I'm having a brain fart, someone's gatta set me straight. Does fixed fractional naturally put you closer to 0? My reasoning(hypothetical): 3 wins, 3 losses, $100,000 account. Win/lose 2% every trade. T1: Win...$102000 (+2%) T2: Lose...$99960 (-2%) T3: Win...$101960 (+2%) T4: Lose...$99920 (-2%) T5: Win...$101920 (+2%) T6: Lose...$99880 (-2%) ... If I kept this up it would almost hit 0! Someone tell me why I'm wrong!!
Your reasoning is that you win or lose 100% of your trades, which is course pure fiction. This is trading, not casino betting. If the 2% falls below your Kelly fraction (whatever that is, and which can only be determined empirically for a trading system), then your account will increase, not decrease. http://www.elitetrader.com/et/index.php?threads/kelly-for-traders.102205/
But a trader has to win or lose 100% of their trades, that's the only 2 options, like a coin flip has to be 100% heads or tails(no 3rd option). I'm confused by what you meant
You're assuming a fiction. Since when is any real trade a 100% win or loss? That's betting in a fixed scenario, not trading. Worse, since you're assuming the probabilities of a coin flip, you have no advantage and therefore no reason to bet. The Kelly fraction in your example is 0. IOW don't bet anything, not even 2%. This is the problem with using betting examples to illustrate trade sizing. Betting examples are easy to use precisely because they have fixed probabilities and fixed payoffs. In the real world of trading, there are no fixed probabilities and no fixed payoffs.
Is this true?: If I just randomly go short, looking for a 5 point drop, and a stop loss at 10 points up, and I did that 100 times, on average would I win 66% and lose 33%?
Oy. I have no idea where you get these numbers from. But this is personally why I avoid stops and targets. They give the illusion that you can pinpoint your wins and losses when in reality you can't. Trading isn't a casino game; there are no fixed probabilities and no fixed payoffs. But there are timing strategies that can give you a decisive edge. Read the first post of the link I provided. It will at least give you some idea of how to size your trades given a winning strategy.
Recalculate after a period in which your strategy is usually profitable (over the course of a week/month/year etc)?