Hamilton to Dallas?
by Kevin Sawyer on Feb 14, 2010 12:58 AM EST
Email
PrintUm, maybe. After the Altman-esque All-star Saturday, I thought we all could use a reason to get drunk and surly.
Today, the Dallas Mavericks and Washington Wizards announced a seven player deal that will send Josh Howard to the Wizards in exchange for Caron Butler and Brendan Haywood. Drew Gooden and some other guys are in it for numbers reasons, but that is the gist of the deal.
On it's face, a Howard for Butler swap doesn't change the picture much, and even the inclusion of Haywood doesn't improve Dallas' chances of getting out of the West. That's where Detroit comes in.
For months, Dallas has seemed a likely trade partner should Detroit make the decision to move Rip Hamilton. The Mavericks have a hole at the two, and what essentially amounts to a perfect fit expiring contract in Erick Dampier. They also have a closing window of opportunity with their core of Dirk and the Jasons, and an owner who is willing to spend money during a recession.
Alas, in addition to having a nice contract, Dampier happens to be a valuable player. Trading him for Hamilton would have left an even larger gap in the front court. Now Dallas has Haywood to fill the gap, and with better defense to boot.
This has all the hallmarks of an "other shoe to drop" trade. As it stands, the Mavs are doubled up at the Center position. The trade was made early enough that there is some breathing room for negotiations.
Hamilton hasn't played particularly well this year, but, then, neither has Butler. If Cuban wants to win a championship with this squad, he is going to have to take a risk.
The Mavericks are solid enough that Hamilton could ease back from the injuries that still seem to be plaguing him, saving his best work for the playoffs. Hamilton's playoff experience has to be alluring to a team looking to get over the hump.
As it stands, acquiring Hamilton might be the last big play Cuban has left. He could make a push for Andre Iguodala, but that would mean taking on Dalembert's contract, which makes it about $30 million more for Cuban to deal with them, even if the two sides can Memphis away the contract discrepancies. They could pursue Ray Allen, but what do they have that the Celtics need?
Of course, any number of things could happen. Having spent money to get today's trade done, Cuban might be tightening his purse. Another team could step in to offer a star player for nothing. Joe Dumars might ask for too much (anything beyond Dampier and a first rounder is too much, and even the latter should result in a blinking match) or decide he wants to (ugh) give the existing team time to gel.
But, from a Pistons perspective, this is certainly the most eventful news of all-star weekend, and something to think about as we watch a second straight night of missed windmill dunks.
67 comments
|
0 recs |
Do you like this story?
Comments
I think Dallas is done
This is a big enough move that Dallas likely feels they have upgraded at the 5 and 3 (although how will Butler and Marion work out?). Any further change might be too much for the time to, uh, cohere before playoffs. I think it’s safe to assume that if there is a trade partner for Rip Hamilton, Dallas has just taken their name out of the running.
Of course, I’d kill to be proven wrong.
However, I think this trade is roundly significant, moreso than what it means for Dallas and Washington. This is a huge trade, with a lot of player movement and some big names involved. If anything, I think this trade has broken the seal for the punch drunk NBA GMs who are hoping to take a piss before the deadline. With just a few days left, I’m thinking this is the first step before the rest of the GMs pony up for the “me too” line. Hopefully Joe can make us proud. If these last two years are any evidence, however, its safe to assume that Joe operates solely to not make us happy.
Would you like me to give you Hubie Brown's home telephone number so that you can actually call him yourself to confirm that he is someone who knows me?
by Mike Payne on Feb 14, 2010 1:12 AM EST reply actions 1 recs
It depends on where Cuban thinks his team is. He has seen some great Mavs teams go down in flames. If he wants to go over the top, he has to make another move. This is not a “preparing for matchups” kind of team, yet.
honestly
Dallas has two all star SFs and an all star SG they prefer to bring off the bench. Will Terry start after this, and will Butler come off the bench? They now have a logjam, but I doubt they’ll trade Marion to balance out their roster. In terms of the primary pieces of this move, Howard and Butler, this move is pretty lateral as talent is concerned. If they were to make a move for another SG, you’d think they’d already do it.
Dampier’s contract, as you pointed out, is intriguing to the trade market, but I’m not sure if they’d actually do anything.
Would you like me to give you Hubie Brown's home telephone number so that you can actually call him yourself to confirm that he is someone who knows me?
Caron would start over Terry
That’s the only way it makes sense to me(albeit I don’t see how they have 2 all star SF anymore, both Butler and Marion are at end of their prime). Think about it, Butler has a career 85% FT%, that shows that he’s a very good midrange shooter. Combine that to the fact that he’s 6’7" and 230 pounds, you have a very big starting 5 (minus Dirk, of course).
