Hits are added.

You don’t want to adjust the averages, SDs or replacement levels based on your bench players, since they’re not actually contributing stats.

When calculating value you only care how much better Albert Pujols is than the worst 1B who is in a starting spot. Pujols’s stats don’t become more valuable in relation to the other contributing players because there’s a bunch of scrubs on the bench not contributing. Replacement level isn’t really “replacement level,” it’s more of a baseline adjustment to who the bottom contributing player is, since the worst contributing player at a position is worth $1. Think about it – if you’re drafting for 2008 league today, you’d pay $1 for the worst player at a position that will be in your lineup. No reason to draft bench players at all, since you know exactly what each person will produce.

With that in mind, you’re never going to spend more than $1 for a bench player (or at least you’re not going to value them more than $1).

So you assume you’ll spend $1 per bench player and just take that out of your available budget.

Of course, this is different than how things work in practice. The person who drafts Chipper Jones might also pony up for a better-than-replacement level 3B to have on his bench for when Chipper gets hurt. Maybe Mark Reynolds or Jorge Cantu or something, rather than Bill Hall. So maybe you should change replacement level based on bench players a little.

I still think the best thing to do is to assume that your bench players will be $1, and adjusting your budget accordingly, and making any emergency adjustments based on what happens on draft day.

Also, how do I deal with bench slots? I’m in a league that auctions off those positions as well. I’ve just been splitting them evenly between Util and P, but I’m concerned that may throw off the replacement levels for the other slots.

Thanks again!

Blank teams get counted as either AL or NL. That was a bigger deal earlier in the offseason when there were lots of free agents that could really end up anywhere.

Injured players will skew the rankings, but not if you edit their projection to something closer to what you think will happen.

Also, what about the blank teams – don’t those get included in both leagues when you do an AL-only or NL-only analysis?

And, what about Ben Sheets, who’s going to miss most of 2009? Doesn’t having his projections included skew the values of the other players?

What I’d really like to see is a way for us to upload our own spreadsheet of raw stats and then apply the price guide to them. Might be a tall order though…

@Tyler: ERA (as well as WHIP and AVG) is computed based on how that player did compared to what an average pitcher would have done in the same number of innings.

For Lincecum in a standard league, an average pitcher would have given up 78 ER in 227 IP, because the league average is 3.08. Lincecum allowed 12 runs fewer than an average pitcher.

An average pitcher would have given up 23 ER in 69 IP (a 3.08 ERA again). Papelbon only allowed 18 ER to finish 5 runs above average.

So the Price Guide gives about twice as much credit to Lincecum for his ERA, because it prefers his 12 runs above average to Papelbon’s 5 runs above average.