TRADE MACHINE ECONOMICS

So, yesterday a trade happened. I'm not going to talk a whole lot about the #BasketballReasons of the trade (other, than, perhaps to take a 5-minute break laughing myself silly at those of you who actually thought Toronto got better here) because I think Ben Gulker and Devin Digham both do a good job of that here and here. I just want to briefly talk about the economics.

 

  POS Min WP48 PoP48 Wins PTS DRB ORB REB AST TO BLK STL PF
Prince SF 1457 .087 0.0 2.7 17.4 5.5 1.4 6.9 3.7 1.8 0.4 0.7 1.5
Gay SF 1541 .040 0.0 1.3 22.5 6.0 1.7 7.8 3.4 3.2 1.0 1.7 2.7
Calderon PG 1273 .235 0.0 6.2 18.9 3.5 0.5 4.0 12.6 2.8 0.2 1.1 2.2
Davis PF 1087 .192 0.0 4.4 19.3 9.1 4.2 13.2 2.4 2.0 1.7 1.1 4.9
Daye SF 348 .137 0.0 1.0 16.8 7.3 1.4 8.7 2.9 1.8 1.1 0.7 5.9
Average SF SF 996 .099 0.0 2.1 19.8 6.0 1.6 7.6 3.7 2.5 0.8 1.6 3.3
Average PG PG 1004 .099 0.0 2.1 19.8 3.9 0.9 4.8 8.4 3.4 0.4 2.0 3.4
Average PF PF 978 .099 0.0 2.0 19.3 8.1 3.7 11.8 2.5 2.4 1.6 1.3 4.5
  FG% 2FG% 3FG% FT% eFG% TS% FGA 3FGA PPS FTA
Prince 44.4% 44.5% 43.4% 79.6% 46.8% 50.2% 15.9 1.7 1.09 3.1
Gay 40.8% 43.1% 31.0% 77.6% 43.8% 47.8% 21.4 4.0 1.05 4.9
Calderon 47.0% 50.5% 42.9% 90.4% 56.9% 59.4% 15.0 6.9 1.26 2.0
Davis 54.9% 54.9% 0.0% 64.7% 54.9% 57.0% 15.0 0.0 1.29 4.5
Daye 44.3% 38.6% 52.5% 83.3% 55.2% 58.1% 13.4 5.5 1.26 2.5
Average SF 43.5% 47.2% 36.6% 79.0% 49.9% 53.8% 16.5 5.7 1.20 4.1
Average PG 43.1% 46.4% 36.1% 80.6% 48.8% 53.0% 16.9 5.4 1.17 4.1
Average PF 47.7% 49.4% 34.8% 70.6% 49.7% 53.1% 16.2 1.8 1.19 4.6

 

Of these players, Gay has by far the biggest salary, at $16.5m annually with 3 more years. The next biggest, Calderon's, is $10m but he is in his final year. And that is why Memphis and Detroit made out like bandits in this trade, and Toronto got hosed badly.

Let's examine this for a minute. Under the new CBA, maximum contracts for most players after their 4-year rookie contracts run out are (ballpark) $15m annually (and will be even lower in a couple of years, when the 2011/12 rookies, who make a lot less in the new CBA, start looking for extensions). Players who qualify under the Derrick Rose rule can get a little more, approaching $17 million (again, this will lower in a couple years). So, essentially, Gay will be making the quivalent of a max contract player AND a midlevel exception player. Given this, Rudy Gay is not worth anywhere close to his salary; this is true even if you think "those #wagesofwins guys are nuts". It's true simply because a) Gay is not a max player and b) Gay's contract is from the old CBA and thus costs even more than a max player. 

Do y'all remember a few months ago when James Harden wanted the max, and the internet was awash with fools folks asking "Does the earth really resolve around the sun?" "Is James Harden really a max player?" That was the James Harden who in 2011/12 scored 16ppg on a mind-boggling 66% TS% (4th in the NBA). Forget about everything that happened this year, and the fact that Harden is still a relentlessly efficient scoring threat even now that he is the "featured" guy, and just transport yourself to September 2012 and ask yourself if you'd trade then James Harden for then Rudy Gay straight up.

I don't think you would find a GM in the league who'd have made that trade (other than one GM, the one who's giving up Rudy Gay). Will you force me to point out that Gay in his best season didn't shoot as well as Harden in his worst?

