Computing R$ for HR/RBI players

[Michael@HQ]

Can you explain why the R$ calculation on the projections seems to devalue HR/RBI guys compared to if I use standings gains points? Here's a specific example.

[RobR]

I wouldn't be able to take a hard look at this until the weekend, but here's what immediately comes to mind.

1) Aviles' lead in SB is worth more than Thome's lead in RBI. The difference maker in the SGP view is the HR. I would have expected 9 points of AVG over 130 more AB to be worth a bigger difference in SGP. That's where I'll start looking.

2) Did you calculate SGP for marginal stats above replacement level? There's an implicit level of 270 built into the AVG formula, but not in the counting stat formulas. Doesn't introduce a difference in their separation, but does affect the absolute levels

[Michael@HQ]

I calculated SGPs using the formulas from Gerald Holmes' recent article. Send me a private message with your e-mail address if you would like my simple Excel file.

I understand that BBHQ isn't using SGPs, but this was a big enough discrepancy that (i) I wanted to make sure there's no error involved and (ii) it might be worthwhile for users to know that an alternative rational valuation method would assign a higher dollar amount to a guy like Thome.

[RobR]

Took a deeper look and here's an approximation of how CDG allocates the total dollar values over the scoring categories:

Name	Avg	HR	RBI	SB	Total
Thome	$0.01 	$7.89 	$5.56 	-$0.15	$13.31 
Aviles	$2.74 	$2.76 	$3.96 	 $3.71 	$13.16

Does that help at all?