Low Risk, High Reward Trading Experiment

Mathematics of low risk / high reward trading

For those who haven't already done the math, here is the basic math, as I see it, of our low risk / high reward experimental trading system that is in development.

(Of course, "low risk/ high reward" is only true of the parameters of a single given trade set-up; if very few of the trade set-ups are successful, then one's account will diminish. The risk of the overall trading program is a product of a combination of the reward/risk ratio and the win / loss ratio. If the win / loss ratio and reward / risk ratio aren't high enough, or if the risk factor employed is too large, then the account will diminish or be liquidated.)

Let take yesterday's Trade Signal - "TS#67 Sell EUR/AUD at 1.6878" as an example (entire trade signal as sent out is reproduced at the bottom of this message for reference).

Let's assume for this example that you are trading in a standard Forex mini-account with 200:1 leverage. Your initial capital is $1,000.

Let's assume that the trading system has been fully developed and back-tested, so the performance characteristics combined with your personal risk tolerance should logically determine the risk factor you will employ. (Risk factor is the percentage of your capital you will lose when you are stopped out in a trade.) Let's say you are an adventurous person, and you are trading with this system as an alternative to buying lottery tickets or day trading in the options market. In other words, you are highly risk tolerant and want to create an opportunity (or chance) of making a fortune quickly - like buying a lottery ticket, except with much better odds. This isn't your way of paying your rent, it's your way of opening a way to possibly becoming a millionaire in a short time.

So, if you are that kind of person, and in that situation, then you might employ a risk factor of 20%. (Our projected system characterisics when the system is perfected suggest that a 20% risk factor could be realistic.)

So, we will assume a risk factor of 20%. That means that you will be risking 20% of your capital going into each trade.

In the first trade, you will therefore risk 20% of $1,000, which is $200.

In this example, the stop plus the spread equals 33 pips. If you hit the stop and thus lose 33 pips, you want to limit your risk to $200. So the number of lots entered into will be 200 divided by 33 which equals 6.06. Therefore you will enter the trade with 6 mini-lots. For each mini-lot, one will either win or lose $1 per pip.

If one hits the stop, one will lose 33 pips, which is equivalent to 33 pips times 6 lots times $1 per pip per lot, which is about $200. It one hits the stop (loses the trade) then one will lose 20% if your capital.That is what you want - to limit your risk to 20% of your capital.

On the other hand, if one wins the trade (hits the first target at least, and exits there), then one will win 172 pips. With 6 mini-lots at $1 per pip per lot, the result of the win will be 172 pips gain times 6 lots times $1 per pip per lot equals $1032.

So in this example, with a 20% risk factor, entering the trade one is risking $200 to create an opportunity of earning $1032. The ratio of $1032 to $200 is about 5.2 to 1. That is the Reward : Risk ratio.

A win of $1032 starting with $1000 means one's account will grow to $2032. That represents a multiplication of one's account by 2.032 times.

If the trading program has a win : loss ratio of 2 : 3 (in other words, one wins two trades for every three trades one loses), then the number of winning trades in a fifteen trade sequence of trades will be about six, and the number of losing trades will be about nine. In this example the multiplication factor of the capital over 15 trades will be 2.032 multiplied by itself six times, and .80 multiplied by itself nine times.

2.032 (exponent 6) X .80 (exponent 9) = 9.45

That means that based on the above assumptions of a 20% risk factor and a win:loss ratio of 2:3, a trading program consisting of 15 similar trades to TS#67 will produce a multiplication of capital of 9.44 times.

One's capital after 15 such trades will be $9,440 using such a trading system.

After 30 such trades the capital will grow to $89,000.

After 60 such trades one's capital will grow to $7.9 million.

Eureka!

_________________________________________________-

Original trade signal as sent out yesterday:

Friday 5/19/2006 9:00 ET
TS#67
Sell EUR/AUD at 1.6878 or higher
Stop is 1.6896
Target #1 is 1.6690
Target #2 is 1.6650
Target #3 is 1.6585
Reward is min. 172 pips
Risk is max. 33 pips
Reward : Risk ratio is min. 5.4 to 1

This is a trade that might not be tradeable by others, as the market is moving very fast, however, it is system compliant and appears like a good trade if it is tradeable and meets the system criteria of independent traders as well.
For research and testing purposes only. Trade at your own risk. Trade risk capital only.

Hmm... don't know what to think of this...

The risk factor (percentage of capital one loses when one loses a trade) that a trading system can tolerate is a product of a combination of the reward/risk ratio and of the win/loss ratio. The appropriate risk factor is the highest risk factor that the system can handle without creating an excessively high probability of the trader having his capital liquidated before he or she hits his or her financial target.

If a trader is willing to accept a 9,500 to one or less probability of losing 95% of his or her capital in the process of the trading program, then he or she should use the corresponding risk factor based on the probabilities involved. With a win / loss ratio of 2 to 3 ( a 40 % win rate), a reward / risk ratio of 5:1, and a 20% risk factor, the probability of losing 95% of one's capital is only 1 in 9,500. That is a very acceptable risk, I believe, especially if one is only risking $250 to $1000.