Besides, Butler is a player that needs to get into rhythm whereas Terry is instant offense. With how short Terry is, it just doesn’t make sense to start him.
The only question is if Caron is still quick enough at his age to guard a lot of the SG that love to utilize screens, like Hamilton and Allen.
by bearded thundar on Feb 14, 2010 8:38 AM EST up reply actions
Logically, IME, the move seems exactly like a “match-up” trade.
Kidd/Fisher, Caron/Kobe, Marion/Artest, plus Dirk and two 7ft defensive centers, and Terry coming off the bench. How is that roster not specifically designed to match-up with LA?
It’s also matched up fairly well to Denver, I think, the only team standing between them and conference finals.
by Apocalyptic0n3 on Feb 14, 2010 9:20 AM EST up reply actions
I personally like the trade
Dallas got a steal in this trade. Even if you aren’t a Caron Butler fan, they pretty much got Brendan Haywood for free. I think it does make them better, at the very least addition by subtraction.
Josh Howard just isn’t the same player he was back in 06-07 after all those ankle injuries. It certainly didn’t help that he rushed his recovery and ended up re-injuring it several times, or that he injured his other ankle in the playoffs last year. Any time you can get a 10 and 10 defensive center, with a good contract, for a player that has clearly declined, you make the trade. Dallas was going to let him go over the summer anyways, so I think it was a great move on their part, to get something for nothing.
Dallas lost lost a worthwhile player Drew Gooden, who was quietly having a good year. He was averaging 15 and 12 per 40, so when you take into account his small contract, Dallas certainly got their money’s worth out of him. However Haywood is a much better player, so they still got the better end out of this deal.
You know, the funny thing is, with their number 5 pick, Washington could have had Brandon Jennings, Stephen Curry, Ty Lawson, Omri Casspi, DeJuan Blair, JJ or Darren Collison, all of whom have turned out to be solid, if not very good future rotation players. Washington was supposed to be a contender this year, and now they’ve dumped everything of value but Mike Miller. So as much as we fault Joe D on his drafts, just take a look at some other GMs blunders (especially Orlando, who the hell in the right mind would pay $40 mil per year to Richard Lewis and current day Vince Carter?).
by bearded thundar on Feb 14, 2010 9:53 AM EST reply actions
Good Point
We get so caught up in Joe’s fault simply because we pay much more attention to him. If we take the time to look around, it becomes clear that the NBA is filled with GM’s who tend to suck at their jobs.
I think this is much more than a lateral trade.
Howard has been playing poorly and has been injured frequently. Heywood’s having a career year — practically averaging a double double and more than 2 blocks! Dallas is clearly better after this trade, but I’m not sure how much.
There’s a logjam at the 3, but it would seem Butler could play at the 2 if necessary. Still, it’s not ideal, especially against some of the smaller, quicker teams in the West. It would seem to me that they still might need to do something at the SG position to complete the roster.
Butler is very versatile, having played the 2-4 successfully in his career. They’ll probably start Kidd-Butler-Marion-Nowitzki-Haywood and have Butler float around the rotation, allowing for Dirk and Marion to get more rest down the road. They also have two great backup combo guards in Beaubois and Terry, so Butler floating around would work very well I think.
You have to admire what Nelson has done with that lineup. The only players in that entire lineup that are one-position players are Kidd and Dampier. Everyone on that team can play two-three positions very well and that will really help them match up against the Lakers and Nuggets or, perhaps, Cavs, Magic, and Celtics. When someone doesn’t work playing against Kobe, Artest, Gasol, Shaq, Dwight, Lebron, Pierce, etc., they can just move another player to the position and it should work flawlessly.
by Apocalyptic0n3 on Feb 14, 2010 10:19 AM EST up reply actions
Getting Dampier would be a mistake, I think. We would likely start him and Ben together. His contract is only expiring if he averages less than 30 minutes over the 72 games he plays the most minutes in. He’s averaging 26 right now. We would either have to take on an $11 million, second-to-third tier center or play him little and let him expire, which does absolutely no good for us.
by Apocalyptic0n3 on Feb 14, 2010 10:14 AM EST reply actions
Why doesn't Detroit do that?
I don’t get why Joe just hands out guaranteed money.
Wouldn’t it make much more sense to give a 3 year deal with bonus clauses (for scoring, fg%, etc) and a team option rather than just a straight up 5 year deal? I’m pretty sure Ben Gordon would have been willing to sign an $8 mil /year with bonus clauses raising it up to $12, that way we wouldn’t be massively overpaying for years like this when he’s injured or has little to no impact on the court.
