Tuesday, June 05, 2007
WHAT MAKES A SUCCESSFUL FANTASY GMPosted by Patrick DiCaprio at 5:38pm
By way of introduction, I run The Fantasy Baseball Generals blog, in which the writers try to illuminate the underlying core principles of fantasy baseball by considering it as a “game” (in the mathematical sense) or warlike conflict. The core principles are viewed through the prism of various disciplines such as game theory, military strategy, economics, and even the writings of Sun-Tzu. When I was asked to write a weekly column for THT’s Fantasy Focus I naturally jumped at the chance.
For my first weekly column I wanted to address what I believe is the most important goal of any Fantasy player. The goal happens to also be the answer to the question posed in the headline.
The answer is obvious to most: a successful GM is a “winner.” Or at least that’s how most fantasy players would answer. Yet, they are wrong. Surely, being a winner is a hallmark of the expert fantasy GM, but it is only a necessary criteria not a sufficient one. Winning requires luck. I have yet to meet the fantasy GM, expert or otherwise, who could properly judge the vagaries of chance. No one can win without a modicum of good fortune. So mere winning is not the answer.
Another popular answer is that “knowledge” makes an expert fantasy GM. Again, we only achieve half the battle. Knowledge is certainly a necessary condition; however mere knowledge will get you little but perhaps the show money (to borrow a horse racing term). An owner could know the batting average, by heart, of every player on the Marlins roster; this doesn’t make him an expert or a success. In fact, the true expert needn’t even know any of this information since it is readily accessible to all at the stroke of a computer key. Even Einstein, perhaps apocryphally, didn’t bother to memorize his phone number.
The answer is that to be a successful GM one must become an “expert,” and there are plenty of supposed experts out there. Just searching Google for “fantasy baseball blogs” I found hundreds. There are scores of podcasts and analyses from supposed experts. One should be cautious of taking advice from non-experts, and yet how can you tell whether one is a real expert? The criteria below must inform one’s assessment of expertise:
1. Knowledge. The expert knows things that others don't, and I don’t mean batting averages but analytical tools and strategies. The expert has knowledge on many matters relevant to the process at hand and is always gaining more knowledge. An example would be that an expert can provide a reasoned answer to many questions, such as "why is it a fallacy that hitters hit worse with two strikes," or "in fantasy, why should I not chase wins" or "who will emerge from the Phillies' bullpen with saves." They may come to a different conclusion than other experts but they will evince a thorough knowledge of the issue.
2. Strategy. A true expert will know and use strategies that non-experts don't even know exist. To borrow from the writings of David Sklansky and Mason Malmuth on poker (from whom this example comes,) an expert may use a game theory based bluffing system that non-experts don't even know exists. In fantasy baseball, an example would be knowing about the Low Investment Mound Aces strategy. Whether a particular strategy is good or bad to use in a given context, the expert knows them and can use them should the need arise.
3. Independence. A true expert will deviate from conventional wisdom and even his own typical strategy should the situation be favorable; he will not be tied down to any preconceived set of rules or the conventional wisdom, and will trust his or her judgment over all else. It may just be that he has a different opinion than everyone else, or it may be that it is for a specific tactical reason. An example: an owner that rarely spends significant money on pitchers in an auction league is willing to bid $30 on a pitcher this year (an example dear to my heart since this year was the first year in ten that I paid $30 for a pitcher; it was Jake Peavy). If you are bidding against that owner, and you have the knowledge of his bidding tendencies (and how can you be an expert if you don’t have this knowledge?) you would be well advised to bid $31.
4. Flexibility. A true expert can use many different strategies, can tailor his planning to given situations, and can take advantage of a situation as it presents itself through judgment and knowledge. The expert doesn't need to read specific advice from trusted websites or other sources of information in order to come to a conclusion, though they may be invaluable; he or she can make good decisions based on a thorough knowledge of fundamental theory. There are an infinite number of situations that can arise. One would be foolish to, for example, follow trusted dollar value projections to the letter regardless of their judgment, but many owners do just that.
5. The Thought Process. Last, and by far the most important, is the thought process. The expert considers many variables, and can sift through them in a logical fashion to come to a reasoned decision.
Here is an example of expertise in practice:
In a high stakes auction league that I am involved in, (in which each owner has one AL and one NL team) this year Tom Gorzelanny and Adam Loewen were auctioned for $12 each, and James Shields went for $9. In all three cases, two owners that I consider to be experts or close to it were involved in the bidding, and no other owners were seriously involved. In all three cases other owners not involved in the bidding were incredulous and making their disapproving comments and opinions known.
In Loewen's case, (which hasn't worked out so far--but I focus on process not results; more on that later) his poor superficial numbers hid what was an excellent pitcher. He struck out approximately 8 batters per 9 IP, and got two-thirds of his outs on ground balls and strikeouts. In Gorzelanny's case, he had superficially good numbers, had good minor league translations and had a FIP of 4.16, again showing that he was for real and not a fluke. Shields had surprisingly solid core skills; including 7.5 strikeouts per nine innings and 2.7 walks per nine. More importantly he was unlucky last year with a 33% hit rate, causing there to be a perceived discrepancy between what these two experts saw as his potential and what the others saw as a rookie with a 4.80 ERA. There is of course more to it than that, but we are discussing general theory here.
To be sure the book is out on all three, and the masses may eventually be right. But I doubt it. The two owners involved saw what others did not. Suffice it to say that in my opinion the two owners involved in these bidding wars could be considered experts, despite how these individual moves may work out.
Your goal: not just to win your league(s) but to gain knowledge, improve your judgment, consider more variables and to learn about the core principles of the game. When you do this you will be a “successful” GM; namely an expert. Expertise requires a confluence of factors that are all within your grasp, so go get them.
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.