A few days ago the Minnesota Timberwolves told their fans that they will probably not be contending for a title in 2011. Most people – except for Michael Beasley – would tend to agree with the T-Wolves. So although we are surprised by the honesty, we are not surprised with this assessment.
Such honesty, though, leads one to ask: Who else in the NBA should offer a similar confession? The beginning of an answer was offered by Arturo Galletti, who is currently working on a forecast for the 2010-11 season.
This past week he offered a very preliminary view. Essentially, he ranked each team according to how the team’s veteran players performed last year. In other words, he ignored rookies and injuries. He also did not consider how minutes could be allocated or how age might impact performance (and please read his post carefully for all of the details behind what Arturo did).
So the following table is not really a forecast of next season. But it does give someone an idea of which teams have any chance to contend. And of course, that also means we now have some idea which teams might want to follow Minnesota’s example.
The key column is in the middle (Adjusted Wins – Minutes and Available Wins). According to this view, only three teams – Philadelphia, the LA Clippers, and the Washington Wizards – rank below the T-Wolves. Again, rookies are not considered and the Clippers do have Blake Griffin. Nevertheless, one rookie is not going to transform this team into a title contender. So an “honest assessment” should be forthcoming from these franchises as well.
One could also add to this list the Knicks, Pistons, Raptors, and Pacers. These teams are not favorites to land in the playoffs. And you can’t win a title if you can’t make the playoffs.
Beyond this list we have teams that have better odds of making the playoffs. Of course, just making the playoffs isn’t good enough. In other words, every team that makes the playoffs isn’t a title contender.
But if we start noting the non-title contenders in the playoffs, then we might start expecting “honest assessments” from virtually every team in the Association. After all, Miami is the clear favorite to win the 2011 title. And right now, the list of teams that might prevent LeBron and company from being happy next summer is fairly short. From Arturo’s list, we might place the following teams on that list: Portland, LA Lakers, Orlando, and maybe Boston. We might also consider Chicago, Dallas, and San Antonio. After these teams, though, Arturo’s first look is less optimistic.
It is very important to remember that this is a first glance at these teams, and much has been left out of the analysis. But I do think we can see that even in September, Minnesota is not the only team that can promise their fans that a championship parade will not interfere with city traffic next summer.
And that is what makes basketball very different from the other major sports. In the other sports, fans can believe until the games start being played. In basketball, we know (again, except for Michael Beasley) that the season for some teams is over before it begins. Yes, it is sort of sad. But I think that is an honest assessment.
– DJ
Wages of Wins Journal 
tywill33
September 18, 2010
Arturo is “doin work” as Spike Lee would say. That is some great stuff.
I can’t quite translate, though. If I’m right, it appears he’s saying the Twolves are looking at a 10 game bump, which is what I also think (and have run my mouth predicting). Although the team made some weird moves (throwing a bunch of money at Darko Milicic), they did a good job of adding by subtraction (the Wolves had a boatload of negative win producers).
It also looks like he has the Milwaukee Bucks, despite their ballyhooed offseason moves, remaining in the same basic NBA social class. Unfortunately, I keep getting that result as well. I think that may be where the safe money belongs.
entityabyss
September 18, 2010
Would it be safe to bet that the washington wizards will be the worst team next year? It doesn’t seem like any other team will be close.
some dude
September 18, 2010
“Nevertheless, one rookie is not going to transform this team into a title contender. ”
Magic Johnson says “hi.”
nerdnumbers
September 18, 2010
Some Dude,
Arturo and Andres from almost four months ago say “Hi”
https://dberri.wordpress.com/2010/05/27/game-changing-draft-picks-since-1977/
dberri
September 18, 2010
If Arturo’s numbers are correct, the Clippers are at 14 wins before Griffin. To be a title contender, you need about 50 wins. So Magic was worth 36 wins? Didn’t he have Kareem on his roster?
nerdnumbers
September 18, 2010
DJ,
Following Some Dude’s logic. . . Kareem was probably overrated as he just stole rebounds from his team by playing poor defense. He only took close shots that were a result of superior players like Magic drawing all the attention to him. In short Kareem wasn’t the type of player you could build a team around. In fact due to his short comings I think he was probably worth negative wins. Magic’s rookie season was probably worth 100 wins easy (which is why he is better than Charles Barkley)
some dude
September 18, 2010
Way to make it glaringly obvious you’ve barely ever watched Kareem play. He only took close shots, eh? Cheated over for rebounds, eh?
