Registration for the Real Accounts (Cents) Championship July 2017 . - page 99
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
There is also a suggestion to calculate profit percentage using the following formula
Percent Profit = (EquityDaily - EquityBeginingTour - (Deposit - Withdrawal)) / (EquityBeginingTour + Deposit) x 100 (%),
where
EquityBeginingTour- initial deposit (Equity at the moment of account registration);
EquityDaily - funds (Equity) at the moment ;
Deposit - funds deposited into the account during the tour;
Withdrawal - funds withdrawn from the account during the tour (absolute value);
That is, I can assume that if the FS is "0", it should be replaced by a minimum number, for example 0.000001
Is this correct, or do I need to recalculate something?
Exactly the opposite is true :)
In brief, you need, as you put it, to substitute it for the maximum number available, but not just substitute it, but with some subtleties.
I will not torture the humanitarians and will outline the proposed solution for the part of the formula related to the Recovery Factor.
Disclaimer 1: I give the solution for the case when we do not want to change the components and do not discuss the validity of the coefficients (1; 1,5; 0,5) in the formulas, because at the moment I do not know the reason why such coefficients are introduced. It is possible that the coefficients are correct.
Clause 2: The decision is based on the principle that the degree of difference in an indicator between participants should reflect the degree of practical difference in trading results, rather than a formal difference in numbers.
So,
1. The Recovery Factor shows to what extent the trading system is able to recover from the maximum historical drawdowns. Though by drawdown in the formula we do not mean the drawdown we have in our formula, but the maximal drawdown of the balance, i.e. the amount of losing trades. Therefore, recovery factor cannot be calculated for participants that do not have a single losing trade (it is highly probable to indicate "oversubscription"). (Those who argue that division by 0 gives infinity, I will refer them to the arithmetic course, where they can learn that the operation of division by 0 in operations with real numbers is meaningless, because it generates uncertainty. The popular saying is "you can't divide by 0").
2. At the same time, 1 small losing trade with all other profitable trades can give us a very large recovery factor - 500, 1000 or 5000. But will it greatly distinguish the systems and speak about their quality relative to each other? Not at all. How can we evaluate trading systems/strategies using the Rec.Fact, even if not to the 10th digit accuracy, but without gross miscalculations as applied to short-term competitions? For this, if you refer to many resources, where the trading systems indices are discussed, you can find out that there is a certain range of FS values, which characterizes the trading systems of this or that reliability class. Quite reliable trading systems are those that have FS = 15, professional trading systems have FS = 20 or more. I am quoting these figures to illustrate why I am proposing to introduce a certain limitation on the maximum FS value taken into consideration when calculating the results in our competitions. I would like to propose that the number 20 should be the maximum FS limit that will affect the results of our results calculations. What does this mean? This means that if the FS of any participant is >=20, we make it 20 for our calculations. For participants who do not have losing trades, we also set FS = 20. It will turn out that all participants that have a high PV (including those with 0 at the MQL signals site) receive the maximum of this part of the formula, i.e. 0.5 points. The remaining participants will receive a fair share of 0.5 points, which corresponds to how far off the bar they are and in relation to the other participants. If there is a numerically justified objection to the maximum of 20, another value can be discussed, but the introduction of an upper limit is necessary andin principle.
Thus, before calculating on the basis of the existing formula, it is necessary to normalise this value and not just use the FS from the signals service.
Score(Rec.Factor) =(Rec.FactorNormalized( K ) - min(Rec.FactorNormalized(1:N)) / (max(Rec.FactorNormalized(1:N)) - min(Rec.FactorNormalized(1:N)) * 0.5
K - the participant number in the table
N - total number of participants, on which the calculation is carried out
Rec.FactorNormalized ( K ) = PV[participant]norm=
if at the signals' site PB[participant]=0 or PB[participant]>=20, then PB[participant]norm=20
if on a site of signals FS[participant]<20, then FS[participant]norm=FB[participant]
min(Rec.FactorNormalized(1:N)) - is the minimum value of the normalized recovery factors among all participants
max(Rec.FactorNormalized(1:N)) - is the maximum value of the normalized values of recovery factors among all participants
Such a small and uncomplicated transformation adds meaning and removes any uncertainties. i.e. we don't change the formulae, we just prepare the value of FS for its substitution. Two birds with one stone.
