My general rule is that I post at most once a day. Although I am tempted to do more than this, my job as a professor (for which I am paid by the state of California) is about teaching and research. For nine months of the year I teach and in the summer I focus on research. Yes, blogging is fun. But I don’t think it qualifies as research (or teaching).
Unfortunately my one entry a day rule means that some issues don’t get addressed very quickly. Since I didn’t post yesterday, I thought I would break my rule this afternoon and post a few quick notes on some stuff I’ve missed.
Is the NBA burning?
Bill Simmons posted a lengthy Page Two column at ESPN.com yesterday. The column — entitled “One man out, one league in trouble” – finished with the following statement:
Let’s just hope we’re not watching a documentary about the death of the NBA some day, because we’re headed that way. Wake up, fellas. Rome is burning.
I commented on this issue at both The Sports Economist and in this forum. Let me throw out a few facts on the NBA.
Let’s start with attendance. The past three seasons the NBA has set an attendance record each year. More people are going to NBA games than ever before.
Then there is television. Brian Goff, my fellow sports economist at the Sports Economist, noted that television ratings for the NBA Finals peaked in 1998. Since then, ratings for the Finals have declined dramatically.
But I don’t think ratings are the story here. What matters is what cable television and the networks are paying the NBA for the right to broadcast these games. As reported by Sports Illustrated.com, last month the NBA extended its deal with TNT, ABC, and ESPN. The extension calls for the NBA to be paid $7.44 billion over eight seasons (beginning in 2008-09 and ending in 2015-16). That works out to $930 per season and for each team, $31,000,000 each year before a single ticket is sold.
Compare these numbers to the 1997-98 season, when the ratings for the Finals peaked. According to InsideHoops.com, in 1997-98 each NBA team earned a little more than $11 million per team from the national cable and network television contracts. In a bit more than ten seasons, this payment has now about tripled.
But why have ratings declined? I think people need to look at two issues. There is just a whole lot more on TV than there was in the past and increased competition will lead to a decline in demand. Furthermore, the NBA has put a whole lot more games on TV. As we saw with in other sports (like college football), when you increase the supply of games, ratings for any one game will decline. Of course the networks are aware of the ratings issue and must not be too concerned. Well, if they are, paying the NBA more than $7 billion over eight years is an odd way of expressing this concern.
Beyond attendance and TV, I also want to look at the salary cap. The cap is set according to the stream of Basketball Related Income (see the NBA Salary Cap FAQ for more information). According to InsideHoops.com , the salary cap in 1997-98 stood at $26.9 million. For 2007-08 it is set at $55.63 million. In other words, when we consider Basketball Related Income, the NBA has never been better off.
When we look at attendance, TV revenue, and what we learn from the growth in the salary cap, we see a league that is doing better than it ever has in its history. And yet we hear that one referee having a gambling problem is going to bring down the league.
The gambling issue is one of integrity. If people believe games are fixed, or not honest, demand will decline. At least that’s the story we are being told. Of course, the media also told us that if labor disputes take the games away, fans will never come back (and The Wages of Wins and our own published research have demonstrated the problem with the labor disputes story). So apparently, cries that the “sky is falling” haven’t been accurate predictors in the past.
Again I will repeat what I said on Sunday. The real issue is the integrity of the press. When you sensationalize every story your credibility declines. Everything that happens is not the most important event to ever happen in our lives. Please, try and have just a little perspective. Yes, I understand the incentives people face. You want people to read your column and you can do this be making everything sensational. But isn’t there just a tiny amount of interest in getting the story right?
David Lee vs. Kobe Bryant
Yesterday morning I got an e-mail from Owen Breck (who you will note often leaves comments in this forum) alerting me to a comment on The Wages of Wins posted at Knickerblogger.com. The comment — entitled “Trading David Lee for Kobe Bryant Straight-Up: Shrewd Sabermetrics or Laugh Test Flunkie?” — argues that since Wins Produced indicates that David Lee is more productive than Kobe Bryant, and “everyone” knows that isn’t true, Wins Produced is not a very good measure of performance.
Owen has done his best to respond to this argument, but I told him I would ask other readers of The Wages of Wins Journal to chime in. So if you get a chance, read the Knickerblogger essay and leave a comment here (or there). I would love to see what everyone thinks of this essay.
