Random Candle System

I have been backtesting the following and there seems to be about 57-60% wins in EUR/USD M5 on MT4, if the spread is 1 pip:

Enter at random time.
Put a limit buy order at the high of the previous candle and a limit sell order at the low of the previous candle. SL=the high or low of the previous candle. TP=the same amount of pips as the SL + spread.

Problem:
if there are news Oanda(the only MM which I know offers 1pip spread at EUR/USD) widens the spread to 10 pips and as the news move the market, many of the TP's are on news candles, which can turn a winning trade in a losing one.

What do you think about this strategy?

I think it's about the same as the coin toss.
Over time, you'll average losing the amount of the spread on every trade. 200 trades = -200 pips.

I may be wrong. It may be proven that a break of the previous bars high does give a statistical edge for a long (and vice versa), but I doubt it.

Spreads going to hurt you alot on this system for the 5 minute chart. If you moved out to the 4hr or higher, it could factor in that spread a little better. 1:1 ratio with a 51% win rate is all you need to make money.

It seems that it's something like 57-60% on M5 charts (after spread), if we exclude the spread widening during the news (like trading only some mondays or something the like).

It should be something like 65% on H4, but I don't think it will be more profitable, as there will be much less trades.
(It's the same to risk 2% on M5 or on H4; you only need more leverage on M5).

Firs of all I'm new as a writer (not as a reader, more than 2 years) here on FF.

I talking from two and a half years of researchs of markets behaviour. I have a computer degree, and have a lot of know-how with statistics and probability. I say this in order to emphasis that I will try to do my best at these forums.

I did (with my fella) a lot of personal research in all kind of market, forex as well.

After backtesting a lot of systems, taking into considerations real spreads, slips, taxes,... We conclude that the markets, even forex are "almost random".

Find a system with a positive expectancy is not that complicated, the problem is to find one who can beat the inbalance with toss a coin, I mean we have taxes, spreads, ...

In all our reseach we found that according with our believes about statistics:

1) the markets are "almost" random
2) the most important part of a trade is the exit (not the entry)

Due to that we created a system that:

1) Has an almost random entry
2) We focus on the exit (we are not discoring nothing new if we say: lets run the profit and cut losses)

All this is talking only about the strategy, our goal is a positive expectancy who can beat the costs with a good performance to the money management (here is the most important part in our opinion)

Of course we do that in order to focus the issues that we can control (mm, our plan, diversification, risk management, ...) and lets the market do its random walk.

Of course it's only an opinion, and we are trading only in real for 10 months (paper and demo trading around 2 years) with only less than 3000 usd.

Our results are not huge in USD, specially due to we don´t use more than 10:1 leverage in any case.

But they are according to our plan (low risk, automatic trading, low cost ...)

Our win ratio is less than 52%, our highest dd about 30% of the balance and we profit more than 200% in 10 monts, most of the real profit due to the mm.

That's my point, I hope it'll help you.

i do think that time factor will help to increase the accuracy of this system

Whould be interesting to hear more about your method, how you enter and exit..

Of course if someone can hear it.

1) Time-frame. We don't have, but our trades last usually days, weeks and even months (we let the profits run). Our reseach show us than in this "almost random" market the best (but not the only) way to beat the balance between the expectancy of profit and the costs (spreads, fees, taxes, in our cases a server,...) is to reduce scaling the system.

I mean, the most important cost of any trade is the spread + fee, then we try to reduce the impact of them on the result of the trade. I set an example, if we have a trade that make 10 pips and the cost of the spread is 1.5 then the 15% of the profit is rule out due to the cost. If we make another trade with 100 pips with the same spread now the ratio (spread+fee / profit) decreases.

Then we are searching not small trades trying to reduce to the minimum the impact of the costs, and scaling the value of the pip according to our risk/balance.


2) Entry. In our believes/research we aproach that market is almost random. And almost mean almost. We use something simillar to S/R but not exactly, we use this as a way to enter. At some point we made a decission due to we did not want to argue about this is a clear support, or resistance, and because our goal is to make it with an autopilot (without human intervenction). When we programmed we decided to do somekind of easy math definition of S/R. But we even today don't take it that seriosly, because we do think the market is "almost random" and this almost changes quickly than we learn.

Obviosly we compare our entries with random entries (there is a mathematical method to do that with a cientifical aproach) and we conclude that it is not the by far the most important issue.

3) Exit. Take into considerations the first point and add to this a classical "lets the profit run and cut losses". We are not searching for smalls movements, retracement, ...

We input a trailing stop (variable) without TP. That's it.


We do only trade on the eur/usd pair due to the less spread. We add trade to the winner positions, take care of the risk, money management,...

But the system is that easy, and for us is working for two reasons:

1) we mathematically trust it
2) it's making money since paper trading to demo.account and now for more than 10 monts in real (still little) money.

In fact all our goal with this system was to reduce the delta of other kind of investments (portfolio) with low ratio risk/reward.

I hope this will help you !!!

Could you be more specific about your system?

I agree with you 100% but didnt achieve profitable results.. I just need to kick in...