Also CV would probably be much more willing to play defense and rebound if he got an extra $1 mil for averaging at least 10 boards per 36.
by bearded thundar on Feb 14, 2010 10:39 AM EST up reply actions
I think we’ll see a lot more of that in the new CBA. But that obviously doesn’t explain Joe’s lack of doing so.
Honestly, that’s the reason why players get lazy in Detroit. Joe is too much of a friend instead of a boss, and is fairly notorious about taking his players’ word over coaches. I’m sure if our previous starting five’s respective deals had team options for the final 2 years and they got said current money due clauses for production, we’d be seeing a much different Pistons team right now. Probably one that still had Sheed and Chauncey (perhaps Big Ben, not sure if Flip would still be here).
There’s no way a player of Tayshaun’s worth gets a guaranteed $10 mil deal in any other sport, especially not the NFL. That’s why the current NBA is so flawed, people just hand out way too much guaranteed money, which leads to players like Boozer who have miraculous contract years and then get lazy afterwords.
by bearded thundar on Feb 14, 2010 11:31 AM EST up reply actions
Those types of incentives are pretty rare
Keep in mind that this is part of an extraordinarily generous compensation package. Joe D. does need to end his love affair with the player option.
There is no way Damp would push his average above 30 minutes playing for Detroit.
by Kevin Sawyer on Feb 14, 2010 1:31 PM EST up reply actions
Maybe it’s just that hard to get players to want to play in the D. Look how empty the Palace is now since we started sucking (and economics but I seriously paid 20 bucks for 9th row seats). Everyone wants to play in Miami, LA, and other big market teams that are contenders so we have to throw more money at them.
Even though I’m a hardcore Pistons fan, if I was a player I would definitely look at other options first. Awesome weather and tax breaks are both really nice incentives to make a home some place.
by garrettelliott on Feb 14, 2010 5:40 PM EST up reply actions
Dampier
Do you people realize that Erik Dampier sucks? He is no better than Nazr Mohammad. I’d rather keep Rip and have 6 SGs. Detroit can not sell low like this. You want to trade Rip target guys like Jamison. He is available. We could us a PF. Dampier???? No wonder why you are not a GM. You seem to be one of those guys who likes to trade for the fun of it.
by Steve Laukonis on Feb 14, 2010 12:41 PM EST reply actions
…are you Keith Langlois in disguise?
by Gabe F-B on Feb 14, 2010 1:06 PM EST up reply actions 3 recs
Don't think Dallas has to trade Dampier
They only have two centers on their roster, Dampier and Haywood. Then I think it’s more than plausible that Butler and Terry can coexist with Terry coming off the bench and Butler starting at the 2.
Also, don’t forget about Stevenson. When he’s playing well, he’s a pretty good, tough defender. And to get him minutes I can see them playing Jason Terry at the point guard spot a little.
It’ll be interesting, because this team will be big at all positions. As Piston fans we keep hearing the NBA is trending smaller, but I don’t necessarily agree with that.
DeShawn Stevenson is the worst player in the NBA
Literally. He shoots 28% from the field.
I really don’t see Butler at the two. He relies on his speed to be effective offensively, and generally isn’t going to keep up on defense. He’ll get amped to guard Kobe. JR Smith? Not so much.
Right now, instead of being 13 wins worse than the Lakers, they are probably 10 or 11. That’s not close enough to have a shot.
by Kevin Sawyer on Feb 14, 2010 1:50 PM EST up reply actions
I think you're a bit harsh
This trade makes them better than just 3 wins, if only because of their improved frontline. Haywood will make a big impact on their defense, I mean, after Dampier what other defensive bigs did they have?
Just look at the Thunder, Nenad Kristic is a decent to mediocre player, but the impact he’s had on their defense is huge. KD isn’t the only reason why they’ve gotten much better this year.
by bearded thundar on Feb 14, 2010 1:56 PM EST up reply actions
Krstic played 46 games for OKC last year. I think Sefolosha’s presence is more impactful. But if Sefolosha adds wins with great defense on the wings, Butler certainly subtracts with lousy defense.
Taken together, those two only account for maybe 3 Defensive win shares over a replacement player, and they are replacing a combined 45 minutes of squat. Haywood is replacing minutes from some combination of Dampier Drew Gooden, and maybe Dirk. .
re: Durant. When your most athletic player plays 40 mpg, and finally learns how to play defense, in addition to improving on the offensive end, it makes a very, very big difference.
by Kevin Sawyer on Feb 14, 2010 2:12 PM EST up reply actions
Stevenson does suck.
