0

Let's say we have a Binary Options 5-minute trading strategy that relies on multiple indicators and exploits price reversals in currency pairs. Now let's say there is a combination of inputs for the strategy's indicators that work really well together and produce a 65% average win rate when backtested against three years of minute-by-minute tick data.

In theory, can we expect this strategy continue to produce a 65% average win rate?

For one of the indicators, we use a Polynomial Regression Channel of length 250. This tells us when the price spikes outside the recent average price range and thus offers a clue as to when the price will likely reverse.

enter image description here

In testing such a strategy, I have noticed that some days it performs extremely well (sometimes yielding a 75% win rate with 12/14 trades winning) whereas others it bombs (20% win rate with only 2/10 trades winning).

6
  • 4
    Only over the past three years? Nowhere near long enough. Many people felt that pain during the 2008 financial crisis. Dec 22 '15 at 1:04
  • 3
    @ChrisInEdmonton - you do realise that the OP is talking about currency trading (not stocks) on minute by minute data, and the OP would be going both long and short.
    – Victor
    Dec 22 '15 at 1:23
  • 2
    Victor, take a look at even just the US $ vs the Canadian $. Three years is nowhere near enough to capture enough data to understand those correlations. Dec 22 '15 at 13:14
  • 2
    @ChrisInEdmonton - that depends, we don't know what currency pairs the OP is trading and he might be trading more than one pair, which is why I mentioned in my answer to check the amount of back-testing and whether it is enough. But without seeing the trading rules and the back-testing results we can't assume that the period is too short or too long.
    – Victor
    Dec 23 '15 at 12:33
  • 2
    If an indicator is working, advantage is taken of it and it soon stops working. Then a new indicator is created and the cycle repeats. Robot trading is emotion free and must adjust to the markets over time. So if your system is working, use it until it stops. Dec 23 '15 at 15:19
3

The proof is what it does with 750 randomly selected trading days, over and over again.

In other words how does it do over a three year period of time, starting with any day randomly selected with a random time step. Based on several thousand "seasons" of data, what was the best it performed and what was the worst. Now you will get a better understanding of the capabilities of the system.

Otherwise you may have discovered a method of explaining the past, not a system of predicting the future.

1
  • And even then, all you've done is show that you can predict based on what was happening at the time. If something else changes, you may be completely wrong again. (Obvious example: that wonderful investment goes bankrupt or turns out to be a fraud.)
    – keshlam
    Dec 25 '15 at 7:27
2

No.

If you have found a mechanism for producing a consistent income, based on past data, then I expect others will too... in which case I would expect the actions of those other traders to lessen your future win rate.

12
  • 2
    If more traders are taking the same signals in the same direction, then that would increase the chances of price going in that direction, thus increasing the win rate.
    – user9822
    Dec 21 '15 at 23:32
  • 2
    @CQM, you clearly have no understanding of technical analysis. Do you realise this question is about FX which is the most liquid of markets. Who are these so called well funded market participants and insiders undermining the retail traders?f
    – user9822
    Dec 22 '15 at 5:46
  • 2
    @CQM - it sound like you are insinuating that these" well funded market participants" - investment banks, central banks and hedge funds, et al are controlling the markets. I am sorry to say but no one controls the markets, the markets do what the markets will do, and if anyone thinks that they or anyone else can control a market, then "you can tell them they're dreaming." You obviously don't understand how to properly trade using technical analysis which is the reason for your opinion and bias against it. TA should be used as part of a trading plan which incorporates proper risk management.
    – Victor
    Dec 22 '15 at 21:41
  • 3
    Wow. There are people here who believe the TA baloney. I suppose we have to make money off someone! :-)
    – Peter K.
    Dec 22 '15 at 21:44
  • 2
    @PeterK. - so what money have you made in the last year or last 3 years or longer? My returns using TA have been averaging 27% over the last 3 years, 22% over the last 5 years, and currently over 20% this year for my medium-to-long term strategy and over 65% for my shorter term strategy (which is leveraged). So I use TA as part of a trading plan and have used it successfully over the last 5 years. I know others using TA for over 10 years with average returns over 30%p.a. And in a deep downturn like in 2008/09 I would be out of the market (except with shorting on my short term strategy).
    – Victor
    Dec 22 '15 at 22:21
1

No.

