It is a compelling idea, if you could get it to work. It is kind of like equity curve trading, where you stop trading a system when the equity curve falls below its moving average. Except in this case, it is in reverse - you start trading it as heads into deeper drawdown territory.

The first issue I have with this is that you are making the assumption that the strategy will recover, and eventually reach new highs. But, what if it doesn't, or what if it hits a new historical maximum drawdown? You'll be trading it on the way down, and you might get wiped out before it recovers.

The second issue may be the bigger problem. You are now making a "wrapper" system to trade another system. And with your new wrapper system, you are reviewing all the results to make a decision when to turn it on. That means it is all in sample data, with no out of sample. So, of course the drawdown you choose is going to yield good results, because it is all in sample. You need to evaluate it for a while with out of sample results. This will be a problem, because drawdowns of that magnitude don't happen all that often, most likely. It will take a while to collect sufficient data.

I think the second issue is the big one. It is not to say it couldn't be done successfully, but you'd need a lot of out-of-sample data to verify it actually works. One possibility: make your "start" drawdown point a percentage of max historical drawdown, which you can get by optimizing over a bunch of your old strategies. Then, apply that percentage to a bunch of OTHER strategies (your out of sample set). If it yields good results on your out of sample, maybe you have something.

Good Luck. My personal experience is that equity curve trading does not really help, ever if at all. I find it better to just always trade the strategy, until you decide to turn it off for good.

So, I have developed a trading strategy which shows consistent profitability in Live Sim for over 6 months now. 'Problem' is, some weeks are inevitably not so good which is expected with any strategy/trading plan.

There is no holy grail strategy which makes guaranteed profit every week. Markets change day to day, week to week, month to month. Markets are dynamic.

To be a trader, means accepting loss and managing risk, as a prerequisite to making profit. You can not have profit without loss/risk in trading.

So, I have been thinking about how I could minimise my drawdowns and therefore increase my overall profit. There is one idea that I would like to share with you, which can theoretically be applied to any strategy!

It is quite simple really. This will work if you already have a strategy running for a few months at least for enough data to analyse. I will tell you how I am thinking of incorporating this idea:

I have a strategy, it will run on simulation. Once the strategy reaches a certain amount of drawdown, I will start the strategy in my Live Account and leave it running for a certain period of time.

Now, the idea behind this is that every strategy has a drawdown. After the drawdown, every strategy makes profit (the lowest point of a drawdown is when things start to go up again). For example, you may look at your results and see that the max drawdown was $2000. You may see that your strategy typically recovers after an average drawdown of $1500. So this is what you do: Run your strategy in live simulation. Once the strategy reaches a typical drawdown, that is when you begin the strategy in your live account.

So what you are doing is cutting out some of the drawdown in your live account, AND beginning your strategy at a typical time that it will recover, meaning higher probability trades!

Note: This is just an idea that I thought was worth sharing. I know there are some faults and I am not saying this will definitely work, but as long as it gets people thinking and adds to the community, I'm happy.

Thanks, and would appreciate feedback!

6 months of curve fitted data is irrelevant.

In what way did I imply that it is curve fitted data? This is 6 months live simulation testing data. Nothing curve-fitted here.

It`s 100% curve fitted as you`ve built it to the current environment.Test it on at least 8-10 years sample,and see what you`ll get

Wait, I don't think you understand what I am saying. I have tested it in REALTIME, for 6 months. I have backtested it for years.

Sorry for misunderstanding then.Can you tell us what was your period sample and Sharpe?

Now, the idea behind this is that every strategy has a drawdown. After the drawdown, every strategy makes profit (the lowest point of a drawdown is when things start to go up again). For example, you may look at your results and see that the max drawdown was $2000. You may see that your strategy typically recovers after an average drawdown of $1500. So this is what you do: Run your strategy in live simulation. Once the strategy reaches a typical drawdown, that is when you begin the strategy in your live account.

So what you are doing is cutting out some of the drawdown in your live account, AND beginning your strategy at a typical time that it will recover, meaning higher probability trades!