Please, don’t compare Kareem to marcus freakin’ camby lest you want to continue to say foolish things.
entityabyss
September 18, 2010
some dude, we’re heading into the 2010-2011 season. The wins produced method has been effectively for years now predicting team wins and losses. Win shares hasn’t. PER hasn’t. APM hasn’t. At some point you have to, rather than constantly bash this work, consider it.
Why should it matter that the people you believed were the best aren’t considered the best? What should matter is the accuracy of this metric. I’ve looked at this metric, PER, win shares, APM, and other metrics and this one just gets it right. Their player ratings stay consistent over time, and their methodology is from sound logic (wins are dependent on efficiency differentials).
They’ve proven that shooting efficiency is not based on the amount of shots taken, but rather on skill and shot selection. They’ve shown that if you remove a high volume scorer, those extra shots will be taken without players’ shooting percentages going down. They’ve shown that they don’t overvalue rebounding, rather they value shooting efficiency more than other metrics.
Why does it matter so much if marcus camby is rated highly if the evidence shows that the metric shows that it’s right. That’s when you start looking for names rather than results. Go ahead and do that. When golden state wins 50+ games this year, and the wizards end up a bad team, you’ll know why.
Anyways, I was asking earlier. Would it be a safe bet to say that the wizards will be the worst team in the league?
Chicago Tim
September 18, 2010
entityabyss — The Wizards might not have the worst record if a number of things go their way. Arturo has examined the possibilities in detail at his website.
jbrett
September 18, 2010
Really, though, the Wizards’ best hope is to stumble upon the Fountain of Youth and fill up a couple of really, REALLY big bottles–’cause that whole time machine thing is scientifically implausible:-)
some dude
September 18, 2010
“Why does it matter so much if marcus camby is rated highly if the evidence shows that the metric shows that it’s right. That’s when you start looking for names rather than results. Go ahead and do that. When golden state wins 50+ games this year, and the wizards end up a bad team, you’ll know why.”
Because the 3rd best player in the league wouldn’t consistently be on a team that isn’t even a 2nd round contender for 10 years in a row. It doesn’t make sense.
I like WP48 but it isn’t that accurate. Is it more accurate than other measures? yes, but it’s not accurate enough to assume it doesn’t need to be tweaked. It does. If the good professor doesn’t realize this, someone will come up with an even better statistic (the SPM people claim they already have, though I’m not convinced). The measurements are good, but could be better. That’s my point.
And as mentioned numerous times, defense is not properly attributed (and this is where Camby leads in the overrated category). When will the WP48 proponents admit that the metric doesn’t measure everyone accurately and is still in need of tweaking.
And the Warriors will not win 50 games. This i can give you a guarantee on. And no one needs WP48 to convince them that the Wizards could be the worst team in the league next season, I’m pretty sure a lot of people think that anyway.
some dude
September 18, 2010
sorry for a 2nd post in a row but…I keep asking for a WP48 differential number (one WP48 vs their counterpart WP48) number because I think this might be a more accurate reflection of a player’s value.
By counterpart, I refer to the actual player they defend, not the opposite player at his position. basketball prospective tracks the oppositions production and quality right now, but i doubt its accuracy.
Chicago Tim
September 19, 2010
“In the other sports, fans can believe until the games start being played. In basketball, we know (again, except for Michael Beasley) that the season for some teams is over before it begins. Yes, it is sort of sad. But I think that is an honest assessment.”
Fortunately, this blog has me excited to see whether all these predictions come true. Why else would I follow the performance of the Washington Wizards? It also has me excited about great players on bad teams, like Kevin Love on the Timberwolves. And yes, I’m excited to see whether the Heat are simply better than everyone else this year, or maximize their potential (read, play Bosh and center and James at point) and challenge some all-time records.