Forex markets are probably one of the most efficient markets out there. The assumption essentially is that the markets are inefficient, which is a very dangerous assumption.

Furthermore, the three years of testing is not enough. This problem is similar to 20 tests with a confidence level of 95%. Approximately one of the tests will show a statistical significance. See https://xkcd.com/882/.

If you try enough many strategies on the past 3 years, you will with certainty invent a strategy that has beat the market. However, this will not mean that the strategy will beat the market in the future.

If you have a priori assumption that the strategy might be good, and 3 years of testing shows it is good, then I would at least consider running the same tests for 10 years, 20 years or 30 years of data. But remember that in any case, you're betting against the efficient market hypothesis, which is a very dangerous bet.

Furthermore, as Peter K noticed, if you invent a strategy that outperforms the market, chances are others will invent it too and thus eat away your excess returns.

No matter what you do, do not take positions that might get you into a debt problem. So, selling short is an obvious no-no. You must have a bound for the maximum losses that the positions you have taken might get you into. A stop-loss mechanism may not be good enough, as it may not react quickly enough.

1
  • 2
    No market is efficient, the FX market is one of the most liquid market, but it is not efficient. No market is efficient because different market participants have different biases and make their decisions based on different information, and they will also interpret the same information in different ways.
    – user9822
    Dec 27 '15 at 2:32
0

I do not believe in this fact. How can one underestimate the current situations and only relate with the past. Decisions should be made on the basis of technical analysis and forecasting methodology by taking in account the recent past as well as the present. Three years is a very short period to determine or plan your future course of action. You should probably look for the long term trends and then make your final decision or else you will be really surprised how you met with such a drastic downfall

1
  • By "long term trends" do you mean actual price trends? Or do you mean patterns? Jan 20 '16 at 22:23
-2

As long as you have a set of rules which you follow to the letter every time there is a signal to take a trade, and you take that trade, then your win rate should remain relatively constant. Two things you should however be wary of is that:

  1. Your emotions can interfere with the results. If you have a bit of a bad run where you lose, say 5 in a row, and then you start questioning your system. So next time when a signal comes you decide not to take it, or you take a position in the opposite direction, then you are going against your written rules and your results will suffer. So from your 3 years of back-testing you should work out you maximum drawdown and your maximum number of loosing trades in a row. This will help give you confidence that even though you are going through a bad run the system will produce you profitable results in the long run.

  2. Is your back-testing long enough that it has captured as many market conditions as possible. A robust trading system should be able to produce profitable results over various market conditions. If the back-testing does not do this you might find that when market conditions change then your actual results might start diverging from your back-tested results. The longer you can back-test for the better. You can even try a period of forward testing the current market conditions with a virtual account.

So yes, if you have a robust system that has been tested over various market conditions and you follow your written trading rules to the letter and keep your emotions out of your trading as much as possible, you should be able to achieve similar results to your back-tested results.

6
  • 1
    Why do we think that the same rules would produce the same win rate if the underlying behaviour of the market were completely different? Jan 18 '16 at 18:43
  • @DJClayworth - do you understand what a set of trading rules consists of? And what I have written is that the backtesting should be done over various market conditions. Certain set-ups may work better in rising markets, while other set-up may work better in sideways markets. Other set-up may work well in both conditions. This is what backtesting can tell us.
    – Victor
    Jan 19 '16 at 9:05
  • The biggest issues lays with point 2. No matter how much you back test. Due to the nature of man, the future will always be (even alittle) different Jan 20 '16 at 12:48
  • @reaper_unique - that is where you are somewhat wrong, history always repeats to some extent. The markets boom and then they bust and then they boom again, and so the cycle continues. Why because people are greedy and they are fearful. Backtesting does work, I have trade various strategies that have backtested well, and the results of the actual trading was within 15% of the backtested results. Average backtesting gains = 28% per year. Actual results; Yr1 = 26%, Yr2 = 24%.
    – Victor
    Jan 21 '16 at 8:52
  • Then how come not one EA works after multiple years even after back testing it for as long as possible? If it were really that simple should a lot of systems just work? I believe there are elements that you cannot put into a system which can cause major set backs. Think about central banks that announce an intervention because their currency has taken too much of a beating (or not even of a beating)? Jan 28 '16 at 13:49

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.