Please, please, please.... You must not have fully thought of the implications of this new rule. It is completely absurd. Good trading is about managing risk.

There is NO way one can manage risk effectively if we are forced to use a fixed-size position for all trades in a given LENGTHY period of time. That is completely ABSURD ! nobody in his right mind would ever trade with always the same position size for one full month !!!!

What, if I have a drawdown of 50% of my initial account, I am supposed to trade the same size ??? that is completely and utterly absurd ! As equity goes down, obviously a proportionate risk has to be taken, not gamble the same amount hoping thigs turn around, possibly destroying all equity !

It is obvious to any trader that if the risk is high, the position size chosen will be small, and vice versa. Forcing the trader to choose a position size, for a FULL ONE MONTH of trading is completely unrealistic. It would essentially be based on luck, therefore doing the exact opposite of your intended goal !

A perfectly reasonable and realistic option (because that is what most real-life traders do) would be instead to limit to trade size to a maximum percent of the available equity.

Say for example, any one trade can not increase the use of leverage to more than 15%. That totally avoids the heavy gambling type positions, while still allowing the strategy to manage risk flexibly.

There are clear market conditions where one would want to trade higher, and others where one wants to trade less. Eliminating this make the contest just a programming exercise and not a trading contest.

Also, to manage risk is not only to choose trade size, it is also about choosing stops. One can trade big, with tight stops, or trade small, with wide stops. Who is to say one way or another is better ? not only that, but depending on market conditions the trade-off will change. Choosing the right trade-off is what can make or break a strategy. Eliminating that choice means we just have an artificial game not at all related to real trading.

I certainly hope you revise the rules to not enforce fixed size trading, and instead enforce maximum proportional risk trading, like explained above.

I totally agree with brainstorm.
Why exclude money/risk management in the contest?
Better increase the amount of time the script must run; 20 or 25 days would be appropriate I think..

Kind regards,
Sacrelege

Hi,

One question and a comment.

Q. May I trade different instuments ? eg. EURUSD, GBPUSD, etc. I understand to have only one position at a time, but I want for example to trade EURUSD if I get trade signal today and tomorrow AUDJPY.

Comment:
I think the simplest way to prevent gambling is to use fixed lot and more, specifiy it size during competition eg. 0.01. So, all strategies will use the same lot: 0.01, a reasonable value. The winner is the one which achieve the largest number of pips,
not the money. I think in this way you can remove the other limitations and let a profitable strategy to prove its power.
There would be no more need to have 1 open position or close an order profit is larger than X% of equity.
All strategies have the same '$ pip power'. If one wants to bet for 200 pips TP in one trade that is, it cant be in advantage with one which trades 20 orders betting for 10 pips or viceversa. The pip is the pip. The contest should plot the achieved pips and not the money/balance/equity. I think also in this case the answer to my above question would be 'Yes'.
I still think there should be a minimum number of trades/month , eg. min 5 trades.

Currently, if you let the strategy to choose its own lot and use it the same during strategy, we can still have gambling. One can still choose a big lot and go to the luck against a more profitable strategy. No place to explain more. You understand better than me.

A simple example:
#Strategy 1, makes 100 pips/monts, 10 trades and has a profit factor of 4; during contest it uses lots 0.01
#Strategy 2, makes idem 100 pips, 10 trades has a profit factom smaller, eg. 1.35; during contest it uses lots = 2;
It is clear Strategy 2 will win almost for sure against the Strategy 1 which is obviously more profitable than S2 due to its high profit factor: 4

So, I propose fixed lot = 0.01 for all traders, no other limitations, unless there are hiden things I dont understand/know.

Hello all,

Just wanted to ask some quick questions here. I have read through the contest rules but couldn't find the answers...

1) Do I have to have a live (real funds) account with Dukascopy to participate in the contest?

2) If no, I guess the 2 week demo account will be changed to 4 weeks so we can continue running our strategy all month?

Thanks in advance..

That or what I suggested (max percent risked equity) seems like a good tradeoff.

I definitely agree that fixed lot and winner measured by ending equity instead of ending pips profit, is a recipe for gambling. At least with what darlu72 suggests, the performance of the strategy is measured, no room for luck, and risk management is still part of the strategies. I still think it is best and more realistic challenge to measure performance by ending equity, but limiting the risk of each trade.

Yep. it is the amatuers and not the professionals who always trade with the sam e position size. The pros adjust size, and scale in and scale out almost all the time.

Perhaps a good solution to this would be to have a maximum net position allowed at any time. This would truly avoid gambling while allowing various trade sizes and hedging, techniques which are WIDELY used. Surely Dukascopy must have noticed an influx of customers after America disallowed hedging in the same account? I know it is certainly one of the reasons I'm looking into opening an account here.

Trading 1 million lots all the time and then suddenly trading with 100% of margin at the end of the month would indeed reveal gambling. However, what if one trades around 1 million lots all the time, and then in some cases only trades 100,000 lots, or 1000 lots? How can trading SMALLER than normal be seen as gambling? It is obvious this is taking on LESS risk, not more. It is only taking on huge positions that would truly be seen as gambling, not simply using different sizes at different times. Thus, setting a maximum exposure limit that the contestants must stay under for the whole month would be the best of both worlds for both Dukascopy and the contestants.

1) No.
2) The demo is valid for 1 month.

First of all, we thank those who left comments and feedbacks!

As we can see, the right combination of rules and methods is not as evident as it seemed at the first glance. We have very different opinions even amongst you guys.