Harden, a guy that was universally acknowledged as a star-caliber player, and a player that almost everyone acknowledged was worth more than Gay, had plenty of people doubting that he was worth the max. Meanwhile, most people agreed that Gay's contract was awful; even when the ink was still fresh, many around the NBA were scratching their heads at the price he commanded. The only worse contract in the NBA right now is probably Amare's.

Now, instead of having the horrifically overpaid Gay on the books, Memphis has the moderately overpaid Daye and Prince (it's amazing that Dumars gave out a contract this long to a player this old) and the severely underpaid Ed Davis. The flexibility this gives the team has to be considered a huge win.

Toronto traded two underpaid guys (you can argue about whether you think Calderon is good or not; if you consider that it's an expiring deal, he's a huge value) for a laughably overpaid wing who has spent his entire career underachieving to the potential that everyone thinks he has. They're essentially stuck with $40m-ish in Gay, DeRozan, and Bargnani for the next 3 years. The only bright spot for Colangelo and Casey, whose jobs are now going to suck (seriously who wants the job of trying to win with this roster and contract collection!?), is that both are likely going to get fired before the season is out, and their successors will have to deal with this dreck.

Detroit, on the other hand, traded two average, overpaid wings for a good point guard in the last deal of his contract. That's a huge bargain.

Has anyone checked on Joe Dumars lately? Are we sure he's feeling OK?

 

Picture?type=square

Omar Kreidieh

you ran away from making predictions about what will happen with each team. You know your metric overvalues players who dont shoot much because they only take shots when theyre sure to make them. Whatever you predict will be something close to : pistons will win 80% of their games from now on since they have both Drummond and Calderon. This trade is the ultimate test for your metric which many feel has gaping deficiencies. I think most people watching your site want to know what your exact numbers are. This is the greatest test... please post your predictions based on the new squads.... Detroit now have 2 of your favorite players... will they be champions??!

111 days ago

A673067dd6c317c52e3c417bb63f147a_normal

VSA

Omar,

I understand you're skeptical of the metric Patrick uses here on this site. I tend to agree that using a catch-all metric to determine player value alone is unwise. However, you need to use your head and think sometimes.

"will they be champions??!"

What a terrifically obtuse question. Obviously, as Drummond doesn't play full starter minutes his impact is lessened. He can only provide value when he's actually playing. Jose Calderon averages just over 28 minutes per game for his career, is competing with Brandon Knight for minutes, and is injury-prone. Patrick believes they are very productive players, but neither plays a ton of minutes.

Here's the other thing. When Calderon isn't playing some combination of
Knight, Stuckey, or Bynum is. Those three sport a WP/48 of .012, .040, and -.025 respectively. If Calderon isn't playing a ton and bad players are there is no reasonable way this team will shoot into the upper echelon. This is even before addressing the terrible depth at the forward slots. Nobody is going to take your doubts seriously if you are incapable of figuring this much out on your own.

111 days ago

Picture?type=square

Ben Herbin

I'm sure this point has been made to you before Patrick but by your criteria I'd love to see where Kobe's nearly 60M contract over 2 years ranks on the list of "Bad contracts" by the numbers? Why aren't you talking about him and how 'terrible' his contract is??

The problem with you metric fella's is there is no way to predict a players value to the rest of his team. How Valuable is it to Metta World Peace to have the best defenders on the opposition defend Kobe every game? How valuable will it be to Ross, DeRozan, Anderson to now face average defenders instead of top defenders? Does that change help a Raptors TEAM win games or lose games? How much was Calderon's 3pt% elevated because Amir set good screens for him? Is Amir worth his contract because he sets good screens? Does his value instead come in the form of his ability to play through injury? Can you track that ability to play through injury? How many points on your system to you award a player for setting a successful screen that leads to a basket? How many points are awarded for playing through a sprained ankle and back spasms? Seriously, you don't just pay a player based on their numbers efficiency, you pay them based on what they do for the team as a WHOLE.

