The opinions expressed in these forums do not represent those of C2, and any discussion of profit/loss
is not indicative of future performance or success.
There is a substantial risk of loss in trading. You should therefore carefully consider
whether such trading is suitable for you in light of your financial condition. You should read,
understand, and consider the Risk Disclosure Statement that is provided by your broker
before you consider trading. Most people who trade lose money.
Well, the system just had two stop losses in a row, which added to the drawdown. I’m going to keep it Public, since I hate people that put them private when it starts to go bad. Because of the loss, I have to reduce the size that I trade, that will be the only difference.
I have been trading for a long time. Weeks like this week are why you need at least $15K per contract on the ES, otherwise two stop losses in a row send you out of the game. Sorry about this, but besides reducing the size traded, I am going to continue on until the system exceeds the statistical references. The next two trades better be profitable, or this system will get retired. Ugh
When I had my CTA a long time ago, we never traded from June 15 until August 1. I guess this is a good example of not trading during the summer. The summer break always provides superior performance. Don’t ask why I didn’t do that this year, just a mistake and did not figure it would be this bad. Live and learn.
This is an example of having overall trade rules and even the most experienced traders forget to trade the plan. This was a $6K lesson. The next trades will not be made until August 1. Just too lousy to trade during ths summer time.
The only thing I can say is that was a very stupid mistake. Also an expensive mistake. Even after 40 years, I still sometimes make really stupid mistakes thinking I know better than the market. The market is a very fickle mistress.
Well, didn’t like the portfolio that this system was trading so spent the last 4 weeks re-doing the backtesting, walk forward analysis, Monte Carlo analysis and discovered that the system does work well during the summer break, so, we are back trading. As you can see, we are trying to climb out of a large hole for June, but hopefully we will continue to perform.
We had a large position short the other day for a large profit (still climbing out of draw down so don’t get too excited) but today’s trade is smaller in size and is going back wards. Well, that’s trading for you. Will continue to present this program and hopefully continue to move out from the drawdown. Unfortunately, today did not help the June performance
Still coming out of draw down but today didn’t help
@JimmyT2, the system is now private does that mean that the backtested results are not the same as the live results? Are you closing down the strategy?
Hi guys, as I stated above, I usually don’t trade from June 15 to August 1, in that the summer recess is usually not a good time to trade with automated systems. In developing this system (the first two were operator errors and only had 1 to 2 trades each - getting used to C2) the system was running nicely up until late May and June. I had developed this auto system without excluding the summer session, so I thought trading thru this break would be OK. Well, the actual results were way off of any of the statistical analysis used to develop and measure the portfolio.
I came to the conclusion that the system was failing when it performed beyond the Monte Carlo drawdown limits. Consequently, the killing of the strategy. I have always maintained that for my own trading, any time a system performs outside the optimization, walk forward, and Monte Carlo simulations, it is time to had for the side lines.
I am in the process of re-running all of the data from 2000 with this summer break included as a no trade area. Everyone needs a vacation anyway to spend some time with one’s family and friends. So, the system had a great run till May 2018 and a big time loser in June 2018. Need to avoid these summer slowdowns.
I will be back with a new strategy developed with this summer shutdown included. I am in the process of doing the research now.
Howard, I never use just backtesting. As a matter of fact, I use backtesting, then verification (unseen data) and then finally validation (unseen data). If I have a data file I usually only use the data from 2000 to 2015 to do backtesting and testing. Usually, 60% backtesting and 40% for verification. I then run a number of strategies developed in a Monte Carlo analysis with 5000 random changes. After selecting the top two to three strategies (meeting a number of criteria), I then test them on unseen data from 2016 thru 2017.
If they are still good and have not deteriorated too badly, I then check for 2018 to see how they would do. If it still works well, then I used to put them in the “Penalty box” for 90 days and watch them in real time. If this still remains solid, I then start trading.
Since I have been retired, I have dispensed with the “penalty box” but it looks like I need to add that back into my methodology.
Lol…penalty box is referred to as forward testing. The strategies that are the most robust will be the ones that have the least amount of drawdown.
Also, the strategy must have an advantage to be more robust such as high percentage winners (which is for scalping systems) or profit targets greater than stops (just examples).
Not if you don’t change any parameters. There is no optimization. You are just trying to see how it will perform eventually in new data. Of course, there is no assurance, but we do try to select portfolios of strategies that have a good chance of seeing signals rather than noise. Hopefully, we are not fitting strategies to noise but to signals. We will see
Hi @JimmyT2, It still will be the same almost since you will be selecting the system that has best performed over the entire data period but it does not assure that the top performing system will continue to perform well over the next data period (which is the future).
What you should concentrate on is producing a system that can still profit with under a 50/50 win/loss ratio since most trading systems cannot produce much better win/loss ratios over long period of time if it does NOT involve averaging down and does not scalp only.
But anyways these are my suggestions. You can ignore them if you feel your idea is better.
