A Brief History of my Strategy Development

F

fairvalue

I'm a newbie, paper trading since 2016 and approching the trading world like so much others did it before: holy grail searching via programming code, flipping from one strategy every now an then, overoptimizing backtesting results and curve fitting until no loss is left and of course the Youtube (or other) Vendors.

Some hints for other newbies who read this and what I would have found helpfull in the beginning, knowing or reading it all in one thread:

  • There is no holy grail system - find out for yourself or listen to expierenced traders; there are systems that tend to work good in a certain market environment. I found none, so far, but hey, that why i am here ;)
  • pick a strategy and work on it - don't change it, but besides do not overoptimize it or curve fit it. Candlesticks is a easy one, because it is easy to spot and to understand. There are so much other strategies out there. Some even trade without charts, can't remember the thread, but this is just intended to be a eye-opener.
  • The Vendors: They are there and always will be. I understand the following: Someone who lives from selling information material (not knowledge) benefits from selling you more information material . So where is the incentive of a Vendor making traders profitable, if at this point he can't sell anymore. Just think about it. Anybody can make up his own mind. There are Vendors that may be different. I will not mention anybody. Enough information can be found here, if one is interested.
The newbie-kickstart package. This is what I would have found usefull. But I am in a banking related business since 20 years. So your knowledge basis may be different. But be reliefed, many of the concepts where new to me, too. So if it helps, great:
  • Go to babypips, do the academy and have a solid basis for your trading endeavor
  • The forum is full of potential helpers, already answering so much (same) questions. Look for them. Try to understand what they were trying to tell us, or at least, what I did, try to understand what efforts it takes to become profitable. At the end, I have to develop and trust my own system. The reason why I am writing this down aswell: raise commitment and boost confidence.
  • Some links from me to the above mentioned:
https://www.elitetrader.com/et/threads/profitable-pattern.339206/page-4
 
Last edited:
I digged in this forum for quite some time, before I decided or felt comfortable entering. I view this as "the pit", requiring one to have a base knowledge of trading. I have deep respect for anyone who is sharing insights, pointing in the right direction and taking the time to write a related post.
I understand that no one has the obligation to proof anything to me.
 
Basis Concepts
  • This is no backtesting. I am already trying to do it manually (right). It is time consuming and hindsight can be a problem. Anyway I will post a link to a backtest thread once it is finished
  • This is forward testing. This can be a problem. I know that. Especially when tweaking and changing a strategy. Therefore I want to use a clear set of rules, related to market context.
  • Candlestick pattern used is a Hammer Pattern
  • Markt Context: Hammer Pattern in a trend after a countertrend-movement
  • Principles of Price Action
  • Timeframe: weekly and daily
 
Last edited:
Hi, fairvalue-I would like to share with you a little of my opinion. IMO, backtesting has value in finding what does not work but has less value in finding what does. In my position in Institutional Sales, I'm often sent marketing material from nascent CTAs and Hedge Fund managers. I have noticed that when they add backtesting to their marketing material, then add live trading, the results are almost 100% different. I have seen backtesting that averages 30% yearly over 5 years but real trading that is +/-2% of break-even before fees. Without trading in live markets on a small scale, you never know.
 
Robert Morse said:
Hi, fairvalue-I would like to share with you a little of my opinion. IMO, backtesting has value in finding what does not work but has less value in finding what does. In my position in Institutional Sales, I'm often sent marketing material from nascent CTAs and Hedge Fund managers. I have noticed that when they add backtesting to their marketing material, then add live trading, the results are almost 100% different. I have seen backtesting that averages 30% yearly over 5 years but real trading that is +/-2% of break-even before fees. Without trading in live markets on a small scale, you never know.
More...


On Backtesting
Appreciating your thoughts on backesting. That is the same, what I heard about it from others aswell. Find link here: https://www.elitetrader.com/et/threads/profitable-pattern.339206/page-4
Key fact in that backtesting was a in-sample and out-of-sample backtesting, including also a forward testing of the strategy. If anybody is interested, please find details in the link below:
https://www.investopedia.com/articles/trading/10/backtesting-walkforward-important-correlation.asp

A solid backtesting via that method could save some time. I made a decision based on a recommendation and my time constraints to do forward testing as the method. Finding the weaknesses of a strategy via backtesting is valuable. Finding a way to use it in my process.
 
And thats the one I am looking for:
  • daily uptrend, forming Higher Highs and Higher Lows
  • even better: the Signal forms above a former broken restistance line, now a support level
  • box ticked: before the Signal ( Hammer) a correction was evident
  • Take Profit: the beginning of the correction marks the price target
Just to be clear: the Signal is the Hammer (pattern) touching the (red) Stop Loss Line.
 

Attachments

  • pic1.jpg
    pic1.jpg
    242.5 KB · Views: 59
Robert Morse said:
Hi, fairvalue-I would like to share with you a little of my opinion. IMO, backtesting has value in finding what does not work but has less value in finding what does. In my position in Institutional Sales, I'm often sent marketing material from nascent CTAs and Hedge Fund managers. I have noticed that when they add backtesting to their marketing material, then add live trading, the results are almost 100% different. I have seen backtesting that averages 30% yearly over 5 years but real trading that is +/-2% of break-even before fees. Without trading in live markets on a small scale, you never know.
More...

Interesting. Are there any categories of systems, instruments, markets, advisers, etc., that you've seen that are more likely than others to fail after presenting successful backtests?
 
The Rationale behind it:

What is a Hammer? It has a long wick, a small body and can close green or red.

  • A Hammer is the most obvious signal for a change in the current market direction. Because in a single time unit, price moves in one direction and reverses in the opposite direction in an instant.

  • Based on the simple principle of “Trading with the trend”, I only want to take trades that are aligned with the trend. In that case: an uptrend. So, if
    • a) in an overall uptrend
    • b) after a retracement

    a Hammer signal occurs, it is qualified as legitimate buy signal.

  • What can support the signal, but is not yet verified by quantitative analyses:
    • the pattern (signal) occurs in relation to a major resistance zone / support zone. An ideal zone would be a former resistance, that was broken and now, by a retest (touched or nearly touched), is a valid support.

    • The signal closes green

    • Price action before (the left) the signal shows, the following
      • the signal has a Higher High than the re-test of the support zone

      • There occurred a visible retracement in a uptrend; defined as followed:
        • The last episode of the uptrend [definition] is a series of consecutive green candles.

        • The [minimum] retracement manages to form a close under the open of the last green candle as part of the uptrend.
      • The retracement shows signs of weakness
        • long wicks, Dojis, scattered green candle

        • a slower downward momentum. I guess this is a key fact. In the sample picture can be spotted quite easily: The 8th candle before the signal (hammer) is the first candle of the retracement. It retraces ultra fast within one day, almost touching the former resistance line (now support). Then followed by a test to resume the uptrend (2 green candles); that failed. The following retracement reaches the support line zone within 3-4 candles. This is a sign for a slower downward momentum.

        • the upward momentum: thoughts will follow
  • A candle can close red or green. But it seems, as mentioned before, a green close could offer a higher probability or quality.
 
Last edited:
Take the time to try and prove the system doesn't work. If you can't then possibly it does.

Right now you are looking for examples of hammer signals that hit their targets. Look for hammer signals that didn't.
 
You must log in or register to reply here.
Back
Top