Attached is the screen shot for those who may not understand what I'm saying.

www.myfxbook.com

Do you know how the accuracy parameter is calculated? Does it combine predetermined amount of initial profit or predetermined amount of drawdown or both?

We can theorize on this.

Firstly, what defines accuracy? Minimal drawdown and defined amount of profit. Trade gets opened and it goes in profit in reasonable time without producing heavy DD irrespective of any closing point. There you have it, another variable - time! So, for example, a trade going <-5points + spread and >5points+ spread in 15 minutes could be defined as highly accurate. Personally I don't like this approach, too much discretionary input needed to set the "right" amount for points and time.

Another idea - lets look the chart as a series of up and down swings. If entries cluster at turning points, entries are accurate, aren't they? I like this idea. Defining swings is actually pretty easily done, I think (a note to Ahmed: we can forget ZZ because it needs a period setting, therefore another "bad" input needed).

I like the test of actual trading on a live account using very small amounts of money.

All the ratios etc are quite helpful, however tha rubber meets the road when trading for real.

I did use a few tools and found that I could recognize problems more quickly when an unexpected loss appeared on a real account.

I am just starting to place 26 copies of an ea on live charts.... I will know very quickly the merits and bad points of that EA when my equity gets hit too hard.

Profit missed and Time in profit in bars is easy.

Entry and Exit accuracy looks very ambiguous to me to the degree that they sounds like marketing tricks.

The accuracy stuff is very similar to my swing high/low theory, it's not marketing gimmicks. I've seen quite remarkable probabilistic calls on forecasting future swing highs and lows, it just might provide a good edge. A good objective way to identify swing highs and lows is all that's needed. Then if trade is opened at the swing bar or one bar before or after, it counts as an accurate entry!

For me 100% accuracy means that you have caught 100% of the possible profit on the given time frame. All other is not 100%.

If the market started an uptrend from 1.2 to 1.6 and you have entered on 1.4 to 1.5 on profit with zero DD is it 100% accuracy?

I see it as long entry at 1.2 = 100% acc., long at 1.4 = 50% acc. and long at 1.6 = 0% accuracy.

But what about previous swings?

If the market was 0.8 -> 1.3 -> 1.2 ->1.6 then is long at 1.2 = 100% acc. or 50% acc.?

Then is long from 0.8 to 1.6 = 100% accuracy?

What about long 0.8 to 1.3. Short from 1.3 to 1.2 and long from 1.2 to 1.6?

It is good to start with the right definition of 100% accuracy.  Because different people has different expectations or definition.

If we are to measure how accurate a trade is then, my opinion it is how well the trade entered with right "bet" (accuracy) and hence it's call entry accuracy.

If we are to measure how accurate a trade can capture the "whole upward trend" and hence, how near it entered at the right pricing, then that can be named as Price or Profit accuracy?

I can't say for sure how MyFxBook measures entry accuracy but I do know that exit accuracy is how well it exit at the lowest price within the trade time frame. 

For eg, if the trade exit with the lowest price "recorded" during the entire trading period, it's 100% but if the price continue to go up (for a buy position) and the trade lost the extra 150 pips from closing too early, it's still considered at 100% exit because at that point of closing, the system cannot calculate future prices or data. 

From a FSB Pro point of view, everything is of "hindsight" hence, we already have all the historical data to do the kind of 100% accuracy that you mentioned Popov.

Popov, if you are going to include this Entry Accuracy and/or Profit Accuracy, can the calculation be separated for Opening conditions and Closing conditions?

Eg.  Since your definition of 100% accuracy is based on how well the opening conditions enter as near as the start of an upward trend, then it doesn't matter at this point of time how well or what are my closing conditions.  Thus the 1st objective or aim is to develop strategies with as good a opening conditions as possible (using the start of an upward trend as the point of reference or calculation).  Once, this 1st objective is done, we can move on to focus on developing a good set of closing conditions.  In this way, we have a better chance of improving and making profitable strategies.

Currently, the strategies are calculated based on profitability and often I want to work on my opening conditions and tweak them here and there but this can't be done because as long as I haven't determine the closing conditions, the system can't calculate how well the strategy is. 

Hence, if you can separate the calculation into 2 parts, Opening conditions Accuracy and Closing Conditions Accuracy.  That would be awesome.

This was something I had in mind from the first day of FSB.

To test the entry, set the exit on Bar Closing and "Random Filter" as a Closing logic condition. After that press F5 to recalculate the strategy. It will give different performance on every hit.

Other option is to use "N bars exit". You need to preset it for 5 or 10 bars and to work on your entry rules. Or just change the number of bars by scrolling with the mouse wheel.

You can do the same with the exit rules. Set Bar Opening and Random Filter for entry and you'll see how good is your exit on a random entry.

If we figure out a proper formula or method, I'm able to implement it in the program. Probably you can ask in the MyFxBook forum for clarification on the topic. I can add all info you find useful in MyFxBook in FSB Pro in next releases.

Popov,

The Random filter isn't a good idea.  It's give random exit signal and hence, the measurement of the strategy performance is based on that random exit signal as a point of reference or measurement, it like shooting blindly.

Your idea of 100% accuracy, using a trend movement as the reference point is a better idea.  It's like looking at the 4 hour time chart and we can see when the upward trend begin and see how well we can set an opening conditions based on that reference point.

That's what I'm doing right now with my own strategies.  I put all the various indicators on my MT4 different time charts and I change the parameters here and there to see which set of indicators over many up and down trends are able to have "crossings" that consistently predict or signal the start of a trend.  I guess with this method I am able to get 70% to 100% entry accuracy for my opening conditions.  Now my challenge is to work on a "dynamic" closing that I mentioned early in previous discussion thread.   http://forexsb.com/forum/topic/5305/dynamic-closing-condition/

Two types of Profit Missed calculation.

Method 1.  The profit missed earlier.  When the trade enters, did it enter at the lowest possible point?

Method 2.  The profit missed after trade closed, how much potential profit missed for premature exit, from the highest peak.

Since you said calculate Profit missed is easy.  Let's use it to measure Entry/Exit Accuracy, if we can't find a better method or formula to measure Opening conditions and Closing conditions accuracy.  And we can measure opening condition or entry accuracy independently from closing condition. or exit accuracy.

For measuring opening condition or entry accuracy, we use the profit missed in method 1.

For measuring closing condition or exit accuracy, we use profit missed in method 2.