SHOCKandAWE wrote:By dividing the gap of homers by teams you get a unit per point. for instance in your league it take 10 homers to gain 1 point in the homers cat. So by just having A-rod on your team all year he would give you 4.7 points higher in the homer cat than not having him. Like I said before there are many more factors, however this is a simple way of explaining it to you. Let me break downyour units first.

It takes 10 homers to gain 1 point in the HR cat. It takes .0017 to increase 1 point in the AVG ct.(harder to predict) It take 21 Rbi to gain 1 point in your runs cat. It takes 22 runs to gain 1 point in the rbi cat. It takes 12 sb to gain 1 point in the sb cat

Now lets see what A-Rod is really worth:

HR is worth 4.7 points AVG cant be determined because its the team average. Cumulative is tougher to use. RBI is worth 5.6 Runs is worth 5.6 SB is 1.6

If you add this up then A-ROD is valued at 17.5 roto points in your league. The key is to do this for all your players to get a ranking. Once you get a ranking you can see how a player helps or hurts you team if he is plugged in. They can also have negative points per cat. Take a guy like Pierre who hits no homers. He may have a negative point total. So if you have a spreadsheet you can determine th league average and base it on this.

I think there's a detail missing here: you need to include a baseline performance. A lineup of 14 hitters each hitting 10 HR isn't worth 14 points in that category.

Similarly, a player projected to hit 10 HR and steal 12 bags (similar to Derek Jeter's '03 numbers), shouldn't be considered to be worth one point in each of those categories, since he'd effectively hurt his team in HR, yet help (slightly) in SB. Also, without that baseline, you won't be able to assign the negative values you mention.

If Arod is worth lets say 17 points
and the 10th best SS is worth 8 points then you can say that having Arod instead of the 10th best is going to give you roughly 9 points more. Now there are so many factors to conisder but I think this is a decent way to evaluate a player

Also lets use your example of 14 players averaging 10 homers with an overall total of 140 and Team B is averaging 5 homers a person for a total 60 points. You can expect to be roughly 6 points high in the home run cat than team B

$42 Curt Schilling $36 Pedro Martinez $33 Mark Prior $30 Mike Mussina $29 Kerry Wood

$42 Eric Gagne $27 Billy Wagner $26 Octavio Dotel $24 Keith Foulke $20 John Smoltz

$32 Todd Helton $25 Jason Giambi $23 Jim Thome $23 Carlos Delgado $21 Richie Sexson $21 Derrek Lee

$33 Alfonso Soriano $19 Bret Boone $17 Jeff Kent $13 Orlando Cabrera $13 Marcus Giles

$32 Alex Rodriguez $24 Nomar Garciaparra $22 Edgar Renteria $20 Derek Jeter $19 Miguel Tejada

$21 Eric Chavez $20 Scott Rolen $14 Mike Lowell $14 Morgan Ensberg $13 Hank Blalock

$14 Javy Lopez $14 Ivan Rodriguez $11 Jorge Posada $12 Mike Piazza $9 Jason Kendall *Funny how everone thinks that catchers are worth much more. Is the top catcher worth the same as the 50th best OF??

$36 Albert Pujols $32 Gary Sheffield $31 Carlos Beltran $30 Barry Bonds $28 Vladimir Guerrero* *Vlad was ranked lower because I lowered his projections due to concerns with health.

When the top two players in your valuation system are Schilling and Gagne, you know there is a major flaw in the algorithm. Your system obviously didn't factor replacement value into the calculation among other things.

All this number crunching and statistical analysis is nice, but personally I don't want to turn my fun little hobby into a big math problem.

People claim to win leagues with their formulas and player projections, but the league winner usually is determined by good drafting and shrewd player management, not 18 hours of number crunching and generation of a 'holy cheatsheet'.

[b][size=18]70*[/size][/b]
* Steroid induced record

The system itself seems to be flawed, and so do your projections. Curt Schilling worth that much more than Prior and Pedro? Mike Lowell and Morgan Ensberg tied in value Lowell is probably going to be the 2nd or 3rd best 3B behind A-Rod and maybe Rolen. Ensberg is probably 9th or 10th best 3B.

Making rankings that are based on replacement cost, position scarcity, batting trends, 3 year averages, age, etc are way too complicated for any normal person to compute. I'll leave that up to Xeifrank and Hootie. If somebody found a proven formula to come up with dollar values, I'd be interested in seeing it. Xeifrank, got anything like that? Although personally I'm pretty convinced that no formula can't calculate a player's worth. There are way too many factors, and I doubt anybody has the amount of time it would take to come up with and consider them all.

"Jack, will you call me, if you're able?"

"I've got your phone number written, in the back of my Bible."

dannahann wrote:I'll warn you right off that most of these cats had tight bunchings and then huge gaps. The pointbreaks wouldn't have been evenly distrubuted. RUNS 1088-852 (236) RBI 1075-804 (271) HR 307-213 (94) SB 156-50 (106) AVG .297-.273 (.024)

WIN 100-79 (21) SVE 119-36 (83) K's 1310-951 (359) ERA 3.27-4.46 (1.19) WHP 1.18-1.33 (.015)

LBJackal wrote:The system itself seems to be flawed, and so do your projections. Curt Schilling worth that much more than Prior and Pedro? Mike Lowell and Morgan Ensberg tied in value Lowell is probably going to be the 2nd or 3rd best 3B behind A-Rod and maybe Rolen. Ensberg is probably 9th or 10th best 3B.

Making rankings that are based on replacement cost, position scarcity, batting trends, 3 year averages, age, etc are way too complicated for any normal person to compute. I'll leave that up to Xeifrank and Hootie. If somebody found a proven formula to come up with dollar values, I'd be interested in seeing it. Xeifrank, got anything like that? Although personally I'm pretty convinced that no formula can't calculate a player's worth. There are way too many factors, and I doubt anybody has the amount of time it would take to come up with and consider them all.

I'd say dollar values are the easy part (not that traditional calculation methods work perfectly for many of today's leagues, but that's another matter); coming up with projections to base those values on is the problem...