A couple weeks ago, a discussion arose in the comments section of Paul Singman’s Keeper Reaper article about the relative value of keepers in leagues that have a salary system. Given the polarity of the responses, I thought this deserved a thorough look in an article of its own. To catch you up, here’s how reader abskippers began the discussion:

I think I made this criticism last year, and intend it to be a constructive one, but the Keeper Reaper series would be much more helpful if it considered potential keepers in terms of dollar value. This segment essentially treats Cespedes and Curtis Granderson as keepers of equivalent value, but a lot of owners could be keeping Cespedes in the single digits, and Granderson in the $30's. And Cespedes in the single digits is probably a no-brainer in any league, but there are plenty of more marginal decisions out there. What do you do with a $34 Verlander vs. a $5 Chris Sale? Or a $45 Miguel Cabrera vs. a $4 Adam Dunn? I would just suggest including a range of dollar values at which the player makes a good keeper and associated discussion in addition to the shallow/deep format.

The question here essentially boils down to this: Is it better to lock in a superstar at his going market rate, or is it better to lock in a lesser player at a below-market price? If you ask grandslam28 it’s, “Clearly Sale at $5. Probably Dunn at $4.” DarinRuf18, though, says, “I’m the opposite. I think it’s clearly Verlander and Miggy. You can’t beat having the #1 pitcher and the #1/2/3 hitter on the same fantasy team.” So that leaves us with one reader firmly planted on each side of the debate with another unsure in the middle.

I’m actually rather surprised to see such diverging opinions on this issue. For me, it’s very clear that you take the player who is the best value, regardless of absolute talent level. I can certainly see how it can be tempting to automatically start next season with a couple of the best players already on your roster, but if your goal is to actually win the league, this is a mistake.

The first thing we must understand is that, to win a fantasy league, it is completely and utterly unnecessary to own the best player in the league. In fact, it’s unnecessary to own any top-tier player. In the vast majority of leagues, I’d go as far as to say that owning such a player is a hindrance. Not in the absolute sense, of course, but in a relative sense. I’m talking about opportunity cost. If there is a blonde and a brunette at the bar, unless I’ve got serious skills, taking the blonde home is going to preclude me from taking the brunette home, and vice-versa. The opportunity cost of selecting the blonde is not getting the brunette. The opportunity cost of keeping Verlander is not keeping Sale.

By spending $30-40 dollars on an elite player, it’s removing $30-40 dollars from my available resources to spend on other players.

Player value is relative, and the primary goal of any draft should be to spend your money as efficiently as possible, to acquire as much value as possible. While there are strategic and categorical concerns to take into account, all else equal, the drafter that gets $400 worth of value for his $260 auction dollars is going to finish higher than the drafter that only gets $300 worth of value. By keeping Verlander for $34, you’re not gaining any value. You’re paying $34 for $34 worth of stats. It feels good to have a player of his caliber on your team, but the trickle-down effect on the rest of your roster will have serious consequences. It would be far better to spend $17 each on two pitchers worth $22. You won’t have the name value of Verlander, but you’ll have gotten $44 of value for the $34 you spent.

In Tout Wars and LABR last year, I drafted just one elite player each (Justin Upton for $38 in Tout and David Wright for $30 in LABR), instead focusing on finding values like Ryan Ludwick and Jon Jay for $7 in the middle rounds. In a non-expert league, I’d likely forgo those Upton and Wright buys. Experts are super savvy and bargains are hard to come by, so I have to spend my money somewhere, but in a normal league where bargains are easier to find, I’d ignore all of the top players and build a super mid-range-heavy roster. It won’t be a sexy roster, but it’ll be an effective one. (For what it’s worth, I finished in the money in both Tout and LABR this year).

In keeper terms, the opportunity cost relates to the value of the keeper slot. You may only get to keep a few players, so you need to use those spots to acquire as much free, pre-auction value as possible. If you really want a top pitcher, maybe even Verlander himself, you can always pay the sticker price during the auction without letting go of what might be $15 or $20 of free value from Sale.

Now you see why it might be difficult for Keeper Reaper to cater to auction leagues, as abskippers suggests. Value is simply too relative to 1) your other keeper options, and 2) the price of the player. Basically, your keeper decision in an auction league should boil down to how big of a value you’re getting relative to your other options. It’s impossible to pick a definitive point (or at least one that isn’t obvious to everyone) where, yes, you keep him here, but no, you don’t keep him here. It’s more of a continuum. If you think Sale will be worth $20 next year, it’s terrific to keep him for $5, but it’s also great to keep him for $10. It’s also worth it to keep him for $15 or, depending on your other keepers, maybe even for $17 or $18. It’s not worth it to keep him for $30, even if you have no other options. For Keeper Reaper, the best we could offer would be a break-even point, but those can vary wildly from league to league depending on setup.