turner lays an egg against GT and Favors has a huge game! I can’t wait to watch that game since there is a 50% chance we end up with one of these two players.
by Rascal Flatts on Mar 20, 2010 1:04 PM CDT up reply actions
I thought Monroe had said he was returning next year?
At least that’s what I heard he said immediately after they lost. Time and a clear mind will probably change that though.
I wonder if this kind of applies to Cousins as well. At the beginning of the year, Favors was the second pick in the mocks and Cousins was somewhere around the 20s (I remember because I was hoping he stayed there so we could get him!). Obviously he has proven to be a better pick than the 20s, but is he really a top 5? I think I would still take Favors over him, but after Favors… I think his full season of physical domination (would’ve been a treat to see him match up against Favors this tourny!) probably offsets his attitude issues just enough to keep him at least in the top 8, probably top 5. An argument could be made for a couple people at 4, but I think Cousins has by far the most solid case to be in the second tier of the draft prospects. I think the tourney will actually be a pretty good marker for Cousins’ talent though. If he has a good tournament, it might be enough to push him ahead of Favors.
Monroe
did say he was coming back, but even his coach was dubious. JT3 said that could change.
Lots of guys say they are coming back but ultimately declare.
by Eric in Madison on Mar 20, 2010 1:07 PM CDT up reply actions
Or how about...
Jordan Crawford!
The only thing I knew about this guy was that he dunked on LeBron in a summer camp. He looked like a star yesterday… so I checked out Ford’s Top 100, where he was ranked 93rd. That tempered my enthusiasm a bit. If he goes Top-20 (if he declares) then that would probably be your Exhibit A of the March Madness Effect.
Or a guy playing right now
that someone just posted a fanpost on: Omar Samhan from St. Marys. Dominating Villanova. Almost certainly lacks the athleticism to play in NBA, though.
D/X just compared him to Rafa Araujo. Ouch.
by Eric in Madison on Mar 20, 2010 1:39 PM CDT up reply actions
Favors
Favors will go Top-5, but that doesn’t mean the Wolves would necessarily want him. Let’s say we pick 3rd (I mean, given our history of drafting in the (X + 1) position in X player drafts, is there really any other outcome?) does Kahn sit tight at Number 3, or does he trade away significant assets to move up and draft Wall or Turner? Or does he trade down? I still think Favors’ NCAA performance, however many rounds it lasts, is important to watch, since there’s no requirement that we draft and keep him if we pick third. If he dominates en route to an Elite 8 or so finish, Kahn and fans alike might be excited to have him — moreso than right now, when he’s been a somewhat inconsistent performer with off-the-charts athleticism.
A not unrealistic scenario
the question is whether that 3 game run should have the effect of changing Kahn’s plans.
I get your point; there’s a small sample size on Favors altogether. It’s important to watch as much as possible. I just don’t know how much we should be swayed by a tournament performance, even for a guy like Favors.
by Eric in Madison on Mar 20, 2010 2:00 PM CDT up reply actions
I should add...
that Chad Ford is convinced that Favors can play himself into the Top-2 with a strong tournament, depending on who picks 2nd. Now that would be placing importance on the tournament, given everything that Evan Turner has done over the course of the season.
But they can still
look at the stats and video footage to analyze the improvement curve of freshmen. Where I get scared is when they go to a game with a bunch of other people and watch it like a typical fan (albeit a very knowledgeable fan) and get caught up in the excitement. That’s not the type of cold, sober analysis we need. Players like Randy Foye, Corey Brewer, and Jonny Flynn were picked based on emotional attachments that were formed due to one or two events, whether it be a certain game or an interview or a workout. We can’t make event-based decisions. Each event needs to be weighted darn near equal, except non-conference “silly season” games. Those shouldn’t count for much.
by Rascal Flatts on Mar 20, 2010 1:00 PM CDT up reply actions
Like your point about emphasizing conference play
Getting excited about non-conference college play is the equivalent about getting excited about what rookies do in summer league (see Jonny Flynn 2009).
