Tuesday, July 24, 2007
The fait accompli strategy versus the “large trade”Posted by Patrick DiCaprio at 6:48am
Robert Greene, in his book The 33 Strategies of War, calls the piecemeal approach to winning a war the "fait accompli" strategy. The idea is that a series of smaller moves can be preferable to one big one.
There are many examples of successful piecemeal, or fait accompli, strategies in practice. One famous example is the history of Charles DeGaulle's rise to power as the leader of Free France during World War II. In fact, Churchill and Roosevelt despised DeGaulle, but DeGaulle managed himself so masterfully that once Churchill and Roosevelt figured out that this heretofore unknown and exiled general was now positioning himself to be the leader of free France, it was too late for them to depose him.
Even though they were allies, DeGaulle was imperious and had offended so many people that Churchill and Roosevelt wanted to be rid of him. But it was too late. He had started small, asking for a single TV broadcast. To Churchill's surprise, he ended this single broadcast by telling the viewers he would be on the air the next day! Since it was not worth it to fight a small battle, Churchill ignored it and allowed him to continue. In short order, DeGaulle was able to rile up the populace through these broadcasts, raise an army, occupy territories in Africa and propel himself to the world stage all by single, incremental moves. He executed a "fait accompli" strategy perfectly.
Now for its applicability to fantasy: One thing that I frequently hear on podcasts and see on message boards is the "large trade" where one owner is trading five guys for another five or sometimes more. In higher stakes leagues where there are strict salary caps, positional eligibility issues, etc., this can be a necessity. But in the garden variety mixed league it is much more likely that the owners making the deal are lazy or just not thinking about the process correctly.
As an example, on one podcast I recently heard a proposed deal that would send Ryan Zimmerman, Prince Fielder and Barry Bonds for Gary Matthews, Alex Rodriguez and Carlos Delgado, with throw-ins on both sides. On a message board I also recently saw a trade where an owner was going to trade Miguel Cabrera, Grady Sizemore, Aaron Hill, Freddy Sanchez and Xavier Nady (he also said he was in the middle of the pack in HR and SB)!
These are some trades! Those of us who don't play in a shallow mixed league never have the pleasure of making deals like this. On the other hand shallow leaguers do not have the pleasure of watching a low cost pitching staff get shellacked for a month, and believe me there is nothing quite like seeing a 3.1-6-5-5-3-0 pitching line night after night.
The problem with deals like these is that there is usually no viable strategic plan that requires trading players like A-Rod, Prince Fielder, or even worse Miguel Cabrera AND Grady Sizemore in the same deal. In a non-keeper league there is no point to dumping, and generally a team that wants to win isn't trading A-Rod. In a five-for-five deal the problem is exacerbated; there are few winning strategies that require a five-for-five trade as a routine part of the strategy.
The problem is largely one of statistical uncertainty. When you are dealing five-for-five and all are above average players as in a standard medium to low penetration league, it is simply impossible to get a good gauge on how the deal will affect one's team. There is simply too much "noise" in the actual stats for the rest of the year.
Typically the "large" deal is made because of laziness; the owners simply want a quick fix and don't want to spend the time working out a better deal or deals. If you are sufficiently strong that you want to trade A-Rod or Miguel Cabrera and Grady Sizemore is there any reason you have to trade four other guys to get a deal done? Worse yet is the situation where it is made for "entertainment" value. If you are in this game for fun there are few things that are as fun as making a big trade. With no stakes, this is fine, so go ahead. But I see this strategy in medium-and-high stakes leagues as a matter of course.
There are two basic iterations of the standard trade. When the league is shallow, the players being traded are mostly quality players. There will rarely be a trade that is so one-sided (that isn't vetoed) that one can make huge gains. You simply can't really change the value much when you are trading five quality players for another five. And yet there is no lack of trying. You simply can't expect to make huge gains in points when you are trading A-Rods for Prince Fielders, except in rare circumstances or if there is some significant luck involved.
The other scenario is a deep league. In the deep league the "large trade" makes even less sense. There is too much risk involved with the more risky players that are typically bandied about in deep leagues. In my high stakes league each of 11 owners has 63 players; there are a whole lot of risky players and who wants to predict what these players are going to do for a five for five deal? Calling this strategy risky is like saying sleeping with Paris Hilton is risky—it's a euphemism of the lowest order. Thankfully though, fantasy baseball rarely requires a visit to the free clinic and if it does, fantasy strategy is the least of your problems.
I have found that the big dump trade, getting rookies for veterans, is where the worst deals come from, and they usually significantly upset the balance of the league, cause ill will, and engender veto threats. In high stakes leagues this may be worth it, but in fun leagues or for lower stakes who needs it? The owner who is rebuilding is usually somewhat desperate to ensure he doesn't get stuck with certain players, and then botches the deal. In almost all instances he could have gotten far more value in a piecemeal approach, especially if properly planned from the outset.
There are three strategic reasons to prefer the piecemeal approach. The first is that the piecemeal approach maximizes the value gained more efficiently. It is my experience that it is much easier to gain $20 in profit over five deals than in one, unless it is a pure dump trade or the owner whom you are fleecing is weak. This is almost always true in deep penetration leagues.
Secondly, it spreads the risk. If you make a "large trade" you are essentially putting a lot of eggs in one basket. If you have any trading restrictions this is doubly true; the large trade usually limits your options going forward and no GM with delusions of victory wants that. It is much better to seek five smaller baskets than one large one; if things go south then at least you have other options. Moreover, against better competition it is very unlikely that a large deal will turn a large profit. If your adversary is astute enough to amass this much talent do you really think he is going to give it all up in one fell swoop?
A third reason is that it allows you much more flexibility in responding to what happens. If you are the first to act and make a large trade, you are essentially now at the whim of your fellow owners; you have fewer ways to respond. If you make a large deal and gain $20, and a competing owner makes a fleece deal for $30, you may have a tough time gaining that extra $10; the opportunity cost of trying for one huge deal can be prohibitive.
A piecemeal approach allows one to fly under the radar so to speak; start out early in the year while everyone is still letting things settle and you can make significant gains before other owners even realize it. No veto threat, no ill will, just the ability to shore up your weaknesses without drawing undue attention, and counterattacks.
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.