| View previous topic :: View next topic |
| Author |
Message |
Mike G
Joined: 14 Jan 2005
Posts: 3618
Location: Hendersonville, NC
|
 Posted: Tue Dec 21, 2010 8:51 am Post subject: Suns-Magic-Wiz trades |
 |
|
Including aggregate stat profiles for 3 players each way.
| Code: | per36 rates mpg Sco Reb Ast Stl TO Blk 3s e82 eW e484
Richardson,Jason 32 21.0 5.3 1.7 1.2 1.4 .1 62 6.3 2.01 1.22
Turkoglu,Hedo 25 13.0 6.1 3.4 1.1 1.8 .8 41 3.4 1.09 .84
Clark,Earl 8 12.2 8.7 2.0 .5 1.0 1.4 0 0.5 .15 .96
to Orlando 65 17.2 5.8 2.5 1.1 1.5 .5 103 10.3 3.25 1.05
Carter,Vince 30 20.3 5.7 3.7 1.1 2.0 .2 27 5.6 1.78 1.30
Gortat,Marcin 16 10.1 12.2 1.6 .6 1.6 1.8 0 2.6 .81 .99
Pietrus,Mickael 22 11.9 4.9 .9 .9 1.2 .2 34 1.2 .37 .43
to Phoenix 68 15.2 7.2 2.4 .9 1.7 .6 61 9.4 2.97 .97
per36 rates mpg Sco Reb Ast Stl TO Blk 3s e82 eW e484
Arenas,Gilbert 35 17.1 3.6 5.1 1.4 3.5 .6 47 4.6 1.39 .92
Lewis,Rashard 32 14.9 5.4 1.4 1.0 1.6 .4 44 3.8 1.21 .73
|
Orl gets more scoring, from both Phx and Was; gave up rebounding.
Orl got slightly 'better' players (e484) in both trades; and they'll have a couple mpg to redistribute.
_________________
`
36% of all statistics are wrong |
|
| Back to top |
|
 |
EvanZ
Joined: 22 Nov 2010
Posts: 303
|
|
| Back to top |
|
|
DSMok1
Joined: 05 Aug 2009
Posts: 611
Location: Where the wind comes sweeping down the plains
|
| Posted: Tue Dec 21, 2010 9:19 am Post subject: |
|
|
Here's the ASPM look at it:
https://spreadsheets.google.com/ccc?key=0Ah1NfCUslJwxdHlzZVFXMWRxX29EUnZSamZFN0F1X1E&hl=en&pli=1#gid=0
A synopsis: the Arenas-Lewis trade was rather balanced: both are in the -0.4 to -0.8 range this year, Arenas is projected higher (better track record, also younger), but Arenas' contract is worse. It could be a good swap for both teams, depending on the fit. Both contracts are in the -$25 million to -$35 million range. (In other words, both awful.)
The Phoenix-Orlando trade is a little more one-sided, because of the contracts involved.
* Pietrus is not playing well. He's projected as a -1 player with a contract surplus value of -$5 million
* Carter is playing quite well. He's projected as a +2.2 player, which is almost enough to warrant his contract. The contract is rated as a -$4 million value.
* Gortat is rated as a +0.2 player (about a league-average center), and worth his contract exactly.
* Earl Clark is rated below replacement level (about -4) and his contract ends after this year. Worth -$1.2 million.
* Jason Richardson is a good player, rated at +1.0. He's about worth his contract; it's one year is worth -$2 million.
* Hedu Turkoglu has dropped to right about a +0 player, near league average. Will the change in scenery help? It had better, because his contract is nearly as bad as Rashard Lewis's. It's worth -$28 million over the life of the contract.
Will this trade help Orlando? Doubtful. I don't think Richardson is as good as Carter. They are almost identical players this year, except Carter has over twice the assists and plays on a WAY better defensive team. It MAY be a good fit, depending on how it works--Richardson is more of an outside shooter and less of a 2Pt shooter.
http://www.basketball-reference.com/play-index/tiny.cgi?id=vtFmX
Turkoglu is about as good as Gortat, at a different position, and costing $28 million more.
Pietrus is better than Earl Clark most likely ever will be.
I don't like this second trade for Orlando, unless the fit is really splendid. I don't know that it is, but we'll see.
_________________
GodismyJudgeOK.com/DStats
Twitter.com/DSMok1 |
|
| Back to top |
|
|
DSMok1
Joined: 05 Aug 2009
Posts: 611
Location: Where the wind comes sweeping down the plains
|
| Posted: Tue Dec 21, 2010 10:43 am Post subject: |
|
|
Another point I failed to mention:
Vince Carter in Orlando was assisted on 35% of his made shots. Jason Richardson in Phoenix? 70%.
That's another reason to think Carter, who managed comparable TS% and USG% with a higher AST%, is actually the better offensive player currently.
Maybe.
_________________
GodismyJudgeOK.com/DStats
Twitter.com/DSMok1 |
|
| Back to top |
|
|
Serhat Ugur (hoopseng)
Joined: 13 Oct 2006
Posts: 209
Location: Basketball Research
|
| Posted: Tue Dec 21, 2010 11:28 am Post subject: |
|
|
| DSMok1 wrote: |
That's another reason to think Carter, who managed comparable TS% and USG% with a higher AST%, is actually the better offensive player currently.
Maybe. |
Interesting. His expiring contract might have drawn more attention than he's been playing statistically.