But I think Dallas is a lot closer to the Lakers than before. They have multiple guys who can guard Kobe in Butler, Marion and Kidd (he’s been really good at guarding Kobe historically). Plus, Dirk is actually a good match-up for Pau because he’s a similarly quick and fluid and 7ft tall. Marion can also be moved onto Odom, which is about as close an athletic match as any team will have againt him. Haywood and Damp are both big and long enough to bother Bynum. Off the bench, Terry is a difficult match-up for LA, as the triangle favors bigger, stronger guards, who typically have a hard time staying with small, quick guys. Finally Kidd dwarfs Fisher and the rest of LA’s PG’s production, so the Mavs will have a big advantage at that position.
Talentwise, though, they aren’t close. Sure, their personnel might be well suited to stopping the Lakers (although the reverse certainly applies), but they are overmatched.
And what if their first round match up is against Utah? Or San Antonio? At OKC? Matchups are more or less irrelevant if a team is starting without home court advantage.
by Kevin Sawyer on Feb 14, 2010 2:25 PM EST up reply actions
It’s a gamble, but it’s been the Mavs M.O. in the past to specifically design their roster to beat their main Western Conference rival. Their original contending team built around Dirk was pretty much created specifically to beat the Spurs. It worked, but they had the bad luck to then get screwed in the finals.
I don’t think the talent disparity is overwhelming. LA’s upside is higher, but they’re also going to be relying heavily on Derek Fisher (old and not very good), and Artest (crazy) in the playoffs. So they’re not exactly a sure thing to put it all together.
Why is Fisher still starting
You might as well start Magic, his lateral quickness is absolutely atrocious. There’s no way in hell he can guard Mo Williams, Chauncey, Beaubois/Barea or D-Will. Not to mention he’s shooting only 39%.
I place a big reason why they aren’t as dominant this year (after Kobe’s injuries, of course) on Fisher’s shoulders, he should be a third string guard for leadership at best, not starting.
by bearded thundar on Feb 14, 2010 2:48 PM EST up reply actions
Dirk is fantastic, Butler is a great theft/addition
Dallas doesn’t have a Kobe.
It remains to be seen what kind of potency Bryant will bring to to the post season though, just given his injuries. There’s at least two that are slowing him down right now. I would’ve sat out All-Star Weekend as well.
Something about Dallas, no matter who they add, just leaves me with a sense that they cannot seal the deal, even if they get to the finals. I’d like to see it happen, though.
"I didn’t even know Elvis was from Memphis, I thought he was from Tennessee." — Drew Gooden.
Ehh
Dirk isn’t a winner. He still doesn’t rebound or play defense. He’s a pure scorer and that’s it. Haywood will help mask this quite a bit, but there’s no way he’s going to slow down Pau Gasol with his usual brand of just standing around.
I just don’t think Dirk is the right player to build around for a title. His teams always do great in the regular season but then end up accomplishing nothing in the post season. You could argue that they were robbed against Miami, but then again so were the Pistons. If Wade didn’t go into Le-I-love-free-throws mode, the Pistons would have surely beaten the Mavs in the Finals regardless. And even still, you could make the case that SA was robbed because the league didn’t want a Spurs-Pistons rematch.
by bearded thundar on Feb 14, 2010 3:12 PM EST up reply actions
Nah, Dirk’s consistently put up superstar production in the playoffs. The “Dirk isn’t a winner” meme is a bizarro cousin to the “KG isn’t a superstar” one. They’re both great players who have performed like great players in the playoffs. I assume it’s because they’re both historically unusual. But yeah, talking about Dirk, any guy who has career playoff averages of 25.5 pts, 11 rebs, 2.6 asts, and only 2.2 turnovers, while having a superstar level usage rate of 26%… That guy is a winner.
What about opponent PER
That’s the stat I’d like to see. I already mentioned he’s an amazing scorer, but his teams still play zero defense.
Just take a look at the Denver series. The lowest they held the Nuggets to was a pathetic 106 points. You can’t win when you’re opponent is putting up well over 110 a night against you.
by bearded thundar on Feb 14, 2010 3:30 PM EST up reply actions
First of all, opponent PER is useless.
But, yeah Denver was a nightmare match-up for Dallas because Nene is too quick for Damp to stay with, Billups is just as strong as Kidd, and the Mav’s didn’t have anyone to match-up with Carmelo (now they have Marion and Butler).
But seriously, last year in the playoff, these were Dirk’s averages:
26.8 pts, 10 rebs, 3 asts, and had a TS% .635%(!!).
There’s just no way a player that productive can logically be blamed for his team losing. It just doesn’t make sense.
What about Steve Nash?
He’s always extremely productive on the offensive end, and yet every time the Suns have gone up against a decent defensive team, they’ve lost. The 08 team was built specifically to combat the Spurs (hence Shaq trade), and yet they still lost in 5 games.
