Hi win rate = winning trade number/ total trade number loss rate = losing trade number / total trade number
joesan, interesting calculation, but this method immediately rejects as "not ideal for realtime trading" any strategy where the average win is less than half of the average loser. So if I have a strategy where my win rate is 90/100, and my loss rate is 10/100, and my average loss is 4x my average win, I get: C=1/4-10/90=0.139 On the other hand, if we look at expectation, lets assume the average winner is $100 and the average loser is $400, with a 90% win rate. The expectation of this strategy is $50/trade, not including commissions, etc. Based on the positive expectation, I would dig further into the strategy. How efficient is it, i.e. how much of your profit are you eating up in commissions? Are you making 10 cents on 1000 shares, or a full point on 100, for example? Are you hitting the bids or asks, or letting the market come to you (and getting rebates)? Now what is the real expectation? Going a step further, what is the likelihood you're going to blow up? As I mentioned earlier, I think risking 35% per trade might be a bit high. But if you're lucky, you'll get away with it... Just think hedge fund manager... But, all things being equal, I would much prefer a strategy with a C close to one! If anybody out there wants to show me how to do that, I would appreciate it.
"But, all things being equal, I would much prefer a strategy with a C close to one! If anybody out there wants to show me how to do that, I would appreciate it." I'm not a math whiz but what would be the numbers to fill in the formula to = >1?
Hi ST Yes , you are right in calculation. But the system in your example ( similar to the system mentioned in the original post of this thread ) is exactly what this "C factor" trying to avoid in realtime trading. However it does not mean that the system is not profitable, it just means if traded in realtime, it can bring a lot of stress due to the volatility of the equity curve. And if you are manage OPM, that kind of drawdown will make the investors frown.
come on hershey don't you get a heart attack on me now start using fish oil omega 3 if you are not already not a joke
Smart, very smart. Too bad the OP won't let you in on the ground floor of this Midas maker. Price undoubtedly goes sky high after the beta.
Chood I don't know how but I have this ability to .........smell if you will people who know their game care to tell us a bit about yourself what do you trade, what type are you, swing, day or position etc. how long have you been doing this
I would question the logic/math behind Joesan's suggestion... To the OP, there's only 2 things that really matter in system testing. The first is your expected profit factor. The second is the frequency of your trades. If you have anything above 3 for your expected profit factor, you will make money over time and have an edge, but only if its based on at least 1000 real, live trades. Even then, that sample may not be enough to determine the value of your system. If you doubt this, ask John W Henry. This is why frequency matters so much to the robustness of a trading method and this is partially why Jim Simons is the richest hedge fund manager of all time.
why not let murray rugerio test this system? he actively posts here. its my contention that the initial post is some kind of a prank, its telling to see who falls for it... regards, surf
"its my contention that the initial post is some kind of a prank," fwiw, I googled around earlier and did find a lawsuit of some sort of filed against similiar claim. I really didn't continue reading or dig deeper but I'm sure there would be others. Prolly not a prank but a system that 95% of the time ends badly. But hey, room for everybody and everything in the market.