It would be interesting to see a GM put together a roster of the 12 best players by the efficiency numbers relative to their contracts... How far would that team go?? My bet is under 20 wins because none of them command referee's attention, cannot create their own shot and don't garner the respect of their opponents! You know why you don't see GM's doing this? How about the fact they'd lose their jobs and probably never get another one anywhere. The most efficient players are smart and play off of the other players on their team. Er say, you need Players who command respect, create for themselves, make contested shots in order to build a successful TEAM. And Yes, by the numbers these players contracts often look poor compared to others, that's the price of winning.. or in the Raptors case, trying to win.

111 days ago

Picture?type=square

Omar Kreidieh

VSA obviously that was an exaggeration. He might not predict a championship but he will predict a high number of wins. His forecast for the season is already significantly higher than what they've accomplished in real life and the difference between reality and his predictions will only climb as Calderon joins the mix.
Dont get me wrong, I like the way he calculates the metric and it applies well to a lot of players... but HORRIBLY to others. The main issue is not about rebounding ... its about shooting efficiency. Shooting efficiency is the same regardless of the shot taken and the time on the shot-clock when it was taken. Players who take a significantly less shots tend to score a higher percentage. That is a fact because them not being the go-to guys affords them the luxury of shooting only when theyre going to make the basket. Thats why the metric says that some of the best players on the pistons are middleton and kravtsov and English, and prior even Daye I think. Now if you put all of those on the floor, or similar support players who dont have much of a talent and therefore only shoot it when they can make it... well that wouldnt be much of a team now would it ( contrary to what he would predict). The problem is that when interpreting the metric, they never take this into consideration. Is Drummond a great player. Yes, absolutely. Is he as good as Thenbageek.com rates him? DEFINITELY not. Im a huge pistons fan but I know that a huge reason Drummond is so efficient is because he only gets the ball when he has a dunk opportunity. In the regular half-court set, unless there's a pick, the ball wont be chanelled through him and thats why all the tough shots ( which he would never have a chance of making) are taken by other players thus lowering their WP48. Now take those " horrible players = bynum and stucky and knight" out of the equation and your left with drummond and the like. They will have the same number if not less easy shots to score, but will still have to shoot those tough ones. When they miss them, and they will miss them at a worse percentage than knight, stucky or bynum thats for sure, their WP48 would have been destroyed. My point is that not all shots are equal, as the metric assumes, and this skews this metric the most because it looks at efficiency and not points. We cant get all dunks or open shots in a game and thats why a team of 5 drummonds or 5 calderons would do horribly. When you give the same credit for a shot taken 2 seconds before the end of 24 seconds by the go-to guy as you would an easy lob dunk taken 10 seconds into the possesion... you produce a serious mis-evaluation. All metrics do this. All analysts say that we need to understand that this metric has a problem of overevaluating or underevaluating this or that player because of this or that reason. The problem is that the writers on this website ignore the problems and insist the metric is perfect. People actually end up thinking coaches and GMs and the experts are stupid because WP48 is the perfect holy grail of basketball and it said Calderon rocks and the pistons will make the playoffs so if they dont its LF's fault ( It is, but not because of this reason :P ).

111 days ago

Picture?type=square

Zachary Young

Nobody's saying that WP is perfect. The only claim is that it explains wins better than any other statistic. You can complain about the methodology all you like, but only if you acknowledge the fact that there is a historical 95% correlation between WP and actual wins. Even if you don't like the narrative it might tell, it's accurate as currently constructed, no matter how flawed it might be.

111 days ago

Picture?type=square

Ben Herbin

Accurate because it does a good job of breaking down the numbers of the individual parts of a particular team and how they relate as presently assembled. Not because it does much to determine the value of said player beyond actual understanding of the game itself and is essentially useless in determining what pieces fit well together. Essentially fantastic to use assembling a fantasy team, next to useless in assembling as actual one.

111 days ago

Picture?type=square

Omar Kreidieh

Thank you Ben, Zacharay and VSA, I suggest you read this : http://ascreamingcomesacrossthecourt.blogspot.com/2012/10/jose-calderon-example-of-how-wins.html

111 days ago

Picture?type=square

Ben Herbin

Very insightful Omar, thanks.

111 days ago

Nba-geek-avatar

Patrick Minton

Omar,

Don't bother reading any of the two links that I posted, each of which make lots of predictions (that I implicitly agree with).

Also, LOVE the crazy overreaction. U mad bro?