The lower curve on the accompanying chart represents the performance of the trading system based on a 20% risk factor.

For the higher risk-taker, a 25% risk factor could be employed to obtain a much higher rate of return. Based on the same win / loss ratio and reward / risk ratio as in the above example, the probability of losing 95% of one's capital with a 25% risk factor goes up to one in 200. For someone willing to accept that risk, they should employ a 25% risk factor.

The resuts of this latter scenario are depicted in the higher curve on the accompanying graph. Over a course of 45 trades, the return on capital is about four times greater.

So the question is: Up to now that the system is still in beta mode, what are the true statistics? What % of the trades have been winners and how many losses? Just curious. It would be great to see if while this period of refinement this ratio gets better with time.

If you look at the timing...I would say the number of trades reaching profitability is close to 90%...with this said...Very few of the trades have reached their intended profit targets with most of them being reaching their stops.

I believe with good money management, you will do pretty well with this system, primarily because a lot of the recent trades have made 100 or so pips before reversing and hitting stops (so extrapolate yourselves). This means that while you dilute the profitability factor of the system, for those with less tolerance can use this system to still obtain a decent amount of profit.

Again...the system prides itself on making a few really good calls and keeping losses to a minimum.

It's actually quite simple to back-test this sytem...simply open up your charts (if you historical quotes) and just scroll to the chart times and see for yourselves...there are only 70 or so trades, so you'll see how the trades themselves go.

Please keep in mind that A FEW of the calls were REALLY time-delayed (often overnight), so, the profitability of being able to trade the systems in real-time from the forum has been very difficult.

This system shows a lot of promise, because I think the built in money management as well as the technical principals are sound. This is my opinion of course.

The low risk / high reward trading system and its implementers are still in the middle of a steep learning curve. There are some major changes and minor adjustments still possible that can have a dramatic effect on the win / loss ratio and reward / risk ratio.

Some considerations:

The trades are either right in principle or wrong in principle (the overal trend analysis and two endpoints of the predicted trend were either valid or not.)

If the trade concept is right in principle, then it's still a giant step to accurately define the entry point, stop, and exit point that will work. A mistake on any of these three numbers will switch the trade from a win to a loss. The entry point might be wrongly placed, so the trade is never entered. Or the stop can be hit. Or the target can be too far away and not be hit, or it can be too close, so potential gains are not realised.

We have made every possible mistake so far. We have imagined trends that never happened. We have missed entry points. We have hit stops (often exactly, or by two to six pips). And we have missed targets by a small or large margin.

However, our overall impression from our experience so far is that we will actually be able to employ this model to achieve a better than 40% win rate at trades with a 5 to 1 or better reward : risk ratio. In fact, I can envision attaining a 70% percentage of wins with a 10 to 1 reward : risk ratio. It could take a year or more of hard work, but it is entirely possible. We haven't accomplished this as yet, but we have done actually quite well we feel. Often when we examined a failed trade, it was obvious why it failed. We failed, not the system. That is actually very heartening. It means we have a potentially great system, but that we need to better understand and be more meticulous and thorough to implement it.

A number of the trade signals were based on the same incorrect analysis. In practise, if we predict a new trend we would choose the currency pair with the best statistics (reward / risk ratio, anticipated time to completion and confidence in) and issue that as a trade signal. For instance, sometimes the GBP/USD will move the most, whereas on another occasion the USD/CHF or USD/JPY might be the big mover. We can determine this through both analysing each currency pair as well as analysing the cross pairs to confirm what the biggest moving currency is likely to be. We are quite good at this.

Generating a number of simultaneous trades in different currency pairs based on the same basic concept is not desirable from an investment perspective. It is better to simply invest in the currency pair that is likely to move the greatest amount, and that we can safely enter with the smallest stop. In our testing program we just generated multiple signals in order to get more practise and experience with our system. For example, there were about four failed trade signals in a row several times, which were essentially the same failed concept of a trend reversal, applied to several currencies at once. In a sense, this was only one failure, not four or five.

There were also many trades which were almost perfect, except for the stop being a few pips too close to the entry point. We feel that we can reduce the incidence of this kind of failure through more careful analysis, and possibly increasing the size of the stops a little.

The biggest problem so far was in the larger trend analysis, insofar as we have predicted the beginning of a larger trend that never actually occurred. The cause of these errors needs to determined, and remedies implemented if possible. We feel confident that with careful and detailed analysis and backtesting, and more learning perhaps, we can surmount this challenge and not mistake the middle of a trend for the endpoint of a larger trend.

When we actually trade with live money or sell signals (if we decide to do that one day), then we will take only the best studied, most optimum, trades in which we have the most confidence. So we will automatically discard many of the less clear and less profitable trades that were included in our testing program.

In fact, we did implement two trades last week. Both did extremely well. So by being highly selective and very diligent, we have gotton off to a very good start. A small account has grown substantially, even with a modest risk factor. We anticipate more big winners and not too many small losses.