The Position Adjustment and Matthew Yglesias
Two weeks ago I posted a column entitled “Rewarding David Lee.” At the end of this post I said “I also wish to offer a response to the latest comment on position adjustments from Matthew Yglesias. That, though, is going to have to wait until tomorrow (at the earliest).”
Well, tomorrow has come and gone and I still haven’t responded. Let me post what Yglesias said (which has a link to what I originally said).
What are your thoughts on the point Yglesias is making? It seems to me, having read this several times, that he’s not making a point that is dramatically different from what he said when he raised this issue the first time. So I am not sure how to respond other than to repeat what I said earlier (which is just a bit silly). Then again, maybe I am missing something. Regardless, I need to end this post and get back to work.
– DJ
Oh, here is what Yglesias said two weeks ago. Again, comments are always welcomed.
I feel like Dave Berri’s missing my point here: “Now we have the argument that the value of Lewis should not depend upon position played. The numbers tell us that playing Lewis at power forward will cost Orlando rebounds. But we should ignore this fact and simply give Lewis extra credit for making an effort.”
No. This is what I’m saying. Suppose you have two players. One is Rashard Lewis, excellent small forward (according to Berri’s numbers) and average power forward (again, according to Berri’s numbers). Now you have a second player. Call him “Lashard Rewis.” Rewis puts up Lewis’ exact same numbers, but if his coach tries to insert him as a power forward he refuses to play. Which player is better to sign — Lewis or Rewis? Berri says it’s Rewis — Rewis will have a better position-adjusted Wins Produced number. I say — and basic common sense says — it’s Lewis.
In any situation where Rewis could help the team win by playing small forward, Lewis can do it, too. But some situations will arise (suppose your starting power forward has fouled out and your backup power forward sucks, while your backup small forward is an above average player) where Lewis is a more useful player to have on your roster. It’s true that teams employing Lewis do well to remember that he’s much more effective as a small forward than as a power forward (assuming that’s true) but it’s also true that it’s better — more useful to your coach and GM — to be able to “play out of position” with a modicum of success than to be totally useless.
That said, it’s slightly absurd to even discuss positional matchups within the Wins Produced framework because it doesn’t deal with defensive matchups at all. Is Player X quick enough to “downsize” and stay with his man? Is he tall and strong enough to “upsize” and not get pushed around? The Wins Produced framework doesn’t differentiate between (very useful) players who can guard multiple positions, and (unfortunate) players who defend two positions because they’re equally ineffective at both spots.
Wages of Wins Journal 
Tball
July 24, 2007
Is there any publicly available statistical information on the performances of the referees?
Jason
July 24, 2007
The notion that the league is in trouble seems a bit premature. The notion that all the fans will go away even if they suspect things are not entirely honest may not be true either.
Nielsen ratings for the NBA reg. season on ABC were a 2.2 in 2005 and 2006. This compares infavorably with “WWE Smackdown” (ratings in mid 2’s to mid 3’s over the same period) even though “professional wrestling” doesn’t even pretend to be a real competition anymore. Granted, I don’t want basketball to be come wrestling myself, but I think that the very simple conclusion that a single ref can spell disaster for a league doesn’t hold up to much scrutiny.
dberri
July 24, 2007
Jason,
Owen (and I) want your thoughts on the David Lee essay. You can comment on Yglesias if you like also.
Okapi
July 25, 2007
Isn’t Yglesia partly just disputing an implicit assumption in PAWSMIN that the replacement player is “average”? If a team has a far below average back-up power forward then the value of Lewis to that specific team is greater. Of course this scenario is still consistent with using PAWSMIN to evaluate players across teams.
dberri
July 25, 2007
Okapi,
I agree that Yglesias seems to understand why adjusting for position played is necessary. I would note that if you simply use the player available on the roster (as opposed to the league average) then I can increase the value of Rashard Lewis by simply employing Dave Berri or Matthew Yglesias. Since both of us are very bad players, Lewis would suddenly be worth what the Magic are paying him.
MT
July 25, 2007
I thought the responses on the Knickerblogger website were unusually devoid of ad hominem arguments and quite well thought out, with very few exceptions. In addition to Owen’s excellent job of pointing out the flawed premises behind the original post, such as using the wrong data, ascribing the wrong conclusions, using WP48 exclusively and ignoring the positional adjustment implicit in the proposed trade. I thought Ben and others did a good job as well pointing out the limits of WoW for comparing players who are used differently by their coaches. The Iverson for Miller experiment showed the probative value of the model when comparing players who are used similarly. But I am not aware of a real world experiment that validates the WoW analysis as applied to players who are used very differently from each other and given that I think the model overweights rebounding (which I think is more correctly viewed as merely the “finished good” of a defensive production line manned by 5 people, and thus should not be credited 100% to 1 of them) and aslo does not take into account the extent to which certain players are asked to score under significantly more difficult conditions than others,I would candidly not expect WoW to have superior predictive power in that application.