Of course, I’m also very excited about the Bulls, who may have as good a chance as any team other than the Heat to win a championship (still not a great chance, though). But even when the Bulls were bad, I still followed the progress of their young players to see whether any of them could turn into stars, I still watched Ben Wallace closely to see if he would justify the Bulls’ unconventional faith in a non-scorer (unfortunately, he seemed to be bothered by nagging injuries the whole time). I still found the Bulls/Celtics series resulting in a record number of overtimes thrilling, even if it was only competitive because Garnett was out with an injury.
reservoirgod
September 19, 2010
some dude:
I will actually be attempting to do that this year for the Miami Heat. It wasn’t an option the last two yrs because I wasn’t willing to pay for league pass just to watch a mediocre team w/ no shot at the title. I’m not sure how successful it will be, though. For example, if Derek Fisher’s job is to run out for easy transition buckets vs. OKC and he doesn’t box out Russell Westbrook who ends up getting an offensive rebound, is that boost to Westbrook’s Win Score really Fisher’s fault? Who gets the counterpart credit for turnovers forced by double teams? Who gets counterpart credit on teams that switch all pick & rolls like the Hawks did last season? I better stop before I discourage myself from attempting this before the season starts.
Cool_Hand_Luke
September 19, 2010
If that’s Fisher’s job then he would presumably get credited for assists/points on the fast break. He also isn’t covering Westbrook and should be duly penalized. If my job is something other than boxing out my man then my value decreases some amount for failing to do that, and hopefully whatever I’m doing instead is worth it.
What could be considered is counterpart production vs expected counterpart production (i.e. if Rondo is expected to perform at ~.300 WP48 and he performs at .250 then you’ve done well). People play different opponents and should be evaluated on how those opponents performed compared to their normal, not compared to the average (which is what WP compares you to).
A possible problem is double counting. For example, the value of a possession is ~1 so a loss of possession is a swing of two points in value (+1 expected value to -1). As it is, WS assigns 1 point of that change to the player who makes the TO and another to whoever steals the ball. If you count for opponent stats you’re credited both for allowing the steal (-1) and getting a TO (-1) while your opponent is credited for getting a steal (+1) and causing a TO (+1). This credits a change of 4 points to a swing of possession only worth 2 in reality.
Michael
September 20, 2010
entityabyss, PER is not adjusted to team defense like WP is. I don’t know why you would expect it to predict team wins (retroactively) without this adjustment.
dberri
September 20, 2010
Michael,
Even when you add in team defense, PER still doesn’t explain wins as well.
Michael
September 20, 2010
True. It’s around 80% I think.
dberri
September 20, 2010
If you use the actual team adjustment used in WP, then it is 60%. If you let the defensive variable take any value, you get to 80%. The results, though, end up being a bit odd. And most of wins end up being explained by the defensive variables (since PER by itself only gets you to 30%).
some dude
September 20, 2010
RG: You have to make a sacrifice somewhere. Look, WP48 doesn’t take into account Camby leaving a man open in order to anticipate a rebound in order to pad his rebounding stats, but he receives the WP48 benefit. So if Fisher is on WB and WB gets the Oreb, it hurts him regardless. The same issue applied when a defender helps because he has to help due to his teammates poor D and his man does better.
This is why i said “more accurate,” rather than just “accurate.”
As for paying a price, synergy tracks points per possession. Perhaps if you could change this into a WP per possession (now that is a small number), perhaps things will get clearer.
Truth is you will always have to make a mental adjustment. The goal is to find a way to make that as small as possible. a differential might make the picture a little clearer even if it’s still flawed (but hopefully less flawed).
fricktho
September 20, 2010
I’m all for the most accurate measure, but at some point the nerdness has to end. Instead of telling someone how to tweak their formula how about do the work yourself. What some dude is proposing is obscenely labor intensive. Who has the time? Your talking a level of nerd-dom that is rarely ventured to, and only by the nerdiest of nerds. The foundation has been laid if you ask me. Now it is up to others invested in it to take it further if they want, not the founder. I’m a big fan of this measure because as some dude states “Is it more accurate than other measures? yes”, which is enough said.
Leroy Smith
September 20, 2010
Some Dude, I’m not sure I get your argument. Help me out. I’m thinking if Camby leaves his man and gets a rebound then it was worth it because the other team doesn’t get a second chance at scoring and Camby’s stats are rewarded by WP. But if Camby misjudges the bounce and his man (unboxed) gets an easy put back then the other team scores and Camby is punished by WP because he gets one less board and the other team scores.
Same thing with high steal guys. If player X goes for a steal and forces the turnover, then WP rewards him because he keeps the other team from scoring. But if he misses then the other team scores or his man gets an assists/bucket and WP punishes him because his team looses or allows a high FG%. I think WP does a good job of capturing these negative and positive contributions.