Here's my prediction: I predict that IF Jose gets the starters minutes and Drummond continues to get 25+ minutes, Detroit makes the playoffs.

Barring another trade, I predict that Toronto will discover new, amazing depths of suckitude.

Somehow, this seems like a win-win for you, because if I'm wrong, you'll be back here to gloat, but if I'm right, I suspect you won't be around to let me gloat, will you?

110 days ago

Nba-geek-avatar

Patrick Minton

Ben,

I've spent plenty of time arguing about how bad Kobe's contract is. One of the first articles I wrote on this blog argued that LA should amnesty him. He's good but he's not $27 million good.

Of course, the difference between Kobe and Amare (or Gay) is that:

1) Kobe is actually a good player.
2) Kobe's deal expires next year

Amare's $20m deal has an extra year on it, and Amar'e was a terrible player the last couple of years.

110 days ago

Nba-geek-avatar

Patrick Minton

Omar,

One last comment:

"Shooting efficiency is the same regardless of the shot taken and the time on the shot-clock when it was taken. Players who take a significantly less shots tend to score a higher percentage. That is a fact because them not being the go-to guys affords them the luxury of shooting only when theyre going to make the basket."

You and I define "fact" differently. This is, actually, not "a fact" at all, and you saying "them not being the go-to guys affords them the luxury of shooting only when theyre going to make the basket." is not actually "proof" or even "evidence".

You got to rant a lot, though, so there's that.

110 days ago

Default_profile_1_normal

Brian Foster

Reading the major media outlets write about this trade has broken my brain, I can't help but ask myself:

Why oh why didn't I take the blue pill??

110 days ago

Picture?type=square

Brett Gunter

Patrick,

My condolences on the outbreak running rampant in your comments section; idiocy is more persistent than most epidemics, but we can hold out hope it will run its course.

It occurs to me there is a very good chance that circumstances will mask the fact that Toronto gave up the 2 most productive players, while getting the least. Their 4 best guys were 2 PG and 2 PF, and while they SHOULD have played them all together, they did not. If we accept that the optimal strategy will NEVER be employed, then the value of all other strategies increase.

Should they have found a better trade than this one? Obviously. However: 1) it does solve their perceived glut at both positions; 2) the presence of Gay and Fields should keep DeRozan at SG, where he will be more productive (well, less unproductive); and 3) the DeRozan-Fields-Gay rotation makes them taller, while still athletic, and everyone in the NBA knows how important it is to be tall enough for your position. Acy may get time in the PF-C rotation with Amir and Valanciunas; that may be a good or bad thing, but at least they'll find out. Haddadi has been hot-and-cold, but could be a solid backup C. I doubt all that can overcome their black holes on the wings, but...

Unfortunately, Lucas is a terrible PG--and Bargnani will be back, and play 30+ horrible minutes a game; they can't overcome that. If there IS another trade on the horizon, Andrea needs to be the centerpiece--and it literally does not matter what they take back.

110 days ago

Picture?type=square

Kevin Sawyer


"pistons will win 80% of their games from now on since they have both Drummond and Calderon."

The WP/48 types did preseason projections, and had no team winning 80% of their games. Drummond only plays 20 mpg, thanks to a coach who thinks quite a bit like you.

"Why aren't you talking about him and how 'terrible' his contract is??"

This is a great point. And I would like to add. Why do I never read on this blog about how bad Rudy Gay is, compared to his contract?

"How Valuable is it to Metta World Peace to have the best defenders on the opposition defend Kobe every game?"

Not very. His offensive numbers have matched his career numbers across the board since he joined the Lakers.

"How valuable will it be to Ross, DeRozan, Anderson to now face average defenders instead of top defenders? "

Not very.

"Does that change help a Raptors TEAM win games or lose games?"

Lose. The Raptors will lose a greater percentage of their games now.

"How much was Calderon's 3pt% elevated because Amir set good screens for him?"

Not at all. He shot the same with or without Amir.

"Is Amir worth his contract because he sets good screens?"

No. He is worth his contract because is he is an efficient scorer and good on defense.

"Does his value instead come in the form of his ability to play through injury?"

No, it comes from his efficient scoring and good defense.

"Can you track that ability to play through injury?"

Irrelevant. What matters is minutes played, if you are talking durability.

