Tuesday, January 08, 2008
More on value and the “learned minority”Posted by Patrick DiCaprio at 7:13am
In response to my last column, the most frequent question I have been asked is whether I am merely saying that we cannot judge whether we have gotten value at the time of the auction. Yes I am saying that, but on a deeper level what I am saying is that the basic paradigm by which players are obtained is flawed especially at the top of the auction pyramid where players are not fungible.
At the top of the auction pyramid where the top talent lies the concept of value has to basically go out the window in keeper leagues where inflation is often rampant. If you are hoping to get a Jose Reyes or Jimmy Rollins type player for under $30 so that they have value then you have little chance of getting the top talent necessary to win in tough leagues.
Anyone who has been in an auction has experienced this common scenario: you have a guy projected at $20. Bidding starts and reaches $20. You drop out and someone gets him for $24. That player goes on to produce a $35 season. Every year there are a few of these guys. Often owners who don't think about such things consider that the successfully bidding owner was just lucky. Perhaps he was but perhaps he knows a lot more than he is given credit for. Evaluation of your adversary is always to be considered.
Question: (and I apologize for channeling Dwight Schrute for a second) why did you drop out of the bidding? Invariably the answer is that the player wasn't a good value at that price. Most times that may be true and in the long run it will be true; namely if I take a group of 100 players all of whom are valued at $20 and pay $24 I will probably show a loss, assuming my projections are good.
But every so often you feel like you have inside information or see something that others do not. Often this might be using a regression to the mean approach for BABIP or hit rate or some other solid information that you have. In this regard you may be a "learned minority." I have discussed at length my trade of Stephen Drew or the fact that two expert owners went head-to-head in a bidding war for Tom Gorzelanny this year, with the other owners denouncing them as crazy (and these two owners finished one-two this year).
Every year there are a whole host of such players that I target at the auction, maybe 5-7 each year. Every year I invariably get almost all of them. Why? Because value is not the proper way to acquire players in every instance and I am willing to go beyond it for these players when other owners are not. I am happy to acquire players I love and feel like I have great info on if they go for anywhere close to my projections.
A few examples aside from Gorzelanny: two years ago I acquired Brian McCann in my high stakes league for $12. I had him projected at $10, and would have gone to $15 (he produced $25). Last year I had Jake Peavy projected at $23. I bought him for $30. I also had Jimmy Rollins at $30 and got him for $33. I got James Shields at $9 but would have gone to at least $13 for him. All outproduced their projections and my purchase price. If I considered value for these guys I would have gotten none of them.
As Wittgenstein might have said if he played fantasy baseball, value is just a symbol. Our goal is not to maximize value but to maximize production and statistics. Value is a shorthand symbolic way to maximize your teams production but it is no substitute for actual evaluation of points to be gained and statistics to be accrued and dollars to be earned.
I am not saying that you should not chase value. What I am saying is that the pursuit of value is a mirage that can cloud your view of the overall goal to be achieved, the maximizing of production. My personal view is that a "go get them" approach with regard to the players you really want will often do better than a value based approach, especially if you are particularly astute and have excellent judgment about players. This advice is not license to go overboard however! We are talking here about non-fungible players at or near the top of the auction pyramid.
The fact is that a value based approach is correct when you are dealing with fungible commodities. But at the top where the big money is spent you will usually do better to target specific players that you desire if (and that is a big if) you really do see aspects of that player that others in your league do not. That means not only must you consider your own values but you must consider what your opponents know and what they will do with that knowledge.
One final anecdote that is somewhat reflective of the same principle: in my days as a horseplayer, I found a particular public handicapper who was very astute at picking longshots. I considered him to be a "learned minority." What I would do is look for situations where he picked a horse to win that no other public handicapper even put in the top three. This yielded tremendous profits. It was a good lesson in thinking outside the box and in realizing that sometimes you must act on knowledge that you have that others don't regardless of other factors.
Patrick is a member of SABR's Statistical Analysis and Science of Baseball Committees and writes about fantasy baseball at The Fantasy Baseball Generals blog. He has achieved the dream of all of his MIT classmates. No, not making millions in the tech markets, but writing about baseball for free. Feel free to send along all insults and comments here.