_________________
http://www.nbastuffer.com |
|
| Back to top |
|
|
DSMok1
Joined: 05 Aug 2009
Posts: 611
Location: Where the wind comes sweeping down the plains
|
|
| Back to top |
|
|
EvanZ
Joined: 22 Nov 2010
Posts: 303
|
| Posted: Tue Dec 21, 2010 1:51 pm Post subject: |
|
|
Do you have to run that analysis manually? Or is it somehow automated for each game? I'd love to be able to use that for my Warriors blog. It would be much more interesting than using a Win Score-type metric like I currently do.
_________________
http://www.thecity2.com
http://www.ibb.gatech.edu/evan-zamir |
|
| Back to top |
|
|
DSMok1
Joined: 05 Aug 2009
Posts: 611
Location: Where the wind comes sweeping down the plains
|
| Posted: Tue Dec 21, 2010 2:19 pm Post subject: |
|
|
| EvanZ wrote: | | Do you have to run that analysis manually? Or is it somehow automated for each game? I'd love to be able to use that for my Warriors blog. It would be much more interesting than using a Win Score-type metric like I currently do. |
Yep, it's automated. I finally polished up the last few bugs (I think) this morning. It's included in the latest release of my spreadsheet: https://docs.google.com/leaf?id=0Bx1NfCUslJwxZjgwNTNjOTctZWYwYy00MzE3LWEzNTEtOTc3Y2Y5NTg2ZjM1&hl=en
Go to the "Single Game Analysis" tab and choose from the drop-down menu of games. Then click "Download Game Data".
There is one other input on the page: how many games of average data to use in normalization. This data is only used to smooth out the "granularity" of some of the data inputs (like assists and rebounds) that are nonlinear and would do weird things if, say, and AST% of 0 was used. What I do is add in the given number of games (somewhere between 2 and 10 seems to work well), compute the new ASPM, and BACK CALCULATE what the required ASPM would be in the game to get that new result. Fewer games for normalization allows the regression to move further out into the nonlinear valuation. I like the results I see with 3 games of average data--it allows an amazing performance (a 100% AST%, for instance) to move well out into the nonlinear part of the regression, but not so far out that the regression breaks.
_________________
GodismyJudgeOK.com/DStats
Twitter.com/DSMok1 |
|
| Back to top |
|
|
back2newbelf
Joined: 21 Jun 2005
Posts: 275
|
| Posted: Wed Dec 22, 2010 6:09 am Post subject: |
|
|
| In the Dallas game, of the players that were already in Orlando before the trade, one shot below 50% (he missed one field goal too much). The new guys shot 10 for 33. Arenas is now at a beautiful 3 for 17 for Orlando. At least you could say that the games were rather hard (@ATL, DAL) |
|
| Back to top |
|
|
bbstats
Joined: 25 Apr 2010
Posts: 46
|
|
| Back to top |
|
|
DSMok1
Joined: 05 Aug 2009
Posts: 611
Location: Where the wind comes sweeping down the plains
|
| Posted: Thu Dec 23, 2010 4:04 pm Post subject: |
|
|
| bbstats wrote: | DSMok, have you posted definitions for your values anywhere? I know ASPM how works in this table, and I know what VORP and Contrib stand for...but I don't really know how you calculate them.
Thanks! |
ASPM is simply performance above or below NBA average per 100 possessions. So if a +3 player is on the floor with 4 average teammates, the team will average about 3 points better per 100 possessions than 5 average players would.
Contrib is the player's actual contribution in this game, above or below average. It is calculated as ASPM*%min. So if a player is a +3 and plays half of the minutes, their contribution is +1.5. This is useful because the contributions must sum to the team's efficiency differential.
VORP is like Contrib, only it incorporates replacement level to get a better grasp of what the player is actually worth-- an average NBA player is worth something! VORP is calculated as (ASPM+3.5)*%min. I have estimated replacement level at -3.5; -3 is also a reasonable replacement level. The study and explanation of replacement level is far more advanced in baseball; look there for some of the theory behind it.
It's all rather simple, really...
_________________
GodismyJudgeOK.com/DStats
Twitter.com/DSMok1 |
|
| Back to top |
|
|
Italian Stallion
Joined: 04 Mar 2009
Posts: 112
|
| Posted: Fri Dec 24, 2010 3:51 pm Post subject: |
|
|
I think when you net everything out Orlando improved their bench from a scoring perspective with Arenas, but weakened it at C minus Gortat.
But more importantly IMO they have a better mix of players in the starting lineup and got a little younger.
Richardson and Carter are more or less an even swap or close, but playing Hedo with Richardson gives you more of a play maker than the combo of Lewis and Carter. To me, there was always some minor level of diminishing returns from playing Carter with Lewis because Lewis is mostly just a scorer and that's a big part of Carter's game also. That's why it never worked quite as well as some hoped.
Hedo was the perfect complimentary piece for a scorer like Lewis. Now he will be a perfect piece for Richardson. |
|
| Back to top |
|
|
Serhat Ugur (hoopseng)
Joined: 13 Oct 2006
Posts: 209
Location: Basketball Research
|
| Posted: Sat Dec 25, 2010 5:27 am Post subject: |
|
|
| Italian Stallion wrote: |
Hedo was the perfect complimentary piece for a scorer like Lewis. Now he will be a perfect piece for Richardson. |
Good point. That's why Otis Smith has undoed (ctrl+z) things by sending Carter and getting Hedo back.
Rashard Lewis has been performing better until Hedo went to Toronto.
_________________
http://www.nbastuffer.com |
|
| Back to top |
|
|
back2newbelf
Joined: 21 Jun 2005
Posts: 275
|
|
| Back to top |
|
|
Serhat Ugur (hoopseng)
Joined: 13 Oct 2006
Posts: 209
Location: Basketball Research
|
|
| Back to top |
|
|
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