By the way I agree with you that WP must somehow be overrating Camby, but guys like Jordan, Shaq, Duncan, KG, Magic are the ones that dominate historical WP top performers lists. Any metric that ranks MJ and that Crew near the top is worth looking at, but WP is the only one that can tell you almost exactly how many wins team X had in year whatever after using those old box scores. Baring injury, it is also the nearest thing to predicting future wins as well.
fricktho
September 20, 2010
some dude, to me, is basically saying every single play needs to be analyzed to figure out who is responsible for giving up the basket to better incorporate a defensive rating into the equation. it’s a borderline futile effort to do such a thing because it is so difficult to do. i could see if you were paid by an nba franchise that valued this information to the extent they would pay for it that it might be worth it, but for even the nerdiest of fans it seems pointless to go to that extent. no problem correcting me if i’m wrong.
in the scenario of camby – if he doesn’t rotate and leaves his man open and he makes the bucket camby isn’t responsible because he is not guarding him according to the formula. it’s his responsibility because he was supposed to rotate so it should go against him. now if the shot goes arrant then yes he made the right decision, but if the shot goes in and he didn’t rotate then he made the wrong decision. since a wide open shot has a better chance of being successful you would have to say camby made the mistake in this scenario. it’s nit picky.
Italian Stallion
September 21, 2010
Camby rates so well because he’s such a good rebounder and rebounding is given a lot of weight relative to other factors in this model.
This has been debated to death.
Italian Stallion
September 21, 2010
Individual defense will always be a tricky thing to measure, but IMO work still needs to be done in other areas.
I still have trouble with the following two 2 things. (I’ll give extreme examples to make the point better.
Assume a team of all average efficiency scorers.
Let’s say that team trades a player that averages 25 points per 36 minutes for a player that averages 10 points per 36. Obviously, the team is not going to score 15 less PPG because of that trade. Each of the 5 players on the court will get more shots that they used to and make up a lot of 15 PPG.
This model assumes they will make it all up without any impact on efficiency.
While this is an extreme case that rarely comes up in the real world, I suspect the team would not remain as efficient and fall short. At the margin things like this should be considered between players that are otherwise rated similarly.
Now lets assume a team has a PF that gets 12 rebounds per 36 minutes and he is traded for a PF that gets 6 rebounds per 36.
This model assume the complete opposite of the assumption with scoring. It assumes the team will get 6 fewer rebounds. While I don’t doubt that trading a great rebounder for poor one will have a huge negative impact, I suspect that a handful of the 6 would be made up by the other players.
entityabyss
September 21, 2010
Italian stallion, the evidence shows that diminishing returns do happen, but not to that extent. Maybe a rebound would be obtained by the other players, but not 6. It would be a big difference. As for the shooting, if you trade an above average (60% true shooting) who scores 25 for a guy who shoots 60% ts but only scores 15, the team would get worse, because the average player on that team likely shoots less than 60% ts. But if u trade some who shoots around 54% ts and scores 25 for some1 who shoots 54% ts and scores 15, you should be fine. That’s average. If you have guys on the team who shoot below that, they’re below average scorers who probably have a negative wp in scoring and will get more time and will hurt the team more. Really though, it all makes sense. Rebounding may seem over valued but it isn’t as has been explained to death. Scoring percentages are just more important. Look at win score compared to nba efficiency. Same rebounding.
Michael
September 21, 2010
Out of interest, does anybody know the probability of a team acquiring a defensive rebound after a miss? What about an offensive rebound?
To hazard a guess I would say that the latter is a lot lower than the former.
Italian Stallion
September 21, 2010
entityabyss,
I agree that all 6 rebounds wouldn’t be made up, but I could easily see one or two being made up. At the margin that might make a difference between two players that are rated fairly similarly overall, but one is the better rebounder and the other is better at other things.
I agree that changing the efficiency of the players would matter way more than their scoring ability, but I am still on the fence about some of these “average efficiency moves”.
If you gave me a choice between two players with a TS of 54% and one was averaging 25 PP36 and another was averaging 10 PP36, I’d take the former without hesitation and feel confident I could tweak his game and improve his efficiency. I would feel less confident I could get 20 points from the other player and maintain his efficiency if I asked him to do a lot more (with some exceptions).