I’m not denying that Dirk is an offensive juggernut, I’m just saying there’s a reason why his teammates jokingly call him Irk.
If you look Paul Pierce and Ray Allen, 08 was really the first time they truly committed to defense. The result was their first championship. Not coincidentally, when KG went out and Big Baby stepped him, the team went no where. Offensively, Davis wasn’t too far behind KG in the previous year, but his defensive leadership was no where near Garnette’s.
You can even look at the 80s Pistons. Isiah not only lead the team that scored the highest points in NBA history (186) but also one of the best defensive teams in history. That was what made him a winner, a desire to scrap for every ball and commit to both ends of the court, not just his unique scoring ability.
My main point just boils down to Dirk has to will his teammates to play defense in order to win a title, and that starts with leading by example.
by bearded thundar on Feb 14, 2010 4:28 PM EST up reply actions
You’re welcome to your opinion, but couple things:
Offensively, Davis wasn’t too far behind KG in the previous year…
This is almost incomprehensibly wrong. Baby Davis is light years behind KG in every single phase of the game.
RE: Nash. This is another example of having to ignore almost every other variable in order to lay the blame on a single player. Remember the “leaving the bench” suspension fiasco? Remember Sarver selling off the draft rights to Luol Deng and Rajon Rondo? Steve Nash has done nothing but play clutch, brilliant basketball, blaming him for the Suns not winning is crazy.
RE: Paul Pierce and Ray Allen. What you call: “for the first time committing to defense,” I call: “for the first time, playing with the one of the three best defenders of the decade.” (IME, the three are Big Ben, KG, Duncan). Remember how awesome of a defender Tay was when he had an in-his-prime Big Ben at C, and Sheed at PF?
I think there’s some weird Bill Simmons-esque pseudo-psychology at work when people blame the great players who spur really good teams deep into the playoffs, but don’t ultimately win a championship. Maybe, instead of blaming guys who have consistently produced superstar level production, we should look at some of the other players on those rosters and see how well they played.
by Gabe F-B on Feb 14, 2010 4:47 PM EST up reply actions 1 recs
Fair Enough
It could just be that I’ve never been a Dirk fan, but it seems that every time he loses he constantly is pouting about the refs when his team just gave up 110 points. Such as that 3 that Melo hit to win the game when he should have been fouled. I agree, that was a missed call.
But in the same token the Mavs had 50 free throws, that’s insane for a primarily jumpshooting team. I understand Dirk getting superstar calls, but Brandon Bass getting 14 free throws? To me, that says the refs blew quite a few calls on both ends, not only against the Mavs.
I don’t know, I just would feel a lot more comfortable calling him a true winner if their point differential was more than a measly 2 and they gave up less than 99 ppg.
Maybe, instead of blaming guys who have consistently produced superstar level production, we should look at some of the other players on those rosters and see how well they played.
Now this I disagree with. If you consider your player a true superstar, it’s their job to create for their teammates. Granted the Mavs were a team of aging stars, but its not like Dirk was playing with the mess LeBron had in 07. The same way the leader gets all the credit for success, he also bears the responsibility for failure.
by bearded thundar on Feb 14, 2010 5:37 PM EST up reply actions
If you consider your player a true superstar, it’s their job to create for their teammates.
…and this is almost exactly the much joked about Petey theory of “deforming the defense.”
Players are responsible for their own production. When Dirk averages 27 PPG’s, with a TS% of .635%, he is doing his job. Pointing the finger at him for not deforming defenses and magically making his teammates better isn’t logical.
by Gabe F-B on Feb 14, 2010 5:47 PM EST up reply actions 2 recs
That’s an argument you hear a lot (“Player X isn’t a true superstar because they couldn’t will their team to more wins”) but I’ve never really subscribed to that theory. There are so many variables involved in the game that one guy can only do so much. And in the case of Dirk, he really IS doing everything he can. There are only a handful of players, historically, who you could say singlehandedly willed their teams to victories and championships. For the most part, even superduperstars had very, very strong supporting casts.
by garrettelliott on Feb 14, 2010 5:50 PM EST up reply actions
Yeah, the simplest way to show it is by looking at the Bulls teams from ‘92, ’93 and ’94.
If you compare the individual regular season and playoff stats from ‘92 and ’93 when the Bulls won championships, with the individual regular season and playoff stats from the ’94 season without Michael Jordan, you’ll see that all the supporting players posted roughly the same stats in the championship seasons with Jordan as they did in the season without him.
Plus, Pippen had probably his best statistical season in ’94 without Jordan.
