Criteria for getting signals to the top - page 12
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Olga, no one disagrees that these are important parameters. They should be put into a formula. Because without the formula, a ten-week signal with a maximum deposit load of 100% is better than a nine-week signal with a maximum deposit load of 10%.
Can you point the finger at the "professional trader" with this signal?
It's not good to point the finger at yourself, so I won't)))
I bring to the attention of the public and admins a fairly simple system of signals rating, which depends on only 3 parameters extremely important for subscribers: signal lifetime, average monthly profitability and maximal drawdown on the account.
Rating is calculated using R=Kt*Kp*Kd, where Kt is coefficient depending on T - time of signal operation monitoring at mql in weeks (all history prior to registration of the signal at the website should be cut off); Kp - coefficient depending on P - monthly average percentage profit (only complete months are taken into account, the current incomplete month is not considered); Kd - coefficient depending on D - maximum drawdown fixed in statistics in percentage value. All ratios are calculated as fractions of one and the final rating value will be in the range from 0 to 1. Formulas for ratios calculation:
Kt=T*0.005-0.06, but not less than 0 and not more than 1. Negative values for signals with lifetime less than 12 weeks are replaced by zero, when the signal lifetime reaches 212 weeks (4 years) the coefficient remains equal to 1.
Kp=P*0.02, but not less than 0 and not more than 1. Negative values for the losing signals are replaced by zero, when the average monthly profit exceeds 50%, the coefficient stays equal to 1.
Kd=1-D*0.011, but not less than 0. Negative values at the maximum drawdown greater than 90% are replaced by zero.
Little explanation on the logic of this rating system. All signals registered at mql less than 12 weeks (3 months) with no profit or 90% drawdown at any point will have zero rating. Maximum rating will be for signals which have been working for 4 years or more and showing 50% and more average monthly profit with minimal drawdown on the account. The value of 1 is hardly achievable, because it is almost unrealistic to show the 50% of profit per month with the drawdown less than 1%. But with maximal drawdown values ranging between 5-10% one can earn 50% a month. I have seen such signals. It is also unlikely that someone will be able to consistently show 50% or more per month for several years, but it is possible to approach perfection - a stable 20-30% per month at low drawdowns can be found on existing signals. Providers trading too aggressively will not get an advantage in the ranking, because with the average monthly profit exceeding 50% their Kp will not grow further, but they will lose Kd to providers with less aggressive trading. And lovers of high drawdowns will be at a huge disadvantage, not to mention the fact that they are unlikely to work for several years with drawdowns of 60-70% and more.
I hope I've stated the basic principles more or less clearly: the providers whose work is stable and profitable for several years and whose risks are minimal will get all the advantages in the rating. Providers who want to make a quick buck on subscribers by making risky trades and showing excessive profits in a short time will be at the very bottom of the rating.
I bring to the attention of the public and admins a fairly simple system of signals rating depending on only 3 parameters which are extremely important for subscribers: signal lifetime, average monthly profitability and maximal drawdown of an account.
Here we go... First clear and specific proposal.
Now let's estimate the signals... (Control, Andrew).
-------------------------------
So signal 1.
18 weeks. Ct = 0.03.
141% during this time. Thus, we have 21% Cd = 0.42 per month (4 weeks).
Maximal drawdown is 19% Кd = 0.791.
Total signal rating R = 0.01.
Now signal 2.
34 weeks. Kt = 0.11.
88% during this time. Thus, we have 7.5% Кr = 0.15 per month (4 weeks).
Maximum drawdown 39% Кd = 0.571
Total Signal Rating R = 0.09
-----------------------------
I think I calculated it correctly.
All in all, not bad in my opinion. I only have one observation. In my opinion, the effect of drawdown on the coefficient is too small.
Well, here goes... The first coherent and concrete proposal.
Now let's estimate by the signals... (Control, Andrew)
So signal 1.
Controlling :)
On profit assumed a slightly different system of calculation: add up the profit values in each month and divide by the number of months.
P = (18.49 +28.08 + 29.76 + 33.84) / 4 = 27.54 (last unfinished month do not count).
Kp=27.54*0.02=0.55
Total R=0.03*0.55*0.79=0.013.
Controlling :)
On profit assumed a little different system of calculation: add up the values of profits in each month and divide by the number of months.
P = (18.49 +28.08 + 29.76 + 33.84) / 4 = 27.54 (last unfinished month do not count)
Kp=27.54*0.02=0.55
Total R=0.03*0.55*0.79=0.013
Yes, I calculated the average monthly profit by taking the total profit and taking the root of the degree N (number of months). From the looks of it - the difference is not too big. But, according to your suggestion, so can I.
Yes, I calculated the average monthly profit by taking the total profit and taking the root of the degree N (number of months). From the looks of it - the difference is not too big. But, according to your suggestion, it is also possible.
It is more important not to take the last month into account, because there can be negative results (as in the first signal) or even prohibitive.
To increase influence of drawdown values, one can suggest to squaring the drawdown percentage, and then the value of D*D would be 100 for 10%, 1600 for 40%, and 8100 for 90%. Weighting coefficient should then be set to 0.00013.
Kd=1-D*D*0.00013
With a reasonable drawdown within 10-15% the coefficient will be high, with the growth of risk D*D will grow exponentially and Kd will sharply decrease, while if it exceeds 85% (level of stop-out at leverage 1:500) the rating will become rubbish or even null.
George, there is a mistake in calculation of the signal 2 rating: its value will be 0.009, i.e. almost 1.5 times lower than signal 1. So, the effect of drawdown has a noticeable effect even without the square in the formula - even a significantly longer operation time minus 12 weeks did not help :)I support the guys above. The current rating is something else.
Before you invent a perfect rating algorithm, look at how subscribers choose signals - https://www.mql5.com/ru/signals/mt4/list?orderby=subscribers
I support the guys above. The current rating is something else.
Here we go again - all criticism. And vague arguments about how to make it better.
What about concrete suggestions? Like Andrei suggested above?
By the way, Andrey Karachev, to promote your proposal - it would be reasonable to open a branch, and once a week to write "your" rating of the best signals, based on your formula. I think if your rating turns out to be more adequate than that of the developers - your developments may be taken into account.