In the grand scheme of things I don’t think it matters all that much because sweeping changes like this aren’t typically made, but it’s another one of those cases where all else being equal, I’d take the higher usage scorer.
arturogalletti
September 21, 2010
IS,
I agree that making the higher usage score more efficient by making him take less shots can be done (but replacing the scoring provided by a high usage,low efficiency guy by sitting him down spreading the ball around is also easy).
By my estimation, there are three critical skills in basketball: Ball handling, Rebounding & Shooting effectively. Given that the effects of Ball handling and shooting relate to how you spend possessions they can be mitigated somewhat by manipulating player possession usage but Rebounding is a different beast. Rebounding is about possession creation per minute on the court, size and a particular knack. Rebounding given the correlation to size is the hardest thing to actually replace.
ilikeflowers
September 21, 2010
For the lulz,
Noah is a promising young big man, but by no metric is his production even close to Carmelo Anthony’s. Many have called Anthony the best pure scorer in the NBA.
Henry Abbott
Technically true even using wp48 since Noah is clearly superior, but just as clearly Abbott means that Anthony is much better than Noah.
C’mon Bulls! Do your part! Get out of Miami’s way! Trade Noah! And please do throw in Deng to boot.
arturogalletti
September 21, 2010
ilf,
*Shakes head*. I thought I covered this already:
http://arturogalletti.wordpress.com/2010/09/11/four-leaf-clovers-and-joakim-noah/
Ridiculously stupid trade for Chicago.
todd2
September 21, 2010
Missed fg’s create rebounds. A great perimeter defender can boost a team mates rebounding numbers.
entityabyss
September 21, 2010
Todd, this is true. More misses equal more rebounds, but the effect is not that big. If it was, you’d notice a drastic change in players’ rebounding numbers by year.
some dude
September 21, 2010
“Some Dude, I’m not sure I get your argument. Help me out. I’m thinking if Camby leaves his man and gets a rebound then it was worth it because the other team doesn’t get a second chance at scoring and Camby’s stats are rewarded by WP. But if Camby misjudges the bounce and his man (unboxed) gets an easy put back then the other team scores and Camby is punished by WP because he gets one less board and the other team scores.”
He leaves his man open to get the rebound rather than defending him. The increased probability of a team rebound by Camby’s team because Camby left his man is far lower than the increase likelihood of that player now making the shot.
I watched first hand Camby do that on a near last second shot and the player nailed the shot because Camby turned around as he went into his shooting motion. My jaw dropped.
Anyway, Camby is not hurt by WP48 if the guy hits the shot because Camby cheated for the board.
Furthermore, Camby takes rebounds away from teammates. I routinely see him tip it away from them when they’re clearly in position with no threat and 99% of other players in the league let their teammates get it in this spot because the tip increases risk.
“in the scenario of camby – if he doesn’t rotate and leaves his man open and he makes the bucket camby isn’t responsible because he is not guarding him according to the formula. it’s his responsibility because he was supposed to rotate so it should go against him. now if the shot goes arrant then yes he made the right decision, but if the shot goes in and he didn’t rotate then he made the wrong decision. since a wide open shot has a better chance of being successful you would have to say camby made the mistake in this scenario. it’s nit picky.”
Camby rotates all the time. He’s an overhelper! It’s his own man where the problem comes from. He’s a solid help defender, even though he does it too much, but his man to man D is highly overrated. Much like Birdman (not to pick on Denver). Someone like Ben Wallace, on the other hand, is superb at both (or was) but his problem stem’s on offense quite obviously.
“Italian stallion, the evidence shows that diminishing returns do happen, but not to that extent. Maybe a rebound would be obtained by the other players, but not 6. It would be a big difference”
The Pistons let Big Ben go (years ago) and ended up with the exact same rebound rate. What they lost with him was defending without fouling.
“By my estimation, there are three critical skills in basketball: Ball handling, Rebounding & Shooting effectively”
I think passing is just as important. People only think of assists when it comes to passing, but it’s not just that. Entry post passing is one of the most important skills for a guard/wing. Just ask Dwight. Unless you consider that “handling.” The thing is, entry post passing can’t be directly measured.
arturogalletti
September 21, 2010
sd,
You’re right, Should have said : Ball handling and passing , rebounding and efficient shooting. Entry post passing can be seen at least partially in assist and turnover numbers.