I've got 1000 USD and a strategy that makes sense and has a good profit factor. But how do I calculate best optimal fixed fraction % for it? Clearly if I risk 1% or 2% per each trade, I am not going anywhere and equity is not utilized at its fullest potential... on the other hand if I trade let's say risking 90% of equity on each trade, I am playing with fire and about to blow up my account, right? So how do you calculate best MM FF% ? What do you use for educated guess then?
You need have an idea of the maximum possible drawdown for your strategy over many trades and market conditions. If you don't know that, then you should be trading with as minimum risk as possible until you generate relevant data. It's paying tuition to the markets. Some do historical backtesting.
one time a guy walked into the office and put on a pork belly spread. Seemed like a good idea at the time. The spread was very wide and historically it had always reverted to the mean. But this spread just kept going and got wider than at any time in history, and the trade got out of hand and he not only lost all his money but owed a lot on the trade. And he was big guy in town, owned a business and a lot of real estate and a strip mall. So he didn't want a lawsuit with his name on it. And every month he came in with some hard luck story about why he couldn't pay off his gambling debt.
Oops. Probable/Likely drawdown. You know what I mean..... And, you might want to do a Monte Carlo analysis of your data.
Agree. Commodities can move several standard deviations more easily than stocks. If you get a breakout on a mean reversion trade, it's can be quite spectacular.
I already know the only strategy which has no risk of ruin is Martingaling with an unlimited supply of money and no limit on the table. The rest of us are just gambling. Probable? Likely? How are you going to make any money betting on that? It lasts for a while when you are young and scalping, but eventually something improbable and very unlikely occurs. And that is where all the money is made. But you have to be in to win, and that is where money management can save your ass.
Friend of mine went short lumber and next 13 days it went limit up, bankrupt, lost home and family on handful contracts. $1000 to trade, you do demo at all? You can lose more than $1,000 even in forex, if the forex dealer has electronic issues and everything freezes, you can lose much much much more. How much was that Swiss Franc spike few months ago?
Half Kelly is not the best option. In fact it is quite easy to find the best % MM if you know distrubution of your outcomes. You just need to define what you want to achieve, the biggest acceptable drawdown and use Monte Carlo simulations.