If there are typos in my comments, I apologize. The boxes for name, mail and website are obscuring the right hand side of the comment box. In the future, I will just paste comments from Word.
Mr. Parker
July 25, 2007
MT,
I present you the Dampier Dilema.
The Dallas
Mavericks had one of the all time great
seasons in NBA history. Their top rebounder
was Eric Dampier who rebounded at a rate of
17.8 rebounds per 100 opportunities. To
give that perspective that is about the career
rate of Shaquille Oneal. Dallas lost a playoff
series to an 8th seed while their best rebounder
played only 8 minutes per game throughout the
entire series. For the entire reg season
Dampier averaged 26 min per game.
Since Dampier has been with Dallas they have a
5-9 record vs. the GSW. In games where Dampier
plays more than 20 minutes Dallas is 3-3. The
point differential in those 6 games is
Dallas +4. In the other 8 games the point dif
is gsw + 39.
My suspicion is that Dampier gave up
his minutes because his coach was looking
for more scoring vs. one of the leagues
top scoring teams. The problem with that
line of thinking is that Dallas was a
more efficient scoring team with Dampier
on the floor.(114ortg vs. 110) To me that
clearly states that Dampier does not
hurt offensive efficiency while on the floor.
Even though he only averages 7ppg.
So not only do we have most fans with a clear
misunderstanding of the player values but we
also have men who have won coach of the year
misunderstanding player value as well.
Now, wp48 scores Dampier as a .191 or damn near
the perfect player.
3 players had minute increases about equal to
the 18mpg Dampier was now sitting. Their average
win score was approx .11.
We should not be arguing about who is better,
David Lee or Kobe Bryant. We should be
wondering why the Lakers aren’t trading for him
right this minute. This is Isiah Thomas we are
talking about isn’t it.
One more tangent. It has become my opinion
that Larry Brown is the greatest gm out there.
The man clearly understand the point that is
being driven home by the wp48 metric. He has
always denounced the type of players who score
low on wp48 and brought in the players who score
higher. One only needs to look at his tenure in
Philly. I can clearly picture Larry Brown
evaluating Isiah’s moves and saying to himself
“Predicted wins 29…SON OF A BITCH!” ala
Master Ralph Parker when he finds out the secret
message is dont forget to drink your ovaltine.
Its not wonder Allen Iverson only good(
championship level) year was with Larry Brown
Larry Brown acting as defacto GM. He understood
that you had to put great defensive Rebounders
around a guy like Iverson. Denver actually
has the potential to create an equally successful
team right now, but they would have to trade
Melo for someone who shoots less and rebounds,
assists, and steals more.
Mr. Parker
July 25, 2007
typos galore,
Here’s my message in short.
1. Rebounding is vastly underrated. One
only needs to rehash the Dallas vs. GSW
series to see that.
2. GMs around the league clearly do not
understand this except Maybe the bulls Gm
who seems to be overdoing it a bit
with his last 3 major acquisitions being
Thomas, Noah, and Wallace.
3. Why aren’t the Lakers trading for
David Lee right now? He is the best player
out there and with the Knicks roster
looks to be mighty available and mighty
productive. Trading him straight up for
Lamar Odom makes the Lakers a 50 win team
right off the bat with David Lee being twice
as productive as Lamar Odom. At least last
year’s David Lee. Im pretty sure the Knicks
would be willing to give up Balkman as well.
Then we’re talking about 55 wins.
4. Larry Brown is the greatest Gm out there.
Someone needs to higher this guy now. He has
a name and he has the undestanding to put
together a winner.
LT
July 25, 2007
tball go to covers.com for game level stats on refs or go to blogmaverick and read old Mark Cuban posts on refs using the link he provides.
Westy
July 25, 2007
As far as discussion of the ‘position played’ dilemma, I thought there were some good thoughts back on your post on that topic: https://dberri.wordpress.com/2007/07/09/adjusting-for-position-played/