That is, I can assume that if the FS is "0", it should be replaced by a minimum number, for example 0.000001
Is this correct, or do I need to recalculate something?
The recovery factor is the ratio of the current profit percentage
to the percentage value of the maximum relative drawdown.
I suggest you calculate it yourself, the data for the calculations is already in the table, then the value will be more accurate than in the signals tab, here is the formula for calculation
Restitution Factor = Percent Profit / MaxRelativeDrawdown,
where
Percent Profit - profit percentage;
MaxRelativeDrawdown - percentage of maximum relative drawdown.
I was attracted by the inappropriate use of the word Restitution. I wondered who had managed to use a term from the sphere of criminal or international law to estimate a trading system. It turned out that the Russian translators from Alpari seem to have made a mess of it. :)
On the essence of the proposal:
Sergey, you can't just change something without linking it to everything else.
Let me explain:
We have now 3 indicators in the total Score:
Score = Score(Balance)+Score(Drawdown)+Score(Rec.Factor)
Score(Balance) shows the relative increase in a participant's balance relative to the relative increase of other participants
Score(Drawdown) showsthe Percentage of Maximum Relative Draw down of a contestant relative to the percentage of maximum relative drawdown of other contestants
Score(Rec.Factor) shows the ratio of a participant's relative gain in balance to the total (max) drawdown when closing the losing trades of a participant relative to that of other participants.
It is not logical to include the same thing in the final score twice.
The recovery factor, by definition, can never be zero, except when profit is zero. Profit can only be zero in one case - when you open a position and the market goes in your direction by the value of the spread. That's it, there are no other cases and this one is very rare and does not deserve to be considered. The problem with FS is far-fetched and it is time to stop exposing your illiteracy and lack of understanding of elementary things.
There is a joke about cowboys:
Two cowboys are riding across the prairie. One says to the other, "Joe, I'll bet you a hundred dollars that you won't eat my shit. Joe ate it, Bill had to pay a hundred dollars. They keep bouncing. Joe was feeling sorry for himself, so he says, "Bill, I bet you a hundred dollars you won't eat my shit." "I will." The bet's on. Bill ate it, Joe made a hundred dollars. They keep bouncing. Suddenly Bill says, "Joe, I reckon you and I had some free shit.If we imagine that the names of these cowboys are not Joe and Bill, but Trader and Broker, then every second argument ends with the fact that the profit of both heroes is zero.
Moral:
Before you loudly and blatantly state something in the thread and call someone illiterate and do not understand basic things, it is worth taking care to cover your own nakedness, to understand what we are talking about, what formulas are used and what values. Then you are less likely to be confused.
Exactly the opposite is true :)
In short, you have to substitute, as you put it, for the maximum available, but not just substitute, but with some nuances.
I will not torture the humanitarians and will outline the proposed solution for the part of the formula related to the Recovery Factor.
Disclaimer 1: I give the solution for the case when we do not want to change the components and do not discuss the validity of the coefficients (1; 1,5; 0,5) in the formulas, because at the moment I do not know the reason why such coefficients are introduced. It is possible that the coefficients are correct.
Clause 2: The decision is based on the principle that the degree of difference in an indicator between participants should reflect the degree of practical difference in trading results, rather than a formal difference in numbers.
So,
1. The Recovery Factor shows to what extent the trading system is able to recover from the maximum historical drawdowns. Though by drawdown in the formula we do not mean the drawdown we have in our formula, but the maximal drawdown of the balance, i.e. the amount of losing trades. Therefore, recovery factor cannot be calculated for participants that do not have a single losing trade (it is highly probable to indicate "oversubscription"). (Those who argue that division by 0 gives infinity, I will refer them to the arithmetic course, where they can learn that the operation of division by 0 in operations with real numbers is meaningless, because it generates uncertainty. The popular saying is "you can't divide by 0").
