Hotsauce: A risk-managed grid-based EA portfolio

[fxozgirl]

Anyway we all know women are muuucccchhhh smarter than men, so nuff said

What a sexist comment! I am so offended!! (flounces off in a huff )

[Thweis]

We are seeing a drawdown of 15% now as our 4 CAD based pairs are in a drawdown. This is your chance to follow the system and decide if the risk exposure suits you or not.

15% DD sound like a lot however it is not much in realation to the 25-30% that we make normally make per month.

We will see how the situation resolves in the days ahead.

http://www.myfxbook.com/members/Thweis/ ... -28/734197

Moreover you should just ignore the balance curve in a grid based system and only use the equity curve to measure performance.

[garyfritz]

15% DD sound like a lot however it is not much in realation to the 25-30% that we make normally make per month.

?? Your myfxbook account "only" shows about 13-14% per month. Are you saying you did better than that before the myfxbook account started?

[Thweis]

garyfritz,

we made 22,8% in November and 17,6% in 22days in december (paused trading over the holidays). This month we are now at 8,5% on 10th January 2014. We added a couple of high growth strategies on 26th November and you an easily see the increases slope after that.
Myfxbook just uses a average and ounted the three days from november as a whole month.

To make a long story short it is very easy to predict the growth with our portfolio as our backtests are always very close to the real thing as we are taking great care in backtesting in an enviroment as close to our brokers settings as possible.

Growth during month where we have no stop loss hits will be roughly 25%. Some month might be better others might be lower (like december for obvious reasons).

v/r
Martin

[macman]

I dug through a bit of their website and I was impressed with their level of transparency and disclosure. They are using Martingales, but they do a fairly good job of explaining the risks -- much better than I've seen with any other M-based vendor. I suggested they should add a description of the true worst-case scenario -- what happens if you get a class-A black-swan event (such as the 9/11 New York attacks) that causes many of their strategy/pair combos to hit their M stoploss......

Yes Gary, agree - there is more information here than is usual with a signal service type operation.

Their approach can work well using multiple strategies until, as you rightly say the "Black Swan" comes along ..... not sure why it called a "Black Swan" when "Black DEATH" is more appropriate for us retail traders.

I almost laugh out loud when I see posts from guys who have ’simulated’ a "Black Swan" event and proudly proclaim how well their strategy held up. So what really happens when a "Black Swan" event hits?

I was trading 3 different EA's using different strategies with different risk profiles on different brokers and doing well, riding the 'news' events nicely and so on. I even used to joke my job had been reduced to a robot technician - what could go wrong?

That all changed within hours when the Tsunami hit Japan recently. I was a bit slow in reacting, as I had assumed my system would handle any event. When I tried to manually intervene it was too late. My main account with a high end brokerage who are definitely not a bucket shop was frozen with Empty4 giving all sorts of errors when trying to close trades. So in desperation I rang their office. After a long wait the phone was eventually picked up and in the background I could hear 100's of phones ringing - then a woman with pure panic in her voice shouted "We have NO liquidity" and banged the phone down. The rest is history. My emergency stops were eventually honoured by the brokerage, but read the small print and you will find that many brokerages are not even obliged to do this.

So if you really want to 'simulate' a "Black Swan" worst case event wipe out your account!

As retail traders we are always at the bottom of the pecking order and the last to get the news, so make sure your risk profile reflects this.

I hope this real world account helps a little with understanding just how vulnerable we retail traders are.

Be careful out there guys

[Thweis]

You are correct that a black DEATH event can kill any grid system while systems adhering to the golden trading rules are less vulnerable. We will move away from grids over time however this will be a slow process. Regarding the broker issues I totally agree with you as well, every trader must try hard to leave the retail forex world as soon as possible. That is possible if you have 40k and above available as this allows you to negotiate some customized trading conditions with an institutional broker.

v/r
Martin

[macman]

Hi Martin,
Just for any newbies around, let me clarify what I consider to be a 'Black DEATH' event, rather than a 'Black Swan' event.

Generally speaking a 'Black Swan' event is defined as "An event or occurrence that deviates beyond what is normally expected of a situation and that would be extremely difficult to predict." (Investopedia).

