Sorry I have no time to waste in trying to replicate your worksheet, when I have used mine successfully for several years. What are you afraid of, for not wanting to post your backtesting trade results in one way or the other? There is no need to post any date, time, trade direction, nothing except each individual trade P&L.
for instance, most mathmeticians start with the assumption 2 is an equal distance between 1 and 3. But if that distance is actually always changing it can mess up your calclations.
most of that is copy paste...it would literally take less than 5 minutes. also it is basically an exact copy of the excel you posted other than different numbers. *i figured out why....there was no simOutput.
If I may ask, where does your December drawdown stand vs the MonteCarlo mean (that is, how many std-dev away from the mean) ? (using a MonteCarlo run of approximately 1 year worth of trades)
they are both the same, they are both based on the assumption what has happened will keep happening. I have to laugh when someone calls anything and "edge" like just because something always happens and I have identified that gives me an edge. All that gives you is a time tested assumption. If you can prove to me things are not changing I will admit you have an edge. otherwise, money management is for me much more reliable than guessing what it's going to do. If it were not so, why do so many smart people disaagree about what it's going to do? There is very little disagreement about sound money management. If the opie had it he would not be blown up.
If you have a positive expectation but large draw downs, then money management can indeed save you from blowing up during those draw downs. If you have a negative expectation however, no amount of money management will save you. It will merely slow down your inevitable march to complete ruin.
if you check my posts, I've said the same thing many times, especially when it comes to relying on tight stops. We are all on the road to inevitable ruin, if you don't know that you have not tested far enough out. The idea is to postpone ruin as long as possible. If there isn't a "quit while you're ahead" parameter in your system you'll just ride it up and ride it back down. If it was a zero sum game it would be fun, But spreads and commissions make it a negative sum game. The first thing that tips me off someone is talking out of their ass is when they emphatically state, "It has nothing to do with luck."
Wow this thread has gotten off track and then back on track like 10x so far. I spoke to the OP a few times over Skype and PM. I'm by far no expert and I still am not 100% sure that I understand his system or his results but he did email me a spreadsheet with a lot of info (no I won't share it). I still say to feng - I'm not 100% sure I understand what you sent me. I don't trade futures so maybe its something stupid I'm missing or overlooking? I honestly have looked at it over & over 1-2 hours each time, every few days - and I still don't 100% understand or get it. OP is manually trading based on rules. I believe that the reason why he experienced such a drawdown is because he was doing exactly what he says and was trading based on his rules and signals without discretion or emotion. Pretty much an automated system but not 100% coded and able to execute on its own. I only own 7 years of tick data and would probably only backtest for a max of 3ish years at most. Then, if I think in the initial design of a strategy **might** work then it gets the full (time consuming) tick database and then various shorter-runs for optimization. The only reason why it helps to backtest more than 3 years (in my opinion) is that it might include something catastrophic or a rogue event (like a flash-crash or massive gap up/down or limit up/down on a stock or index) that could be telling on the extreme ends of the cycle. Most of my focus has been always very short-term (short holding periods) so for longer term I expect that the backtesting and optimization data would need to go back farther. I think the OP's system may be a fluke. His backtest results are very accurate to the real-world trades - winners and losers. Because I don't fully understand it I can't say for sure "look at this date/time when market conditions changed and here is why it stopped working" but it did pretty much fall off a cliff. Feng - i sent you a few questions via skype. not sure if you got them. I don't trade futures so I think i'm missing something basic. I also don't have ANY futures data (historical) so I'm not really able to pull together much outside of simple data from Bloomberg - but I don't know when you switch or roll contracts, start looking at the back month vs. trading the front, etc. I just don't know enough info to tie your data points back to historical that I know to be good. I'm happy to (we aren't supposed to but this isn't that much data) share with you some of the data I might be able to get from Bloomberg - if it'll help. Without knowing more it may be as simple as you got lucky, or it could be something simple like your data is off a day or who knows. Cheers, lots of good (and some bad) info in this thread so hopefully you'll come out of this experience a better & stronger trader.
Sorry for you man, but i love to hear when automatic traders lose money, because i hate them. Thats why i be a manual trader. Manual, WTF ? How you do this, do you have a brain, amazing, could i borrow it ? No sorry my friend......... Automatic traders are not real traders in my opinion, they are just gamblers........but there is a clear difference between a gambler and a trader. the one is a hard working brain using mastermind, while the other is a lazy ass bastard gambler....... LOL sorry for you:eek: :eek: :eek: