I could swear I read somewhere recently acknowledging that steals lose value when the draft ends. It’s something that I came to the realization of the hard way when trying to trade Juan Pierre a couple of years ago, but wasn’t sure if it was correct or if it was, then why. I now believe that steals, mainly from players whose value is mostly derived from steals, are indeed worth less post-draft and actually should be.

There are two things to look at in this issue, and I hope they are good enough to explain what’s happening here. First we need to look at the values we calculate going into our draft and how much the SB-heavy player was auctioned for. I’ll assume an auction league (12-team, 5×5 mixed, standard rosters) since that’s what I prefer and it’s easier to understand what’s going on. I usually find that SB-heavy players could be purchased at a discount to my calculated value, which means that my competition isn’t valuing steals correctly to begin with, but that’s another issue.

Let’s say I valued Juan Pierre (assuming he was the full-time starter) at $25, which is around what he’s earned in previous years. I would probably be able to get him at around $20 at the auction, since as I said above, I used to always get him at a discount to my value. Post-draft, if I wanted to trade him, I should expect to at least be able to get a $19 player in return (since that’s what the rest of the league valued him at, or at least the owner who bid $19), assuming I found an owner who made for a good trading partner based on our needs. I could understand not receiving $25 in value back because clearly the rest of the league only values him at $19. Without being able to provide specific examples, I have found that I am unable to trade Pierre for fair value based on his auction price. Maybe I could get a $12 player in return, but I would expect more than that. After seeing many trades in my leagues go down when a SB guy is included, it’s obvious that I’m not the only one not getting value that matches what is paid at the auction.

So why is this happening? The first thought many of you might have is that “Well duhhh, didn’t you ever think that maybe your value for Pierre is just wrong, and you actually overpaid for him at the auction?!” Well sure, that’s always a possibility, but then I’d have to blame the valuation method I use, which I am confident is the best out there, at least for valuing players for the auction. So let’s just pretend my value is accurate given my projection and the projections for the rest of the player pool.

Post-draft, but before the season, and for the first month or so, I almost strictly go by dollar values to evaluate a trade. I’m still in value-accumulation mode, assuming I could worry about needs later. But once the season gets into May and June and the standings start to take shape, evaluating a trade is easier because you could actually estimate how many points in the standings you would gain/lose if you swapped hitters. Most likely, the gain you’d enjoy from the SB boost will be negated, or close to negated, by the loss you’d suffer in HRs and RBIs. If you were the one acquiring Pierre, you might expect to gain 6 points in SBs, but then lose 3 each in HRs and RBIs, of course depending on the player you’d be trading. If we were to estimate the point changes by trading Pierre for a $12 OF versus a $19 OF, you’ll probably see that the swap for the $12 OF actually might be completely fair, wherein both teams’ gains and losses in different categories nets out to 0 gain or loss in the standings, while the trade for the seemingly fair $19 power OF might net a gain for the team acquiring him and a loss or no gain for the team acquiring Pierre.

This might actually seem like an argument to completely throw out your auction values once the season begins. I’m not arguing this, however, because they should still be used when comparing players who contribute in the same categories, and especially during the first month of the season where you are still just looking to acquire the most value, figuring you’ll worry about needs later in the season.

I still haven’t really answered the question of why Pierre’s in-season value seems to fall from $25 to like $12. One explanation for this is that dollar values are calculated in a vacuum, and assume you have no other players on your team. Once you start constructing your team, everyone’s value changes based on what categories you have accumulated too much/enough/too little of. For example, when you acquire Pierre, all of a sudden the value of steals in the auction drops for your team, while increasing the value of HRs, which unless you are using a dynamic value spreadsheet that adjusts values during the auction, you wouldn’t realize or be able to quantify exactly how much values change. Since you have acquired Pierre, you are smart enough to make a mental note to start buying power and not continue to buy guys like Reyes and Figgins (unless they come at large discounts), and are therefore building your team around Pierre knowing you will be competitive in steals.

OK, after sitting here staring at the screen realizing I really have no idea what I’m saying, I think I have figured out the answer- One of the most popular valuation methods I’ve read about deals with the standard deviation between the stat totals of each team in each category. This is the Standings Gain Points (SGP) method devised by Art McGee that looks at how many of each stat would move you up one place in the standings in that category. I used to use this method before reading about its flaws, at which point I stopped and switched to Todd Zola’s. Although steals are rare, which increases their auction value, the number of steals during the season that you need to gain a point is similar to HRs, and probably smaller than Runs and RBIs. If we go by that, then in essence 1 steal should be just as valuable as 1 HR!

So maybe I am proposing having 2 sets of dollar values for your league- one using replacement level theory for the auction, which is what I use, and one using the SGP method during the season. The criticism of the SGP method has always been that it doesn’t value steals correctly (not high enough) and that you can’t assume 10 steals (or 10 of any category) will gain you a point because you first need to acquire enough steals to get to at least a typical last place steals total before you could possibly gain points in the category.

I wish I was able to look at this mathematically, because to be honest, I’m really not sure how to best explain the issue or if the above made any sense. But I’m very confident that steals do lose value after the draft and owners should understand this so they don’t get frustrated, like me, when they can’t get seemingly fair value in a trade. I’d love to hear your thoughts about this and any explanations you might have for the phenomenon.

Questions, comments, criticisms, praise, expert league invitations? Email me at FBGeneralsMike@gmail.com.