I am so dissapointed in the bot right now I want to break something..... it is down nearly $900 right now.. this is the BIGGEST down day in all of my 4 years of data.... I don't know what went wrong.... but oviously I have a problem somewhere... I am interested to see what my method #2 would have done today..... I am not sure what I am going to do at this point.... I am tempted to pull the plug and go back on the sim account, but I have not lost all my faith in the system yet........ Horrible day
The day has ended at -$768...... not a good day at all.... I have some soul searching to do tonight to decide what I plan on doing... One thing I am considering very heavily is switching to ES as many on this board have suggested for the lower risk until I feel more confident in the system... This is what makes someone want to completly redesign their system...
frosty, 2 things: 1) backtesting is somewhat important, but as you are seeing, that only does so much. Personally, I don't put much into backtesting and now you are seeing why. 2) If you switch to the ES, make sure you realize the volatility is much, much lower than the ER2. Yes, your losses will be smaller, but so will the wins. I would also take a look at the NQ or EC, maybe even the QM. If you have access to foreign exchanges, you may want to look at the DAX and STOXX as well. Good luck!
Would one bad game make Peyton Manning or Tiger Woods quit? One losing day and you are willing to walk away from your modelling and robots and scrap it all??? Did you expect to make money every single day? I think we all can exepct our "system" to have bad days. If it keeps happening then fix it. But one bad day?
brownsfan, I wouldn't be so dissapointed if my backtesting was great and then forward testing sucked.. cus that is the process sometimes.. but my several months of running live in the sim account had such great results... its like I picked the absolutely wrong time to actually use my real money.... optioncoach, I know what you mean about quitting.. That is one of the main reasons I started this thread was to stop myself from throwing in the towel too easily.. Tomorrow when I wake up my system will be up and running.. its just what variant and what contract it will be running on, but SOMETHING will be running
My take is that it is a comment about position sizing. I'd run with a smaller size for awhile and see if the future does indeed look like the past as far as Frosty's system is concerned. I'd also add that the last few months have been somewhat atypical. We've seen a huge rally. That's got to skew results and expectations.
A system trading approach has to have a long-term view. YOu might have a losing month maybe but what if you end up +30% for the year? Only looking at that month would mean the system was horrible, but for the year, pretty decent. The key is to track losing days, make sure they are not consistently $800 down while winning days are only $300 up unless you have way many more winning days lol.. Look at the big picture not one day. Today had a lot of chop and long swings..
Do you find that your losing trades are clustered together? This may signify that specific market conditions are getting the better of it. You might want to play around with the idea of having your system become more defensive (switch to method 2) when it gets a bad trade. and switches back to offensive mode after a few good trades. Id be interested in the difference this dynamic approach would have over the 4 years of back testing.
Walter, I will try to take a look and see if I can find a correlation where loosing trades come in bunches... also I cant remember if I posted this yet today, but had I ran method #2 instead it would have lost -$500..... better than what method #1 did...