The conference play is why I really like Turner, he dominated playing in a tough, defensive minded conference.
no synergy
well, SNP noted the other day that the wolves don’t use synergy so there’s no chance of watching college video. Bastards.
I believe I read an interview with Sam Presti where he said that he never watches NCAA tournament games live. He watches them a month or two later. He does send front office people to the games but he does not go.
He’s the decider, you see. And he does not want his ultimate decision to be influenced by the excitement.
What’ so F**king smart about that is the recognition that you can’t control yourself. You can’t go to the games because the atmosphere WILL affect you even though you know in advance it WILL affect you you cannot protect yourself. Smart dude.
David Kahn, jury’s still out.
Synergy stats are different than college video
Every team uses a combination of tape and in-person observation. And to imply that Kahn or any other GM/scout can’t separate the atmosphere from their job and then tie it in to an overall evaluation is dubious analysis at best. Presti’s not exactly a savant; he’s a guy who’s done a good job of stockpiling picks, but his draft record is above-average and aided by lottery luck.
by pagingstanleyroberts on Mar 21, 2010 11:46 PM CDT up reply actions
The odds are likely
That Dr. Berri will soon tell us on his blog how DeMarcus Cousins and Evan Turner are the two best prospects in the draft based on College Performance. How many fans let alone GM’s would argue for those two being the Top 2 ahead of Wall and/or Favors. I think you could make a strong argument that this is the way it should be in an unemotional assessment. I wonder as I read littleboxes blurb on Presti not going to NCAA games because he doesn’t want the atmosphere to adjust his assessments, whether 12 months of the Media Echo Chamber telling all of us how John Wall was the unquestioned #1 pick affects our ability to discard that “known truth” and make a stats based conclusion that flies in the face of the media created conventional wisdom.
Cousins is the one who is flying too far under the radar from a statistical point of view. I ask myself whether his passion and quick fuse on the court is any different from someone like KG and how I’d rather have someone who cares, albeit too much, than someone who doesn’t care at all.
So Cousins has put up better stats than Wall, forced teams to gameplan around him rather than wall, and plays a position of greater value with a scarcity of good players; yet, Wall is the #1 pick.
Whose side are you on EiM? Dr. Berri’s or the Conventional Wisdom?
It seems to me that one of the questions here
is: can we statistically model everything? Or at least, everything that matters, from performance, to size, to body type, to combine type measurements, to quality of opposition, to quality of teammates, and then draft based on that statistical model(s)?
This is a real question, by the way. The immediate reaction is: of course we can’t. We have to see these guys. But I’m not 100% sure this is true, though it’s true today.
We don’t do that now; and we make plenty of mistakes. Would we make more or less mistakes if we tried? How much more information do we need to come close? It seems to me that some of the stuff that Synergy tracks is relevant.
As for Wall….when everyone says he’s the number 1 guy, I certainly don’t feel qualified to say that he isn’t. It isn’t as if he doesn’t have good numbers. As a general rule, I believe in data, but data can take a lot of forms.
But you’re right; Wall has been the #1 guy for so long that it’s hard to look at it dispassionately and determine if he is the best guy. And even if you can put everything aside and come to a conclusion, do you have the guts to act on it as a GM with the #1 pick if you decide he isn’t the best guy? Because if you take someone else, and you are wrong, it costs you your job, whereas if you take Wall, and he doesn’t work out, well, everyone would have taken Wall.
One hopes that one’s GM uses the appropriate evaluation tools to reach sound conclusions about draft prospects and acts accordingly.
by Eric in Madison on Mar 20, 2010 3:42 PM CDT up reply actions
The thing about stats
is that some guys’ games are just more suited to the NBA. More spacing and less focus on an individual. Most of your success depends on work ethic obviously, and Turner seems to have plenty of that. But Wall’s game is definitely more suited to the NBA.
It’s not rare that a guy all of a sudden becomes more efficient in the NBA than he was in college and IMO Wall will be one of these guys and Turner will not. Not that Turner will be bad, I just think his production will take a slight hit and Wall’s will get a slight boost.