The Bulls won 6 championships because Jordan was an amazing players (and so was Pippen), not because Jordan had a special ability to make his teammates better.
Re: opponent PER
PER essentially tracks PPG and rebounding, and while I don’t like it for evaluating personnel, it might actually be kind of useful for evaluating how well a player defends and blocks out.
If a player consistently (PF/C) holds an opponent to a low PER, that means he’s keeping his opponents from scoring points and grabbing boards.
No
It means his opponent is not doing a good job scoring points and grabbing boards. Can’t really say why that is happening.
If it were consistent, I think it would say something. The larger the sample size, the more confident one could be.
Consistency does not equal causation and Large Sample Size does not eliminate Bias
Check out thislink.
It does a pretty good job of explaining the Fundamental Problem of Causal Inference. This is mainly a problem in social science where it is very hard to create randomized control experiments. It’s relevant to any discussion of advanced basketball stats, because you can’t assume independence among observations. In other words, your sample may be biased in a number of ways, because you didn’t draw it randomly. That, in turn, will likely bias your estimates.
When thinking about something like opponent PER, you’re looking at SOME measures of performance. That really just tells you how well the player did according to that metric. You can’t just take a big sample size and say, “look, he caused his PER to go down.” Large samples decrease variance, but don’t eliminate bias. You simply can’t control for every variable that might be correlated with both the opponent’s PER and the other player’s defensive performance. Therefore, you’re likely to have Omitted Variable Bias.
Due to these issues, you will NEVER EVER be able to say, statistically, that Player X caused Player Y’s PER to decrease.
by Colin M on Feb 15, 2010 4:53 PM EST up reply actions 1 recs
What you’re saying is kinda the half truth. It’s definitely true that you can never say with certainty that player X caused Player Y’s PER to decrease.
IMO, PER isn’t the best example because it does a poor job of weighting possessions and efficiency, but still, tracking counter-part stats for players individual match-ups doesn’t necessarily need to explain the “why” to be useful, as long as it reliably explains the “what.”
The utility of any counter-part stat is to compare matched-up players production, not to explain/analyze the results.
The nice thing about basketball is everything on one end of the court is mirrored on the other. So theoretically, two players matched-up on opposing teams should have roughly the same factors “biasing their sample,” to whatever degree that it occurs, canceling the effect out over a large enough sample.
Which half is false?
Everything I said is 100% true, as far as I know. So, I’m not sure what you mean by saying it’s half true.
Your point about match ups is fine, to a point. I don’t think there’s anything wrong with looking at head to head stats and saying, “Gee, Paul Pierce seems to dominate the shit out of Tayshaun.” If you’re not going to analyze WHY that happens, then I don’t see a problem, but Brgulker attributed a cause to an effect. That’s a “WHY” interpretation, not a what, IMO.
But, your argument that the biases cancel each other out is just wrong. That’s not how bias works. Bias is generally caused when some omitted variable is related with both what you are trying to measure and at least one of the other variables that you included in the analysis. There are other causes of bias but let’s ignore those for now.
Suppose that you observe that Player X performs below his averages when he is matched up with Player Y. Now, suppose that you observe that there is a negative relationship between Player Z’s (he is Y’s teammate) presence on the floor and Player X’s performance. If you don’t include Player Z in your analysis, your estimate will be biased. Suppose Ben Wallace is Player Z. I know that you agree that Player Y is going to look like a better player if he’s on the floor with Ben Wallace. I don’t see how that will “cancel out” on the other end of the court. Ben Wallace sucks at offense, but I don’t know if we can say that his sucky offense will negate the decrease in Player X’s PER. IMO, the bias remains.
Of course, in a regression analysis, you could just control for that relationship. HOWEVER, you can’t control for what you don’t know about. Basketball is extremely complex and dynamic. I just don’t see how you can possibly be aware of all the potential confounding variables and control for them.
Finally, I understand that I’m being a nitpicky son of a bitch. However, I hear the word objective being thrown around a lot in reference to stats. I also hear a ton of causal language, when it isn’t justified. So I decided to start dropping nerd bombs.
Here’s a fun example of meaningless correlation: This study had over 10 million participants and found a relationship between astrological sing and medical disorders. Incidentally, this one has nothing to do with bias, but it is a nice example of how easy it is to find a statistically significant relationship.
Heh, that’s a funny study.
When I said half-truth I didn’t mean that anything you said was false, just that there was another side to the coin.
My point was that the same complex relationships exist on both sides of the ball.