@brainstorm
Of course, limiting the use of leverage is a good idea. For a live trading! But not for this contest. It would be very similar to a driving a sport car with a 30 km/hour speed limit. After all, we have a competition and everyone should compete in the environment which is similar to a real trading environment. Let me remind you that the use of leverage is already limited to 100%. Lowering the limit does not make any effect but reduces the maximum position size. But the contestant still can use the single-lucky-winning-trade-strategy.

@darlu72
Pips approach is a little bit more efficient in regard to eliminating the gambling but it destroys Money Management which we would like to keep in the contest. The contestant still can average their positions, use a martingale strategy, and increase the exposure to its maximum level.
Concerning the profit factor method, it’s not efficient too. The single-position strategy described by Ocire can skyrocket the profit factor to a very high value.

@inkexit
A maximum net position rule is more or less the same as the idea by brainstorm to limit the use of leverage. Besides, 100 000 USD balance also has a certain exposure limit. However, we appreciate your comment.

Attention everyone! We are ready to offer you an alternative rules for the strategy. We can remove a requirement to have a constant trade size. There will be one simple rule only:

A profit on a single position must not exceed 50% of the total profit of all positive trades.

In case if we have such a position, we remove it from your balance. What are the consequences of this rule? You need to trade with more or less even targets trying to win not with a single trade but with many even trades. The martingale strategy does not work because the last winning trade may be cancelled(until you have a really good strategy with stable signals). The single-trade strategy described by Ocire also does not work as the first big trade will be cancelled too.

We don’t really have a lot of time till the start of the next contest! We would like to hear your comments on this proposal as we consider it to be almost ideal for eliminating any gambling or accidental luck.

What about the 10 trades minimum rule?

Wouldn't these rules pigeonhole strategies to short timeframes or even scalping?

good idea, but if there are no other rules, there is a simple way to modify my previous "strategy": instead of one big trade, make 10 small trades at exactly the same time for each 1/10th of the amount the big trade would have been.
that way, you have 10 trades each having 1/10th of the total profit... that would be legal according to the proposed rules.

how about: a profit on a single day must not exceed x% of the total profit?

works for me. Now If I can only get some good answers to my java questions elsewhere on the forum and finish up reading my intro to java book...

1. I agree. Rewrite final clear rules so we all know them before start.
2. I propose to add some fun and give the Contest a name: eg. 'The Pips Battle'
(I am tired and sick off now of coding, java, exceptions, references, threads, testing, crashes, etc, so some relax would be welcomed

We would like to make a precision in the rule!

A profit on a single position must not exceed 50% of the total profit of all positive trades.

So, the rules which will be removed from the Contest Rules:
1) You do not need to have the constant trade size (Trading Rules, Nr. 6.2)
2) There is no a requirement to keep the profit on a single position below the level of 25% of your current balance (Trading Rules, Nr. 6.4)

Added Rules:
1) You need to make sure that the profit on a single position is below 50% of the total profit of all positive trades.
The maximum amount of simultaneously opened positions is still 1. No change in this rule!

As a result, you can choose any trade size, you can have any profits you plan! Keep positions long-term or short-term! The only obligation is to have even profits on positions.

No change. you still need to have at least 10 positions.

I like the new rule! In my opinion, it's more flexible. Do I need to trade with a single position only?

Now I'm confused.

Are all the other rules not listed above still in effect?

I thought you said the "no more than 50%..." rule would be the ONLY rule?

Hello,

I think the new rule is a good step is in the right direction, but we still have to do something about those filler trades. Suppose I got lucky and win 2 positions with very high leverage, each position will earn me $25.000 for example, then I will trade with a very small amount of money to respect the 10 positions rule. In the end of the month, with some luck, I'll get a balance around $150.000 in perfect compliance with the new rule.

Regards,

Jakil

I think you misunderstood my argument. I was refering to leverage as expressed at the bottom line of the JForex client (among equity, profit/loss, Use of leverage... which is basically an expression of the position size in proportion to the account equity), not the account leverage.

Sounds like the new rule does something similar, it essentially limits the risk/position size in proportion to past performance, which is close to equity fluctuations.

It seems like a flexible and realistic rule.

just to make sure I understand the new profit limit rule, complying strategies could have code kind of like:

public void onMessage(IMessage message){
    switch(message.getType()) {
        case ORDER_CLOSE_OK:
            mTotalPositiveProfit += (order.getProfitLossInUSD() > 0) ? order.getProfitLossInUSD() : 0;
            break;
       .....
    }
}

public void onTick(Instrument instrument, ITick tick){
    for(IOrder order : engine.getOrders(instrument)){
        if((order.getState() == IOrder.State.FILLED) && (order.getProfitLossInUSD() > 0.47 * mTotalPositiveProfit) ){
            order.close();    
        }
    }

    ....
}

Here we make sure subsequent trades comply with the trade profit history.

But that is not necessarily the only way to comply, since one could trade huge at first and keep trading huge, within the 50% total profit, or progressively scale up or down position as risk or time or profit goes by, all the while staying within 50% of all profits made. Seems like a very realistic rule.

Question:
if I restart my strategy (it happens... reboots, etc.), how can the strategy know how much the total positive profit has been so far ? Strategies do not have access to the trading/profit history.

The only ways I see around that, is to have the strategy read this total from an @Configurable variable, or store it in some file in OnStop() and read it back in OnStart().

Please advise...