Because the goal of your exercise is less to predict the price every player goes at, and more to get yourself a team you like. (Here's hoping you like good players!) The auction dynamics are impossible to predict 100% (though knowing your league-mates can help), but giving the "extra" money to players you have a feeling will exceed others' expectations can get you where you want to go.
I hope I didn't give the impression that the goal of setting your bids is to predict the price of every player. I agree with Mike that, if you do this, you'll wind up with a team you don't like and you won't be drafting based on your hunches or intuition. Generally, you don't win this way and, worse yet, you won't have fun rooting for someone you consider a stiff.
However, you do want to have an idea of what each player will go for. If you have Jermaine Dye down on your sheet for $45 pre-inflation this year because a) you believe he'll exceed last year's numbers b) you're a huge White Sox fan and c) you believe that 33 is a magic year for hitters, you will be sorely disappointed.
This goes back to my earnings/predictions/bids discussion. You want to know that Dye earned $36 last year. You'll also want to know that he earned $25 in 2005 and $16 in 2004. While it's possible that Dye might duplicate his 2006 career year, you shouldn't predict that he will. Your bid probably should shave a couple more bucks off of whatever his prediction is, given Dye's age, injury history, deep level of talent in A.L. outfield, and the fact that Dye is pretty much a two category guy.
Another reason that you don't have to worry too much about figuring out what each player will be purchased for is that we've finally reached the point where almost every expert under the sun has almost the same exact opinion about each and every player. And this is reflected in the fact that the bids are getting closer and closer to par, across the board, every year I've played.
This is what Alex Patton calls Stage 3; I'll leave the long description to him. The short description, which I'll provide, is that we all are within $1 or $2 of each other on what every damn player in the universe is worth. Or what he'll go for. This isn't to say that valuation isn't important. It's extremely important. It's just that there isn't much wiggle room left on how much a player is worth.
So, in a way, you almost are predicting what you think your league thinks a player is worth, because your league has the same opinion that you do. In a manner of speaking. And you can tweak your bid $1-3 up or down to express preferences and draft a team you like. But that's usually about as far as you can go one way or the other. Too many $45 Jermaine Dyes and you'll lose every year.