I understand what your saying
But if Turner were to become slightly less efficient and Wall becomes slightly more efficient, Turner is still the more efficient player based on actual on the court performance versus projection.
How much more “athletic” do you have to be where we should draft you above someone who outproduced you? I think it’s clear when the non-athletic player is also undersized; however what if the non-athletic player has enough size to defend his position. I don’t know if there are easy answers to these questions, but I also don’t think it’s easy to dismiss Dr. Berri’s theory that the more productive you are in College, the more likely you will be more productive in the NBA.
I love both players...
but let’s not forget that Wall is a freshman and Turner is a junior.
Turner’s freshman year, he averaged 8.5 pts 4.5 rebs 2.5 ass. in 27 mpg. His a/to ratio was 1:1. Not exactly stellar stuff. He’s clearly improved, and I think John Wall will do the same in two years, with a much-better starting point.
He’ll be damn-near impossible to guard in the NBA game.
He's pretty freaking good
What you notice about him is:
1. He has unreal talent, but
2. He’s remarkably patient, and
3. He really guards people
Sometimes I think he’s almost too patient, but when you have that much talent around you, you have to feed it.
by Eric in Madison on Mar 20, 2010 8:13 PM CDT up reply actions
And you hit the nail on the head
that’s exactly what it comes down to. Do you think Wall (who Andy G points out is 2 years younger) will improve enough to climb to where Turner fell to or not? I think he will. And I think he will exceed Turner by quite a bit in a few years. I also think Turner will be on a championship team before Wall, but I think Wall with no talent could get an eighth seed and Turner with no other talent couldn’t. The tradeoffs are tough to sort out. Both project to be great players, and more productive NBA players were definitely productive college players as well. But at the same time, the most productive college player [probably] won’t be the most productive NBA player. So finding that fine line and choosing on the right side of it is the hardest part of the draft. There’s correlation, but it’s certainly not 100% or even 75% IMO (as in, I think about 3/4 players in the lottery meet or come close to their average expectations. Not that 3/4 meet their lofty potential).
Hope this made sense, didn’t do any editing and my thoughts were kind of jumbled.
If what you say above is true
I hope the Wolves get Turner.
BTW, everyone here talking about Wall being more talented in a couple years is why I want players to have to spend a couple years in college. What if Wall has already reached his potential?? Of course he probably hasn’t, but colleges are filled with guys who didn’t get much better over a four year career.
That’s why I have been favoring Turner for the Wolves, even if he doesn’t get much better he is a starting SG for a decade on a good team. Of course we hope for more.
On another note
Rubio went 5/5 from 3 today as Barca won on the road against the 3rd place team in the ACB (Caja Laboral). This is Tiago Splitter’s team, though, for unclear reasons, Splitter only played 10 minutes (maybe he got hurt, he wasn’t in foul trouble).
by Eric in Madison on Mar 20, 2010 4:02 PM CDT reply actions
Glad to hear it...
if we’re lucky-enough to get Wall or Turner, Rubio will fit in a lot better if he can shoot well from outside.
I think scorers should be rewarded
although not over-emphasized. Scoring is ultimately what makes the game fun. If games were regularly 30-28 at the end, attendance would be about 2,000.
Comments For This Post Are Closed
Secondary Sidebar
User Tools
FanPosts
Community blog posts and discussion.
Recommended FanPosts
Recent FanPosts
Next Game
In Association With
Ball Don't Lie - NBA - Yahoo! Sport
- Days of NBA Lives: Wherein Roy Hibbert tap-dances to Maine
- Grizzlies Valentine’s Day cards are the key to love
- Kevin Love would like the NBA to take the MLB’s approach on All-Star games
- Kobe Bryant fires a few hilarious salvos at Jeremy Lin
- Brandon Jennings states the obvious regarding his 2014 free agency, everyone goes nuts
Certain photos copyright © 2012 by Associated Press or Getty Images. Any commercial use or distribution without the express written consent of Associated Press and Getty Images is strictly prohibited.