For many traders when a currency pair moves unexpectedly let's say 1000 pips, this is a 'Black Swan' event, others may call it a sharp market adjustment - as usual in forex these terms are vague.

By contrast my definition of a 'Black DEATH' event is when there is market panic and the system just breaks down and during these fairly infrequent events you can easily lose your whole account.
So how often do they occur?

During the fairly recent past 'Black DEATH' events include the Flash Crash, 911 attack, Japanese Tsunami etc. For anybody designing a long term trading system this brings up a problem - do you design to try and include 'Black DEATH' events or just accept that they can (and will) happen sooner or later?

Trading forex is anything but an exact science, but it is essential for your long term health to anticipate.

Hope this helps someone.

[Thweis]

We see some high correlation here in our four CAD related bots causing a drawdown of >30%. Our NZDCAD just hit the stop loss and GBPCAD is close to hitting the stop loss.

If you considered joining our signal service this is good news for you as it allows you to really see the drawdown/risk without being blinded by a steady upsloping equity curve. We will sit tight and are not too worried here. We might give away some or all profits that me made so far over the last 2 month however we know that the portfolio can recover quickly.

Quote from our website:

There is no free money to grab on this planet and most things are in balance. If you want to create insane profits like our method tries to accomplish, there is the probability of proportionally large drawdowns and losses. Both are part of the system, you cannot have only one and neglect the other.

We will sit tight as we believe in our method. Like always the situation is fluid and it is important to have a fresh mind and act accordindly to whats happening around us. If this worsens we might remove on of the four CAD bots from the portfolio and reduce the portion of another one slightly.

@macman: I traded through the mentioned events in stocks, however the reactions in stocks are easily manageable but forex is a different story hence I agree that a pure gird system is vulnerable to these kind of events and we already working on a way to protect us against it. Usually this means moving away from a pure grid based system which is exactly what we do.

v/r
Martin

[macman]

Martin,
Interesting post as it ties in with something I have been showing the students on my own training course recently.

I checked out your myfxbook link as I was interested in why your account was suffering high draw down at this time. I was a bit surprised to see you don't trade any JPY or CHF crosses but are CAD and commodity heavy.

By coincidence I have been discussing Risk vs Reward with my students recently and as part of the discussion I started the EA available to students back in November. To emphasise the point I was making, this is running a lot size which is 10 times too high; basically a non intervention suicide account which I had expected to be margin called over the Xmas period (that was the point of the exercise). It is a fairly intelligent grid trader, designed to be used with manual intervention.

Below is the trace with notes (it just survived Xmas), which shows the pairs I set up, the draw down, profit etc.

C3PO_demo_rr.png

If you compare this chart with your own (below), which I have cut/paste from your myfxbook page, you can see the similarity - the Xmas dip is there, but cushioned by your choice of pairs, with this recent draw down amplified by your choice. Both charts show the same time period.

HotSauce.png

So basically for anyone looking in, Yes a grid based approach can work and if you run it high risk it can create high returns. The better the 'balance' between your chosen trading pairs the lower your risk is. However without intervention any grid trader will sooner or later collapse, so IMHO you must be able to trade manually so that you understand the signs that come along indicating when you need to step in.

Best of luck with your service Martin & I hope members here find this post useful.

[Thweis]

@macman: The problem is that I am fully busy trading stocks and that we will only trade forex automatically for that reason. I like to keep things simple hence I won't do both together. However we do intervene if we see a chance to get out of a basket without too much loss.

We work on our own EA which is basically a short term trend following system based on non obvious price&volume action (the volume part is important). It will employ sound pyramiding and a rigorous stop loss. What we will do is basically trading in between the crowd by enterning trades on non obvious points and take profits right before the obvious breakouts and support/resistance levels are breached.

@Grid portfolio: Yes we are very commodity heavy and also trade exotics like EURZAR, USDZAR, AUDSGD, EURNOK and the like. Trends are our worst enemies hence we prefer the commodity crosses and based pairs.

Exotics are just less likely to have the "shakeout" events as less eyes are watching the charts.

@X-Mas period: In stocks it is usually a good idea to be mainly in cash during dezember/january even if this means that you will miss out on some profits. However in forex I am still not sure if only the christmas period should be avoided or also january? I will look into this and we might come up with some safetly measures for next year.

v/r
Martin