2. At the same time, 1 small losing trade with all other profitable trades can give us a very large recovery factor - 500, 1000 or 5000. But will it greatly distinguish the systems and speak about their quality relative to each other? Not at all. How can we evaluate trading systems/strategies using the Rec.Fact, even if not to the 10th digit accuracy, but without gross miscalculations as applied to short-term competitions? For this, if you refer to many resources, where the trading systems indices are discussed, you can find out that there is a certain range of FS values, which characterizes the trading systems of this or that reliability class. Quite reliable trading systems are those that have FS = 15, professional trading systems have FS = 20 or more. I am quoting these figures to illustrate why I am proposing to introduce a certain limitation on the maximum FS value taken into consideration when calculating the results in our competitions. I would like to propose that the number 20 should be the maximum FS limit that will affect the results of our results calculations. What does this mean? This means that if the FS of any participant is >=20, we make it 20 for our calculations. For participants who do not have losing trades, we also set FS = 20. It will turn out that all who have a high PV (including those who have 0) receive the maximum by this part of the formula, i.e. 0.5 points. The rest of the participants will get a fair share of the 0.5 point, which corresponds to how far off the bar they are and in relation to the other participants. If there is a numerically justified objection to the maximum of 20, another value can be discussed, but the introduction of an upper limit is necessary andin principle.
Thus, before calculating on the basis of the existing formula, it is necessary to normalise this value and not just use the FS from the signals service.
Score(Rec.Factor) =(Rec.FactorNormalized( K ) - min(Rec.FactorNormalized(1:N)) / (max(Rec.FactorNormalized(1:N)) - min(Rec.FactorNormalized(1:N)) * 0.5
K - the participant number in the table
N - total number of participants, on which the calculation is carried out
Rec.FactorNormalized ( K ) = PV[participant]norm=
if at the signals' site PB[participant]=0 or PB[participant]>=20, then PB[participant]norm=20
if on a site of signals FS[participant]<20, then FS[participant]norm=FB[participant]
min(Rec.FactorNormalized(1:N)) - is the minimum value of the normalized recovery factors among all participants
max(Rec.FactorNormalized(1:N)) - is the maximum value of the normalized values of recovery factors among all participants
Such a small and uncomplicated transformation adds meaning and removes any uncertainties. i.e. we don't change the formulae, we just prepare the value of FS for its substitution. Two birds with one stone.
Here I can agree with what has been written
Here's where I can agree with what's been written.
Great.
Vitaly, to be honest with you, I am still unclear about the final calculations and nuances for the participants in the first nomination + some other things.
1. I'll start with Something.
According to the terms of the competition, all trades must be closed at the end. How will the participant control closing of transactions? Does he have to be able to close his trades at the end of the competition? What will happen if he did not have time to close, etc.? How counting is done in this case (if there is and no disqualification).
2. Here we have closed the deals.
Why and for what reason is the calculation shown in the spoiler:
If we have closed all trades, the Balance = Equity.
I do not see the dependence of the calculated value on the drawdown. The result is the same.
3. In general, as far as I understand from available information, now nothing stimulates participants to keep a small drawdown in the fight for the first nomination. After all, if you can pursue and 10% and 70% of drawdown, the main thing - it is forcing the risks just do not merge into a complete zero...
Is it so?
4. A wish.
We would like to have 2 tables on the main page for each category separately. - This is for clarity.
In the 1st table, all indices are calculated and sorted by balance (as the future total index). Skilled traders will estimate the chances of losing a position after closing trades at the end by themselves on the basis of drawdown and equity.
In the 2nd table, we simply leave the participants sorted by their current balance that can still claim a prize in the 2nd category. There is no need to recalculate anything here, just take the data from the 1st table.
5. Proposal for discussion
In case none of the remaining participants could stay within the limits of 10% drawdown, but they want to give the prize in the second category, the prize to the participant who has the maximum score among the 20% of participants with the minimum drawdown. If it is considered that the Second Prize should not be awarded in such a case, it would also be understandable. After all, then those who have applied for the second prize, realizing there is nothing to catch in the second nomination, can increase the load on the deposit and with their super-strategies will start to "tear up" everyone in the first nomination :)
There is an anecdote about cowboys:
If you imagine that these cowboys are not named Joe and Bill, but Trader and Broker, then every other argument ends up with both heroes making zero profit.