Using your example of Paul Pierce and Tay to make a general point: Pierce’s ability to defend Tay is influenced by KG, and at the same time Tay’s ability to defend Pierce is influenced by Big Ben (and vice versa). But if Tay struggles to keep Pierce from being productive (and if Tay’s production is held down when they are matched-up), while that might somewhat reflect on Big Ben and KG, (IMO) it clearly reflects more on Tay and Pierce
I agree that trying to account for every single variable leaves you with almost total mush (take a bow, adjusted +/-), but it still seems like a simple counter-part comparison is useful.
From what I understand of his metric, Ty, formerly of bucksdiary, uses a per/possession calculation of every box score stat by league average for each position, then compares each individual players production vs. their counterpart to see if they are holding their counterpart above or below the league average in each stat, and by how much. I think his method is a big step forward, but it hasn’t been widely adopted yet.
by Gabe F-B on Feb 15, 2010 6:27 PM EST up reply actions 1 recs
Fo Sho
Like I said, I don’t think comparing match up stats is worthless.
Your point about KG is useful, but only to a point. Those interactions are really complex and are not necessarily going to be equal and opposite. Does KG help Pierce as much as Ben helps Tay? What about Rasheed, or having an elite level coach like Khandor or Larry Brown? The options are pretty much endless. Intuitively, the idea of bias canceling out makes some sense, and my gut instinct is that it is good enough for our purposes, but from a statistical standpoint, it is absolutely false.
My point is that every single statistical analysis of basketball is vulnerable to bias because you can’t do randomized experiments. There’s nothing you can do about that. Sample size increases precision but does not correct bias. That is a fact.
This isn’t a stats blog, and I’ve already conceded that stats are useful, so why am I being such nerd? Well, my problem is that it’s really easy to take a biased estimate, and say it’s an objective analysis. But it’s not! It’s biased. But people take it more seriously than a carefully argued, non-quantitative analysis because there are numbers in it. Also, it seems like some people (not you) are on a bit of a high horse about how awesome stats are, when it seems like they don’t fully know what their talking about. I find it kind of annoying.
Anyways, you’re like the most respectful person to debate on the internets, so I’ll stop beating a dead horse.
by Colin M on Feb 15, 2010 7:03 PM EST up reply actions 1 recs
What about Rasheed, or having an elite level coach like Khandor or Larry Brown?
:) +1
Anyways, you’re like the most respectful person to debate on the internets, so I’ll stop beating a dead horse.
Honestly thank you and I’m glad to hear that. I really just enjoy thinking about and procrastinating via thinking about basketball.
my problem is that it’s really easy to take a biased estimate, and say it’s an objective analysis.
This is a definitely true and something I try to (and need to) keep in mind.
Overall, the reason I like basketball stats (and basketball in general), is IMO the internal logic of the game makes it both complex, and ultimately relatively simple.
Teams and players always have choices on each end. And with certain exceptions (cough… MCIAFI…cough) teams usually make rational, predictable decisions, which IMO contributes to the explanation for why basic basketball box score stats tend to be far more consistent from year to year than the stats from other sports.
An example of what I mean is; theoretically if the Spurs/Celtics replaced Duncan and KG with very poor defenders it’s possible/probable the change would affect Ginobli/Pierce’s defense to an extent, but it’s (IMO) likely that opposing teams would choose to attack the weak interior defenders more than before, leaving Ginobli/Pierce’s net defensive impact roughly the same.
And this comment column is impossibly narrow now and I have to keep hitting preview to see if any of this makes sense, and I’m not sure that it does :(
Colin,
Not to rekindle a debate that’s ended, but I thought that the tentative language implied all the points you raised.
Let me be explicitly clear, though.
If a Player X (who plays at PF or C, because that’s what position was being discussed) consistently holds his opponents to lower-than-average PER’s, then we could say with some level of confidence — not absolute confidence, but some level of it — that Player X is doing a good job of defending his opponents and keeping them off the boards. We can say that because PER, as it relates to big men, essentially tracks PPG and rebounds.
(That’s a little oversimplified for PER, but not by a lot).
I think that gets at your (correct) objections/observations about interdependence, limitations of any given statistical model, bias, and certainty.
I didn’t ever mean to say something like, “look, he caused his PER to go down,” but if that’s how it came across, then I just wasn’t clear enough.
One last thing:
Due to these issues, you will NEVER EVER be able to say, statistically, that Player X caused Player Y’s PER to decrease.
I agree. Correlation and causation aren’t the same thing. I don’t mean to imply that they are. But in basketball, GMs have to evaluate players and make decisions based on those evaluations. GMs simply must look at these types of correlations and draw whatever conclusions they can, no matter how tentative, and then make financial decisions based on those conclusions. Statistical models, including part/counterpart production, can be helpful in making those decisions. Or at least IME.