"How many points on your system to you award a player for setting a successful screen that leads to a basket?"

That and a few others are neglected by metrics as a whole, and by pretty much everyone. However, it has nothing to do with this conversation.

"Seriously, you don't just pay a player based on their numbers efficiency, you pay them based on what they do for the team as a WHOLE."

Rudy Gay has had no impact on his team as a whole.

110 days ago

Picture?type=square

Omar Kreidieh

you miss the point entirely ... 80%, champions, ... are all satirical comments ... the point is even projecting a playoff berth for them is outlandish in my book and im a huge fan. Anyways, it does seem like drummond will continue to get 25 minutes or around that number, and Frank did say that calderon will start; I will be back regardless of the results I can promis you that.

Patrick: Here are the numbers for you : Knight : 11.7 FGA , 0.012 WP48 Monroe : 12.8 FGA 0.094 WP48, Prince : 10.8 , 0.087 WP48 and Stucky 10.1, 0.04 WP48. Go- to guys in detroit's offense for those who dont watch them are Greg Monroe and then the guards/ prince

Patrick's studs : Drummond 5.4 FGA, 0.36 WP48; Austin Daye = 4 FGA, 0.137 WP48, Kravtsov 0.357, 0.8 FGA...... in fact ... even if you adjust for mpg, these players still take less shots than the go-to guys so yes it IS a fact that players who are not go-to guys take less shots ( I would have thought it was a self-obvious statement but ... well ... ) . They also are defended by lesser defenders, tend to play against bench players more often, and are fresher when playing their minutes because they arent asked to do as much or stay in as much. Take all of that away from them and their numbers will drop. Now is this the same elsewhere I wonder?

Orton daniel and anderson chris lead the WP48 statistics but lets ignore them because of their small sample size of minutes. Top three then :
Drummond, Chris Paul, and Tyson Chandler. Paul is a stud and shoots efficiently despite having to take many. But two of the top 3 are there not because of what they do, but because of their efficient shooting/ inability to shoot: Chandler: 6.4 FGA per game .... This is not to say chandler isn't a great player.... BUT ... he is no top 3 in the NBA and nor is Drummond.

again: Sefosha 5.6 FGA per game, Demare Carrol 4.7 FGA per game, Kawhi Leanord 7.2 FGA .... All of those ranked higher than rondo, parker, wade, randolph, harden ect.... And then you call me stupid ...

110 days ago

Picture?type=square

Omar Kreidieh

Theres a bit of a fix I think... you just have to accept the deficiency... Im not sure how technically feasible this is but you can grade the shots made per seconds left on the shotclock. That would be great because it measures a player's shooting as well as his decision making. The idea would be that you take all the shots taken in the first 10 seconds of a possession and grade them against the average FG percentage of shots taken in the first 10 seconds. Then the next ect... that way guys having to take the tougher shots at least get some sort of credit for them. Not a perfect fix, but a bit better. At the end theres no perfect metric. Yours comes close but you really should just interpret it taking the imperfections into consideration.

Just so you get a feel of how much this will affect the metric :
http://www.82games.com/clock.htm
You can see a HUGE drop off towards the end of the shot-clock. It drops from a bout 50% in the beginning of the shot-clock to about 35% at the end. Guess whose taking the shots at the beginning and guess whose having to take those lower percentage ones at the end?

110 days ago

Img_3451_normal

Jon C

Well this is going to be an interesting trade. I personally think WP is a fine metric when generally applied to NBA players as a whole. But that certain NBA players have both skills and deficiencies that don't get captured by box scores. Calderon is held out by many as a perfect example of that because his shooting percentage and assist to turnover ratio is elite. But his individual defense is supposedly so bad that it is a real problem. Now even if we accept that, the fact that Calderon is now going to be backed up by Monroe and Drummond is probably going to help cover up the defense. So yes I agree with Patrick's prediction and I suspect there is going to be some crow to be eaten here when the Piston's make the playoffs.

Question though. Do people really think Bargani is going to be played when he returns from injury? I mean isn't there any amount of evidence in actual Raptor wins when he is out that could convince the Raptors not to play him? If Bargani is played, then I think Gay is going to look really bad because there will probably be a lot of Raptor losses and maybe more than should will be attributed to new guy Gay.

