Topic: Max Equity Drawdown

I often use 'Max Equity Drawdown' as one of the statistics to determine whether or not to keep a strategy.  I understand why one would want this to be as small as possible.  From reading elsewhere on the Forum I try to keep it less than 10% -- even lower.

So -- suppose I have a dynamite strategy with excellent statistics -- e.g. very high SQN and win/loss ratio -- but its MaxDD is also high -- e.g. 25%.  If I am also using a conservative SL, then why do I care what the MaxDD is?   I mean, if all the other statistics indicate this is a great EA and if it wins 98% of the time and if the SL is conservative, then even in the worst case the SL would protect me from the MaxDD.  So, is there really any risk in using this EA?  And if all my EAs include a conservative SL, should I even pay attention to the MaxDD statistic?

I suppose I'm missing something -- I would just like to understand this a little better.  Thanks...

Re: Max Equity Drawdown

It looks like that such strategy makes profit most of the time but it tends to have consecutive losses.

If the other stats are ok, it depends where the DD is located in the backtest chart. If it is in the beginning of the test and the strategy makes profit at the last part of the chart, you may try it for trading.

It is always tricky to decide if a strategy is good for trading or not.  That's why we try to find and trade several strategies for a market. However these strategies must be "different". I mean that they must not have similar indicators and entry rules.

Re: Max Equity Drawdown

Okay -- taking into account consecutive losses makes sense -- thanks.

I've attached an image of a mediocre strategy.  You can see it has only one consecutive loss.  The Max equity drawdown is very high -- greater than 39%.  Yet the SL is only 98 USD.

With only 1 consecutive loss and the SL set at 98 USD, how can the Max equity drawdown be 439.76 USD?  Is it because the Equity chart does not take into account the SL setting?

Or, perhaps it is a bit more complicated.  Even though there is only 1 consecutive loss, there are several losses that are clustered and which, collectively, could account for the large drawdown.  But then why is the balance line horizontal in the area of the drawdown near the middle of the chart?  Shouldn't it also drop because my SL kicks-in?

Post's attachments

question-maxdd.png 30.36 kb, file has never been downloaded. 

You don't have the permssions to download the attachments of this post.

Re: Max Equity Drawdown

I think the Max Drawdown is quite subjective in that if too tight, then what could be excellent trades are missed.

This has to be considered along with the risk level to be tolerated.

Re: Max Equity Drawdown

I agree -- it can be used to weed-out undesirable strategies, but if too strict then you might miss good ones.

But I'm trying to reconcile the statistic reported by FSBPro and the corresponding Balance / Equity chart.  In the previous post I attached an image of some stats along with the corresponding Balance / Equity chart.  I am having difficulty reconciling (interpreting ??) the reported Drawdown statistic and Balance / Equity chart with the fact the strategy includes a conservative SL that is much smaller then the drawdown.  I guess what I am missing is how can there be such a large Equity drawdown when the SL is so much smaller. 

Also, in the middle of the chart where the Equity drawdown is quite pronounced, the balance remains horizontal.  I would have expected the SL to have kicked in and the Balance drop -- but it doesn't.

It is almost as if the backtesting and Balance / Equity chart ignores the SL setting.  Do you see where I am confused?

Re: Max Equity Drawdown

I have been thinking about this for a while now.....  I am not too bright with math and statistics to know how to quantify the problem....
So..... I am running some tests next week, I already have them set up for market open..... and after a couple weeks, so that there are some trades. I will post whatever results I can glean.

For me..... I have to run a bunch of strategies under different conditions and then compare results... 

Maybe someone with a math background can step in and explain some of this.

Re: Max Equity Drawdown

Okay -- thanks for looking into it. 

But it's not a big deal -- so, don't spend too much time.  There are so many details to learn and keep track of, it's okay if I don't understand everything.

And, by the way, all this is really your fault (smile).  I was reading one of your recent posts about which stats you pay the most attention to and they correlated well with what I've been monitoring -- so, it gave me confidence I'm on the right track when it comes to analyzing strategies.  And you did mention MaxDD as one of the few you pay attention to -- so, that motivated me to snoop a bit deeper.

Thanks...

Re: Max Equity Drawdown

I set up mt4 with several pairs.. filtering them 5 different ways.... at the end of the test period.... we can examine which filter did whatever  ie each pair has 5 strategies. haha I selected 5 filtering methods, there could have been many more, but 5 is a start. I will have to see where this leads.

I am not sure how long those will take to show results.

I think it is worth doing because EA Studio presents a lot of options that we did not have in the past with the multiple strategy EA's.

The sample will have many, many drawdowns, that is the way I have set things up. Huge Take Profit and small Stop Loss.

I want to see how to maximize the filtering on EA Studio ie. can I take the top 50 in this setup, or the top 10 in that setup.. the filtering on large samples should show significant variations.

What I am doing now is taking all that appear in 5 different sorts....... maybe refine that further once  there is sufficent trades.

My hunch is that EA Studio has just ended portfolios of ten strategies, that people will be trading something like 1000 strategies at a time. The filtering in future is going to be different from what traders have been doing in the past.

Scalpers, I think, are going to become a dying breed...., as soon as they see the potential of EA Studio and learn how to harness it.

Most important, EA Studio will be the way to take the risk out of Forex trading, there will be a statistical probability of winning once we learn how to filter the output and spread the risk over a huge population.

Lastly, EA Studio will allow small traders like myself to use only one instance of MT4 for trading and be able to generate thousands of strategies to test overnight...  without investing in complex computers.

Since I went to sleep, EA Studio has generated about 2,400 stratgedies for me to work with today, the odds are pretty good that some of them will be quite suitable... I like to run the thing more than 1,000,000 times to get a good selection of strategies. I guess I will have close to 4,000 between three pairs to work with in a couple hours.

This EA Studio will rock the strategy development world !

Re: Max Equity Drawdown

I really enjoy reading your posts.  I think this last one well get more views (and the attention it deserves) if it were re-posted under 'EA Studio'.

Re: Max Equity Drawdown

Popov, can you please duplicate the thread in EA Studio...