Man, I just wrote a reply about the necessity of both qualitative and quantitative player evaluation, the continuum of objectivity and subjectivity, and I accidentally hit cancel instead of Post.
In sum, I don’t think pure objectivity exists. There’s always, always, always bias. I agree 100%.
I also don’t think that stats tell the whole story, and I don’t think any statistical model — regardless of how objective it attempts to be — explains anything completely, or perhaps even primarily.
Finally, I made some comments about how some DBB’ers have found me annoying or condescending at times. I sort of wish that you guys could meet me. That’s not how I am. I’m hoping this is due to the communication medium and the general suckitude of this season (which brings out the pessimist in me). Regardless, I apologize if I’ve been a pain in the ass. I trust and respect the posters who’ve said this, so I’ll try to be more intentional about my tone and choice of language.
by brgulker on Feb 16, 2010 1:12 PM EST up reply actions 1 recs
you're not annoying...
I like reading what you have to say. You’re very good at driving discussion around here. I’m not in anyway much of a basketball person. Most of the stats talked about here are new to me. Driven discussion helps me learn more about the game. I’ve never noticed any abrasive attitude from your comments but I hardly represent a majority opinion that makes any relevant statistical significance.
Basically though my comment here is directed toward a bunch of the DBB’ers. You guys keep discussing and I’ll keep on a learning.
My Music: Now on last.fm!!
My Blog: Strike Three Mechanics
I actually think you seem like an extremely nice guy. You also are obviously intelligent and know a lot about basketball (more than me, I think). I bet we could have a great time watching a game and having a beer.
However, something about your writing style does get under my skin at times. I don’t think you should worry about it too much. I’ll be ok. Secondarily, I do take issue with the way you overvalue anything that has stats in it. I’m sure you would disagree with that, but I think this thread and the fanpost you had on Lebron James’ fouls are perfect examples. That Lebron article was complete dogshit – an absolute misuse of stats – and you called it damning for the nba. As for this thread, in the comment above, you said something about confidence. Now I’m being the condescending one, but in statistical language, confidence has to do with precision, not bias. Precision is usually measured with a confidence interval. You can’t quantify bias, because it usually comes from stuff that you didn’t include in your analysis. Bias would be more akin to accuracy than precision. In other words you could have a very precise estimate that was way the fuck off the mark.
Anyways, you’re absolutely right that it makes sense for GM’s to use these types of stats to get a better sense of a player’s value. In truth, you weren’t really saying anything that crazy. I just felt the need to push back against the statz fever. I suppose this season has also brought out the worst in me. So, I’m also sorry if I was a bit of a dick about it.
I forgot to say -
Don’t go anywhere. Even if you piss me off sometimes, you’re always good for a smart, spirited and respectful debate. Aside from comedic relief, I think that’s why most of us are here.
Yeah, it must be that Bennett Salvatore picked up on Dirk’s inability to win, which is why he decided to award Game five of the 2006 finals to Miami. That was worst example of officiating to the narrative I have ever seen in any sport.
by Kevin Sawyer on Feb 14, 2010 8:10 PM EST up reply actions
Think 80s
By getting smaller, they mean you don’t see nearly as many dominant 7 footers. Shaq appears to be the last of the truly unstoppable centers. As good as Cousins will be, I don’t think he’ll be averaging 30 ppg because of the lack of athleticism that Shaq had.
Also the PF position has gotten much more mobile. You see a lot more Tyrus Thomas and LaMarcus Aldrich like players than old school Sheed (before he discovered the 3 point line).
The NBA has changed, there’s no way B-Consistent survives a single season in 80s as PF, but at same time he could be a dominant SF, because as Bird has shown, that position didn’t require that much athleticism at the time.
Lucky for us, Detroit has had 2 players in a single era that really changed the face of the NBA. Joe Dumars really brought about the concept of lock down perimeter defense and Bill Laimbeer had the best outside touch of any center in the decade.
by bearded thundar on Feb 14, 2010 1:54 PM EST up reply actions 1 recs
Detroit has had 2 players in a single era that really changed the face of the NBA. Joe Dumars really brought about the concept of lock down perimeter defense and Bill Laimbeer had the best outside touch of any center in the decade.
"I didn’t even know Elvis was from Memphis, I thought he was from Tennessee." — Drew Gooden.
by Skylar on Feb 14, 2010 3:22 PM EST reply actions 1 recs
Comments For This Post Are Closed
Secondary Sidebar
User Tools
FanPosts
Community blog posts and discussion.
Recommended FanPosts
Recent FanPosts
FanShots
Quick hits of video, photos, quotes, chats, links and lists that you find around the web.
Recommended FanShots
Recent FanShots
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.