110 days ago

Nba-geek-avatar

Patrick Minton

Omar,

No response to any of Kevin's points? Or are you just uncomfortable in debates that involve actual evidence?

Interesting how you ignore the following players, all of whom have MUCH higher FGA/minute than average, in your cherry-picked examples:

Kevin Durant
Kobe Bryant
Kevin Love (excluding the year-of-the-broken-hand)
Derrick Rose
Zach Randolph
Dwayne Wade
James Harden

probably dozens more. It boils down to this:

- if you are good at shooting, and you shoot more, your team benefits (note how Kevin Love and James Harden didn't suddenly suck when their shot attempts went up, but I'm sure you'll attribute this to "chemistry" or "learning" or some other metric that is conveniently impossible to verify)

- if you are bad at shooting, and you shoot more, your team loses more

- if you are average at shooting, and you shoot more, it won't matter (but you will get paid more for having a higher PPG because guys like Omar think you are now a "go to" player)

Three guesses as to what the players' primary incentives are.

Also, forget shooting. Even if Drummond never got extra looks, if he had the conditioning to play 30mpg, he'd average like 13ppg/15rpg/4 blk/2stl or something absurd like that. Are you honestly obtuse enough to believe that this is not extremely conducive to winning? Really!? Ask New York if they'd be so kind as to return Tyson Chandler to Dallas, then, because I'm sure he doesn't really make that much difference.

109 days ago

Nba-geek-avatar

Patrick Minton

Omar,

"Guess whose taking the shots at the beginning and guess whose having to take those lower percentage ones at the end?"

Actually, I don't like to guess. I prefer facts. I know you THINK you know who's doing this.

I'm also wondering if you are ignoring any of the other factors:

1) what percentage of total shots are taken at the end? If this is small, then frankly who's taking them does not have a big impact

2) what leads to these situations? Every time I watch Westbrook "probe" for 10 seconds before giving it up, then later get it back and shoot at the end of the shot clock, you're praising him for putting up a tough shot with limited options, and I'm calling him stupid for not running an offense early in the shot clock.

It's funny that people always accuse me of making the game simple, yet people who criticize me make ridiculous assumptions like the that it's really as simple as "someone has to shoot the ball".

109 days ago

Nba-geek-avatar

Patrick Minton

Brett,

there's truth to that. But Lowry is injury prone, Calderon got a lot of minutes, and their bench is now awful.

There's some hope if they trade Bargs or if he just sits a very long time.

Also, I like the "rage" in the comments. I'm not making any progress unless I'm pissing off the mainstream :)

109 days ago

Picture?type=square

Omar Kreidieh

OK here goes :
1. "No response to any of Kevin's points": Kevin's post was mainly a response to Brett and the only part that was a response to me was saying that you never predict 80% wins. To which I clearly responded that it was a satirical number. Obviously no one would predict an 80% wins. If theres any specific point you feel was relevant to me please let me know.

2. On having many players with many FGA/minute highly ranked: The ones you chose are not ranked highly enough. Besides, these guys may be just so elite, that even shooting at crunch time, they would have a higher than average percentage and therefore gain wins off that in the metric. The reason why theyre still good on the metric is that their percentage is so much higher to begin with, they likely dont have as much of a drop off towards the end of the shot-clock, and they are some of the few with an effective shooting percentage above 52% ( This is partly a curious question. Im just wondering were you got that threshold from because when I looked it up just 42 players had a percentage higher than that, so for all other players, shooting less often would translate into a higher WP48 :S ... )
3) "what percentage of total shots are taken at the end? If this is small, then frankly who's taking them does not have a big impact": I did the math on the source I provided earlier (http://www.82games.com/clock.htm) by doing the averages for the first 3 teams ( random coz I didnt want to waste much time on this)... about 29% of the teams' shots were taken after 15 seconds had passed on the offensive shot clock, and the average field-goal percentage for those shots was 38% as opposed to the 42.5% for all shot-clock periods ( calculated as a weighted average of FG% and shots taken), and a FG% of 47% for shots taken in the first 10 seconds.

4)" note how Kevin Love and James Harden didn't suddenly suck when their shot attempts went up" they didnt but their percentages went down. James harden's FG% dropped from 0.491 to 0.431 when his FGA climbed from 10 to 17 this season. Kevin Love is now shooting 35% with about 17 FGA per game VS 45% when he was taking close to 11. Source is ESPN.

