System Research, for You?

BarroMetrics View: System Research, for You?

A received a couple of emails asking if backtesting helps our bottom line. My answer, ‘absolutely’! It also asked if I would recommend any services.  My answer: ‘yes, but I need to know what you are looking for’.

For example, if you are looking for a ‘tipping service’, then it’s hard to go past InvestiQuant or  Quantifiable Edges. For example, for today, InvestiQuant considered that:

“a prior unfilled down gap during similar conditions has not been bullish historically.”

In short, yesterday we filled a downside gap, if we gap up today, should we sell and hold to the close? Or should we buy and hold to the close?

Similar questions for a gap down.

The site provides a full set of stats: average dollar win, consecutive wins, win rate, etc.,  If interested in the full answer go to:

Trading Tip for Thursday: Though counter-intuitive, filling a prior unfilled down gap during similar conditions has not been bullish historically.”

On the other hand, if you are asking for a service that will test your ideas, Chris’ suggestion would be worth a try: Alan at http://www.helixtrader.com/.  Chris’ comment:

“I know Alan personally, he trades his own funds for a living as well as now offering his services. Previously he traded his own funds only. He is a great guy and a very talented developer with trading expertise.” 

Final question posed: “What commission do you receive for your recommendations”? My answer, not a dime. I recommend sites because I believe they provide a service.

Oh, almost forgot…..a reminder if you want the updated backtesting system’s report with two new backtesting services included, please do drop me a line – here or at ramonbarros@tradingsuccess.com

11 thoughts on “System Research, for You?”

  1. hi Ray,
    For backtesting, if you are placing:
    1. Limit Order, how do you know where you are in the queue, to ensure your order is filled?

    2. Market and Stop Orders, what spread/ slippage you assume?

    Without knowing whether our orders are filled, without knowing the exact entry prices, how could we trust our backtesting results?

  2. Hi Paul

    I don’t know the mechanics, but I do know the backtesting programs will identify stop and limit fills.

    I also know you can set slippage amounts for stop orders

  3. Ray,

    My issue with Quantifiable Edges (which has some nice ideas don’t get me wrong) is that they regularly post low incident statistics as though they are proof of anything. Statistically, 10 incidences of a given condition (or even 50) isn’t enough to gain any confidence in it. That’s not my opinion that’s first year college statistics there. I would be very wary of basing trades on that.

  4. Hi Scott

    Thanks for the comment.

    I agree that sometimes the population size quoted by QE is a trifle small. And yes 10 is very small.

    For me, at 50, you can place some reliance depending on how much margin for error (confidence interval) you allow.

  5. The same applies to simulated/demo trading.

    Many of the trade rooms etc use platforms that use unrealistic fill rules as it pumps up their performance but you would not get those fills in live trading.

    Here is something I saved along the way i.e. the content is not written by me:

    “If you can successfully make a profit from a simulator, and the simulator requires “limit order pass through” then you have something real. I cannot stress this enough, you need to execute on a simulator that requires that the price pass completely through in order for a trade to record. The simulator at http://www.tradenavigator.com is the most conservative trading simulator on the market, and will give you the most accurate market experience. Consider it.

    Whenever a trading room or educator is using a limit order, then execution MUST BE VERIFIED. By for the most common form of fraud with trading educators or trading rooms is the use of limit orders executed on demo accounts.

    And so I would simply ask you to call your broker and ask them about any system that users a limit order to enter or exit any trade. Also, you can test this yourself by simply attempting to enter and exit a trade for only a single tick profit. Watch in horror as your order just sits there, unfilled…and then the market starts moving against you. And then you are forced to liquidate at the market for a loss

    Many people would say that using a simulator is not really a realistic way to do this. However, a simulator is actually better than a real, live money account. Why? Because a simulator can be set so that no orders can be filled on a limit price. In other words, the entry and exit price must actually trade through and not just touch. A real live money account will sometimes fill a limit order, but a conservative simulator will not, which makes it a much better way to determine the authenticity and expectancy of future trading performance”.

  6. Thanks Chris.

    Agree with the ‘fill on limit orders’. In my testing, the fill only happens if the price trades through.

  7. Best practice?

    So is there somewhere in between that might more closely reflect reality? After all your limit order may have been filled.

    CQG has more a complex rule set that attempts to simulate where you would have been in the queue (paraphrasing).

    I cannot say that this is best practice as I am not an expert however I was informed about CQG’s capabilities by people that I know are the real deal so I then did my own research after that. By way of context the first person who told me of it is openly dismissive of TA of any kind and the sell side of the industry. He was also the head of various bank trading desks (let’s call him “A”). He told me he accepts demo results from CQG as being credible. I came to find out that many prop trading groups/houses require CQG demo results as opposed to results from other platforms.

    CQG’s logic is not in their documentation so I asked them for the rules which they shared with me. I won’t disclose them here as it was part of a support request, may not be the full rule set and it was a private communication etc but suffice to say:

    “We are still trying to fill order in a realistic manner, but we are on the conservative side”.

    Along the way I had another little experience in self-responsibility. Let me explain. “A” had told me that CQG demo trading required that price must trade thru the price for limit orders (see earlier posts). I found out that CQG is more sophisticated than that. Now to me it does not matter that he does not know the specifics as a) he does not need demo results for himself and b) he knows from many years of coal face experience that CQG results are credible. A huge part of trading is taking ownership and being self-responsible and it just reinforced for me yet again that it is my responsibility to verify what anyone says no matter what their experience, by doing that I have to think for myself and fill any important gaps in my knowledge.

    No doubt there are other platforms that I am not aware of that do something similar. If anyone knows of any others then please post it here for the benefit of everyone 🙂
    ?
    Hope that helps someone.

  8. Hi Chris

    I don’t worry too much about the limit fill issue. I take the view that, usually, for me to be filled I’d need to see (for a buy), my price needs to become the offer.

    The chances of my buying the low of a move are remote at best. I have been keeping records since 1983. Do you know how many times I have bought the exact low?

    Once.

Leave a Reply

Your email address will not be published. Required fields are marked *