I'd be interested to hear opinions about my assertions about season-long variance.
Also, anecdotal evidence is rarely useful but it's even less useful in these situations. If anything all these "I had x, y, and z and made it to the championship" statements would only serve to prove that it is a high-variance, and therefore bad, strategy.
Minimizing variance is a great idea in theory, but I'm far more concerned with getting the best players possible on my roster, and if I have to add a little bit of potential variance to do that, then so be it.
Just because it's high-variance doesn't automatically mean it's a bad strategy. Now, taking someone you clearly consider an inferior player for no reason other than to minimize variance, now THAT is a bad strategy (unless you're in a keeper league or dynasty league where you're operating at such an overwhelming advantage that a monkey in a flight suit could win with your team). The simple fact is that, if everyone starts on a level playing field and everyone more or less knows what they're doing, you aren't going to have enough guns to worry more about minimizing variance than maximizing talent.I would humbly submit for your consideration a league composed of the 12 best fantasy footballers in the history of fantasy football. You claim that minimizing variance is a good strategy and maximizing variance is a bad strategy, but this is only the case if you have one of the better teams in the league- and in this hypothetical, 6 of these 12 gods of fantasy football are going to have a BELOW-AVERAGE roster, meaning the smarter play would be to maximize variance. Of course, every one of them is going to think they have an above-average roster until proven otherwise, so they're all going to be aiming at minimizing variance, but I would submit that, since 50% of them are going to be wrong, and since high-variance situations are currently being undervalued by that market, going low-variance is the real "bad strategy" here, whereas going balls-to-the-wall to maximize your variance is actually a very clever and strong play.
Long story short, if you are operating at a CLEAR talent advantage (either because you know you are a significantly better Fantasy Football player than your league-mates, giving you an "off-the-field" talent advantage, or because you're in a keeper/dynasty league and have accumulated a dominant roster, giving you an "on-the-field" talent advantage), then minimizing variance is the smart play and maximizing variance is the dumb play. If you are operating at a talent disadvantage, then maximizing variance is the smart play and minimizing variance is the dumb play. If you are operating at neither a talent advantage nor a talent disadvantage, then the smart play is to entirely disregard variance as a part of your draft strategy, since you have a 50% chance of being above average (and wanting to avoid variance) and a 50% chance of being below average (and wanting to seek out variance). In fact, if you operate in a league where the other 11 players are all convinced that they're the best and are seeking to avoid variance, you can often find great value in high-variance situations simply because they're being undervalued by the market as a whole. Most of the time, though, the "shark play" is to ignore variance and compile as much talent as you possibly can so that hopefully at some point you will
be in one of those positions where you have a talent advantage and can worry about minimizing your variance.
One last point- note the emphasis when I say a CLEAR talent advantage. Every one of us overrates his own talent (both "on-the-field" and "off-the-field"), so if you think you have a talent advantage, odds are that you don't, so you shouldn't be seeking to avoid variance. Unless you KNOW that you have a CLEAR AND SIGNIFICANT talent advantage, then the odds are you cannot be trusted as an objective measure of your own talent or the talent of your leaguemates.
Besides, I find it hard to understand how you can say that the fact that so people have ridden this strategy to a championship proves it's a bad strategy. Isn't the goal to make it to a championship? Isn't any strategy that gets you there, therefore, a good strategy, because it accomplishes your goal? The goal isn't to make the playoffs every year and lose in the first round, is it? It's to win a championship- and as you said, this HIGH VARIANCE strategy apparently seems to have some success in that regard. Seems like a pretty good strategy to me- I'd rather have a 50% chance of being #1 and a 50% chance of being #12 than I would have a 100% chance of being #6 or #7.
