Prisoner's Dilemma
Probably the most rudimentary example of economic game theory stems from the story of the Prisoner's Dilemma. While it's the most rudimentary, it's a bit too long of a lesson for here, so if you have an interest I'll direct you over there. Either way, the implications of the theory are pretty obvious for our purposes.In game theory, a unit can choose to either "cooperate" or "defect". If both players cooperate, the gain for each is "medium". If both defect, the gain for each is "small". Finally, if one defects and the other cooperates, the defector gains a lot while the cooperator gains nothing. Now consider this when applied to fantasy baseball, or any fantasy sport really...
- Cooperating: Making a relatively fair trade that helps both teams. This often involves trading surpluses for surpluses. As a side note, it's good to go into the season knowing at least a few members of the league, or at least arranging some trading partners during the draft. With the mentality of cooperating, you can draft value and worry less about need.
- Defecting: These are the trades when you've ripped someone off. Last year I traded a package centered around Delgado for a package centered around Wright. It was in May, and my team never looked back. Needless to say, these are the harder trades to pull off, but they are the trades that win you leagues.
Labels: Economic Editorials
posted by Zachary Piso at
9:31 PM
3 Comments:
In response to the question of whether one-sided defective trades are unethical, I answer no. Even the trades which seem equal result in one player scoring more points for his team than the other. If a trade results in one player outscoring another by a small amount why should a larger gap be considered unethical? Granted, I don't truly understand how fantasy baseball is scored, but it seems like common sense. Unless a trade is being made by conspiring parties attempting to make one person win and split the winnings, all trades in which there is agreement between parties should be allowed. Who knows, maybe the guy we think is being swindled realizes that Ken Griffey Jr. is injury prone and deals him right before he sprains his knee and is out for the season. Even if he only got a low-level starter from the Brewers, he still ended up with a better deal. You never know.
I'm not big into league regulation of deals. I've seen some good and bad deals go through, and unless there's any obvious collusion, I think you have to let trades go through. The individual owners can't assume they're getting a bad deal if they both agree to the trade, and with baseball, you never know who can get hurt and make a one-sided deal suddenly look great for the other team.
Grabbing Brandon Phillips early last season from the waiver wire was a better heist than any I made from a team (OK, I didn't keep him all year, but it was still nice while it lasted). I also figured getting Manny Ramirez for Roy Oswalt was a huge score, but it didn't pan out. Two years ago, I moved Jonny Gomes at the end of his power streak for Bobby Jenks... again, a good move, but not a true "heist." I guess my greatest ever move in fantasy was getting A-Rod with the 11th pick in the draft last season. 11th. What else can I say?
Now I think one point that is being glossed over is the fact that in a baseball league with an unknown number of trades, we are talking about the iterated prisoner's dilemma. This is relevant because, if you don't know the number of trades, the logic doesn't favor a defect-continuous strategy. I'd like to detail a little bit of the economics here...
If, on the last trade, your opponent has no ability to retaliate against a "defect", the strategic move is to "defect". Hence, both players will logically defect on the last trade. But, if on the trade before that, you know that on the next trade your opponent will defect, you might as well defect already since you have nothing to lose. This continues back to the first choice, since at its succeeding decision the logical choice is to defect. However, if you do not know where the last trade is, then no such argument pertains.
I think this is relevant for a number of reasons, but the one I will give you is this. If I make a trade where I rip off another player, the league will see this as unethical, or in any event consider my team too much of a threat to even cooperatively trade with. Hence, each defect sends a signal that damages my chances of another defect. There is no way of knowing how many of these signals must be sent to result in a "peaked" team, but in any league that number is attainable, and must be considered as an opportunity cost of any deal.
Post a Comment
<< Home