4. "What leads to these situations?" A lot of times its the team playing bad or a bad point guard who is not the go-to guy charged with taking the resultant low % shot. There are exceptions ofcourse and westbrook is a good one. But that is not the case for all teams and often the guy taking the low percentage shot has to do it because the TEAM failed to find any higher percentage one in the seconds prior to that...

5. "Even if Drummond never got extra looks, if he had the conditioning to play 30mpg, he'd average like 13ppg/15rpg/4 blk/2stl or something absurd like that. Are you honestly obtuse enough to believe that this is not extremely conducive to winning? Really!? Ask New York if they'd be so kind as to return Tyson Chandler to Dallas, then, because I'm sure he doesn't really make that much difference." Firstly the actual numbers would be closer to 11.25 points and 11.4 rebounds, 2.4 blocks and 1.35 steals per game. I never said Drummond is not good or conducive to winning. Ive actually sent emails to Keith Langlois ( the writer for the pistons website) several times asking why hes not getting more minutes and to the rookie ladder writer asking him to pull him up a few spots on ROY. Im a huge fan of his BUT I still disagree with you on the DEGREE to wich his stats are conducive to winning. Hes an excellent player but NOT number 2 in the league and not even close...

Also Ive watched every single pistons game this season except for this wednesday's game and the guy is sometimes so lost on defense its beyond ridiculous. He makes a LOT of mistakes, has awful post-up D, gives a few open shots to opponent bigs and forgets to box out on several occasions almost every game. Still hes somehow managed to have a great season, but those points are worth pointing out.

Any ways, those are my thoughts. I dont think we're doing anything here other than just argue without convincing each other. The test is what happens at the end of the regular season. I always visit your site to read some articles so you can gloat and feel free to mention my name if they make playoffs. I wonder if youd let me write my own take on it/ gloat if they dont? :P


PS: "I like the "rage" in the comments. I'm not making any progress unless I'm pissing off the mainstream :)" : Im not enraged bro, I'm just having a discussion. Even though I might disagree with you on many interpretations of the metrick, I still really enjoy the site and the metric/ database so thanks for the hard work :)

109 days ago

Picture?type=square

Michael Motherwell

> "Guess whose taking the shots at the beginning and guess whose having to take those lower percentage ones at the end?"

I find it amazing that Omar can quote http://www.82games.com/clock.htm and then NOT quote anything even vaguely specific. It INFURIATES me - beyond infuriates me. So here are the "facts" according to 82 games:

Memphis Team Shotclock usage: http://www.82games.com/1213/1213MEM3.HTM
Rudy Gay Shot Clock Usage: http://www.82games.com/1213/12MEM9.HTM
Google Doc to make comparison easy: https://docs.google.com/spreadsheet/ccc?key=0AlSEoxg2JxHJdE14RTNIQ0JaSEk3ekNWbGM3cG43MFE#gid=0

Key Data:
Shot Clock Memphis Rudy Gay
S: 0-10 35% 32%
S: 11-15 24% 24%
S: 16-20 23% 25%
S: 21+ 19% 19%

Upshot? Gay takes a tiny fraction less early and a tiny fraction more in the 16-20 range, but otherwise, it is EERILY similar.

Now, I have never looked at this in a statistical, detailed way (scraping the data and running SQL queries) but I've also never made a claim that certain players shoot more late. However, being that I am the kind of guy that likes to go that EXTRA mile in my smack downs, I also included Westbrook's stats (team: http://www.82games.com/1213/1213OKC.HTM vs individual http://www.82games.com/1213/12OKC2.HTM same doc) and, WOAH baby, did THAT prove how true this idea is (alert: sarcasm mode may have been activated):

Secs. OKC Westbrook
S: 0-10 38% 39%
S: 11-15 27% 28%
S: 16-20 22% 21%
S: 21+ 13% 12%

So it seems that Westbrook, the guy who gets the "it is so unfair he takes shots late" benefit of the doubt is shooting more EARLY (although I'll be reasonable and say that falls in the margin of error).

