Page 1 of 1
So ESPN's trade machine now gives projected winning %. . .
Posted: Sun Feb 10, 2008 7:28 pm
by Crackfool
Interesting. . .
Posted: Sun Feb 10, 2008 7:31 pm
by Spykes
Huh?
Posted: Sun Feb 10, 2008 7:31 pm
by RIPskaterdude
Posted: Sun Feb 10, 2008 7:57 pm
by JES12
And this trade machine says if Dallas trades Harris/Stack/Hassell/Ager for Kidd, Mav's winning% would go down by 17.07 % and Nets would go up by 7.32%.
I guess it works just fine!!!!!!!!!!
Posted: Sun Feb 10, 2008 8:25 pm
by loserX
Based on what? I can make up my own numbers too. Unless I understand the formula or rationale, it's worthless to me.
Posted: Sun Feb 10, 2008 8:29 pm
by T-Spot
I guess it factors in the +ppg, rpg, apg of the trade and spits out a % based on that.
It is way flawed though. If I traded a 10ppg player and two 5ppg players, I could get Josh Smith and Atlanta's winning % will go up and mine would go down.
Just guessing here.
Posted: Sun Feb 10, 2008 8:32 pm
by JES12
It also changes too often. I just pulled that same trade again (less than an hour) and not Movs projecting % would drop by 18.29 and Mavs would be 39 & 43 by the end of the season.
Posted: Sun Feb 10, 2008 8:49 pm
by deeney0
loserX wrote:Based on what? I can make up my own numbers too. Unless I understand the formula or rationale, it's worthless to me.
Hollinger computes each change in win% on a trade-by-trade basis using an ever changing formula so that each number reflects his own personal biases and prejudices.
Posted: Sun Feb 10, 2008 8:52 pm
by Schad
Some fun 'stats' that I pulled from the ESPN TradeMachine:
If you trade Kobe Bryant, Andrew Bynum, and Jordan Farmar to Orlando for Rashard Lewis, Pat Garrity, Adonal Foyle, and J.J. Redick, it will result in no change whatsoever to the records of the two teams.
Trade the entire Raptors team to New Jersey for every Net except for Stromile Swift, and the new-look Nets will have a 60-22 record.
However, the Raptors roster would be a 30-52 club if in Minnesota.
...or 36-46 in San Antonio.
...or, 58-24 in Houston.
The ESPN TradeMachine's projections rival 'Dewey Beats Truman' for accuracy (that's my second 'Dewey Beats Truman' reference in two days...).
Posted: Sun Feb 10, 2008 8:57 pm
by loserX
deeney0 wrote:Hollinger computes each change in win% on a trade-by-trade basis using an ever changing formula so that each number reflects his own personal biases and prejudices.
Oh, well see, now THAT makes sense. I stand corrected.
Posted: Sun Feb 10, 2008 9:31 pm
by hermes
neat
Posted: Mon Feb 11, 2008 2:35 am
by blazerduck24
traded Lafrentz and D miles straight up for kobe.....
We won two more games?
Dont underestimate Raef's shooting ability i guess
Posted: Mon Feb 11, 2008 4:38 am
by gswhoops
Trading POB for Josh Smith and Dwight Howard apparently means 0 more wins for the Warriors.
Posted: Mon Feb 11, 2008 10:55 pm
by GopherIt!
Hollinger is a fool. He completely overlooked players like Craig Smith of the Wolves in his preseason PER projections. When I read his PER analysis I wonder if it doesn't actually stand for (this) Pointless Exercise is Retarded.
ESPN stinks on the whole.