Moral:
Before you loudly and blatantly state something in the thread and call someone illiterate and do not understand basic things, it is worth taking care to cover your own nakedness, to understand what we are talking about, what formulas are used and what values. Then you are less likely to be embarrassed.
It does not make sense and is not necessary to physically close positions by the end of the contest, it is enough to achieve virtual closing of positions by fixing the state of trade at that moment. After all, we are dealing with real trading, which can be continued after the end of the contest, and premature closing against the rules of the TS, can harm the future performance of the account.
I want to ask you - an experienced theorist, in what cases can we talk about dividing by 0 in the formula FS = Net Profit/Maximum Drawdown? For your information, the maximal drawdown can never be 0. The moment you open a position, you already have a drawdown in the spread, and even if the market goes in your direction right up to the close of this trade, the maximal drawdown is fixed at 1-2 points, depending on the size of the spread at the moment of opening. Now you can see for yourself what your reasoning about FS is worth. Even in the case where all positions are closed in the plus, the resource "Signals", quite rightly, fixes the maximum drawdown that took place during the trade, the so-called "hidden drawdown".
I have, however, given an anecdote showing that profits can be zero not just once, and not rarely, but often. i.e. as if gently hinting that before you use loud and instructive words (this time, "learned theorist", "to your knowledge", "even theoretically", "what is your reasoning worth"), understand for yourself what we are talking about: which formulas are used, which values. If you want to participate in the discussion, change your tone from emotional and instructive to working and try to really understand first, so that you don't have to feel embarrassed later, when you realize that you were wrong and argued inappropriately. I can understand - young, blood boiling and all that...
For everyone's benefit, here's some background information on the formulas used
1. as it happens, there are actually several different formulas for calculating the indicator called "Recovery Factor", which give different results (Google to the rescue).
When using the term "Maximum drawdown" you should always specify the drawdown we are talking about. Otherwise it creates uncertainties. For example, the tooltip in the MQL signal service, does not specify the maximal drawdown value, and you will not get the value of the recovery factor mentioned there, if you substitute the value of maximal drawdown in the left picture in their formula (Net Profit/Maximum Drawdown) (for such calculation, this value must be set in dollars - this value will appear when moving the mouse pointer to the drawdown field on the left).
For the calculation of indices, the organizers of the competition parse the ready values to be substituted in formulas from the MQL signals website. Including the Recovery Factor indicator
About values, including where the discussed zero came from
3 Now, if you open statistics of any signal with no losing trades, you will face one more thing: for some reason these signals have a Recovery Factor equal to 0. How can we deal with your two statements, Yusuf, that the Recovery Factor cannot be equal to zero and thatthe resource "Signals", quite rightly, fixes the maximum drawdown that took place during the trading process, the so-called "hidden drawdown"?
The answer is below - below the picture.
I wanted to open the signal of the leader according to the balance of Elena, but the signal turned out to be hidden for some reason...
Let's open the signal of the second leader Alexey Volchansky who has no loss trades (I hope Alexey will not mind advertising)
You will probably have to deal with it in the following way:
1) The Recovery Factor should really not be equal to 0 and there is a programming error. Unless some secret coefficient in the secret formula applies :)
2) Resource "Signals" under "Maximum drawdown" on the left means the maximum relative drawdown by equity (and this figure really takes into account what you write between"For your information" and before"Now you yourself understand what your reasoning about FS is worth". However, there is a "nuance" - unfortunately the Maximum Relative Equity Drawdown does not take part in the calculation of FS. You can imagine how embarrassing the situation is. I hope you understand now... Don't take my irony to heart. From the bottom of my heart and for the good of education only.
What is the maximum drawdown that takes part in the FS calculations? - The maximum relative drawdown on the balance sheet. Of course, the calculation formula is available, but the value is not displayed in its pure form (as a separate parameter) in the Signals section of the MQL site.
This is the "bouquet" formed and, as you can see, not out of place. We cannot use the 0 value from the website, because adding 0 to the formulas for calculating results will automatically make traders without loss-making trades less than 0.5 points.
I have suggested a temporary solution, which will be changed and improved later.
But it allows already now to correct the situation without changing formulas (only a preliminary normalization of parameter "a la Factor recovery").
P.S. I did not bring an anecdote about "nuance" - you can google its different variants.