This could be just a lucky two I've happened to look up, that is not beyond the realm of possibility. But as I've NEVER made a claim about shot clock usage the burden of proof is not on me I wouldn't think. Anyone got a detailed, comprehensive and clear study they can link to that shows this adjustment is fair, reasonable and justified? I'll wait.

109 days ago

Picture?type=square

Omar Kreidieh

Michael, I honestly didnt know before how to get to those specific pages. I figured it out after your posts and I have to admit that your right. the few players I looked at did seem to take a similar shotclock selection to that of the rest of their team. Drummond is the only one taking less at the end, but he somehow makes a higher percentage between 16 and 20 seconds for some reason that I think even if we adjust for the shotclock it wont make much of a difference.
So yes I was wrong about the shotclock. I still have a feeling that there are tough shots and easy shots to be had in a game and the more shots taken by certain player, the easier the shots left to others. But I can only tell for sure after I test it out. Il try and run the numbers on SPSS later: For a team, look at the FGAs for 2 players, and then compare that to the FG%s of the other 2 or 3 consistent starters on the team. For Detroit that would be Monroe and Knight, and Il compare them to the FG% of Singler and Maxiel ( Cant have drummond coz he plays with the other unit.). Detroit isnt the perfect team because singler played often with the second unit as well. Any other suggestions are welcome .... Also if someone wants to do the regression instead of me I wouldnt mind it ( youd get to gloat if it turns out not to be significant) :P Lots of work today already :(

109 days ago

Picture?type=square

Michael Motherwell

So Omar if I sum that up, you have a theory that you have never seen all the way through, that you have then used as the basis for another set of opinions, correct? That's even more reason for me to be annoyed and infuriated, because while the premise that shotclock usage varies by players is something that seems to be taken for granted, isn't not taking everything for granted the point of the stats movement? And shouldn't there at least be some foundational basis for a view that claims a statistical basis?

This is precisely what drives me nuts about so much basketball stats commentary. It seems to me that people take a concept, like clutch shooting or shot clock usage or whatever, and proceed from what is an unknown/unverified/untested hypothesis to a statement/opinion built entirely on that premise. If the premise you move from is unproven, then I think the opinion is at best speculation, and at worst simply invalid.

Now, we all have intuitions and hunches and hypothesises, but as much as possible, we should try to falsify them before we use them as the basis of our views. Or at the very least work out how they can be falsified.

As a related example, clutch play gets a lot of hype. Claiming it matters in specific cases, like contract negotiations or player evaluation, needs to start by showing that the specific conditions that constitute clutch - http://www.82games.com/CSORT11.HTM has a list - can be meaningfully understood, predicted and/or utilised. One can then, but not before, build an argument from that understanding. That, to me, seems like the basis of a well structured argument, but it is rarely the way things proceed.

109 days ago

Picture?type=square

Xavier Quach

Omar,

You keep posting that same link like it's gospel. It's an incredibly cursory analysis which is in effect one big rationalization. The fact that it so disparages "fanatical" followers of WoW, yet implicitly accepts +/- variants and SRS as the absolute truth is monumentally flawed logic.

The fact is that EVERY metric has outlying players that are overrated. For instance, RAPM has Nene Hilario as the 9th best player in the NBA, and better than Tyson Chandler. Win Shares has Chandler 4th and Nene 143rd. +/- has Russell Westbrook as the 2nd best player (nevermind that he plays with Kevin Durant) and Mario Chalmers 8th. PER has Brook Lopez 4th among players > 1000mins.



108 days ago

Picture?type=square

Michael Motherwell

> I still have a feeling that there are tough shots and easy shots to be had in a game and the more shots taken by certain player, the easier the shots left to others

I have thought about this a fair bit, and think that the question pre-suposses we can agree what a "tough" and "easy" shot are, and secondarily that certain players HAVE TO take certain shots.

Here is my man Westbrook taking one of these so-called "tough" shots. Notice - there is 19 seconds on the clock, a hand in his face and no NEED to take it:

(JIC: http://i.imgur.com/wKt24n9.jpg)

I think this unfalsifiable, and I fear we'll be stuck with definitional arguments as the "have to" we'll be a constant debate amongst us.

IMHO, giving certain guys a bonus because they "take the tough shots" is really hard to justify, but YMMV, and I'm betting Omar never comes back anyway.

108 days ago


You must be signed in to leave a comment.