And we wish the same for everyone else.

Good night.

Hello, just wanted to throw my micro pip in

When trading my account i think of 20% on a single trade as super very crazy high risk.

say you had an account of $1000 and you sustained 3 losses in a row

1000 x .2 = 200
800 x .2 = 160
640 x .2 = 128

512

I would be too upset to loose 1/2 my account in 3 trades to continue traded for at last awhile.

But seeing you have a system with 90% win ratio I think your max draw down would never be more then 5 losses in a row, but i think 2 losses in a row might come up often.

The way i get larger amounts of risk into play is to trade more then 1 pair at a time. say i was feeling lucky and i decided to go 3% risk per trade, if i get 3 trades in I am actually running total risk of 9%. The odds of them all failing would be 1 in 8 as opposed to 1 trade of 9% which would be 1 in 2.

I will hafta say that your trading system so far is amazing

Hi Witchazel

Like you I have a small account and the thought of risking such a high percentage is scary. In fact, I do not do it.

At the same time, it must be said, that mathematically unless you have a long run of consecutive losses, the higher the percentage you use the better off you will be. If you want I can show you the math behind that statement. In fact, one of the best scenarios is using 85% of your account on one trade with a 15 pip stop. Unless you have more than 7 consecutive losses you end up doing extremely well.

All of that having been said, I am not so sure of my ability to avoid 7 consecutive losses. I have not had that many consecutive losses but I have to deal with my fear of that occurring. In other words, while I see the mathematical wisdom of using a large percentage of the account on one trade, I have never done it.

The Mathematics of Low Risk / High Return Trading (cont.)

Above, the comment is made "But seeing you have a system with 90% win ratio I think your max draw down would never be more then 5 losses in a row, but i think 2 losses in a row might come up often."

The system is currently achieving less than 30% of wins, not 90% wins. However, based on our experience so far we believe that a performance of 50% to 70% of wins is theoretically possible, with a 5 to 1 minimum reward : risk ratio.

Depending on the reward :risk ratio and win:loss ratio, the proper risk factor can be calculated. Also, the distribution of wins and losses is a factor. Are the distribution of wins and losses random in occurence, or do they occur in clusters? The proper risk factor will optimize growth while maintaining the probability of being margined out during the trading program to an acceptable level.

Trade system performance optimisation means adjusting the risk factor based on logical criteria, not emotions. A trade system performing at optimum performance will experience large drawdowns by most human judgement.

The goal is to optimise capital growth while surviving possible multiple losses. The better the win:loss ratio and reward:risk ratio, the less likelihood that there will be multiple losses and the less severe the consequences relative to the impact of multiple wins, which will dramatically enhance the capital.

Here is a graph of the number of consecutive losses required to reduce the capital account by 95% as a function of the risk factor. See attachment.

Next I will deal with the probabilities of losing one's capital in the normal course of a trading program as a function of the various variables, which are win/loss ratio, reward/risk ratio, distribution of wins and losses, and risk factor.

It is true with the math, 1 step back 5 steps forward, you are going to win. I just am not that brave

ITME, Thanks for your comments. Regarding this, what do you think about lets say grading or classifying your trades to A, B, C to keep it simple. Trades A would be the ones you would definatly trade on the information you have right now, B would be the trades that you highly reccomend, and C the trades that have a high probabilty of going right with the biggest chance of failure,
obviously assuming that ALL trades you recommend have a good probability of a profitable outcome.

Like this, some of us could test with only A,B or C type of trades and it would definatley be useful for your further analysis. Besides we could try getting a 5% risk in A, 3% risk in B & 1% risk in C (or 10%, 7%, 5% or whatever)

What do you think?

Thanks, XRod, for an excellent suggestion. We will try to implement it immediately. Don't forget to apply all of your usual standards of analysis to independently corroborate every aspect of our experimental trade signals.

Great!

Monday 5/21/2006 21:30 ET

Revision to TS#67 (Sell EUR/AUD) either exit or raise stop and re-enter higher up

After the sell trade signal was issued, EUR/AUD moved down about 60 pips to 1.6816, then headed up again. We were incorrect in calling the top of this upwards trend. AUD/USD is still descending more strongly than EUR/USD, which is moving up EUR/AUD past our stop. The current price of EUR/AUD is 1.6898. We expect it to rise even higher. For those in this trade, we recommend either exiting it, or raising the stop to 1.6954.

For those not already in this trade, we recommending selling EUR/AUD at 1.6930 with the stop set at 1.6954

Targets are the same as before. Target #4 is 1.6400, offering a 520 pip profit.

Confidence level: B


Original Trade Signal:

Friday 5/19/2006 9:00 ET
TS#67 Sell EUR/AUD at 1.6878 or higher
Stop is 1.6896
Target #1 is 1.6690
Target #2 is 1.6650 Target #3 is 1.6585
Reward is min. 172 pips Risk is max. 33 pips Reward : Risk ratio is min. 5.4 to 1
This is a trade that might not be tradeable by others, as the market is moving very fast, however, it is system compliant and appears like a good trade if it is tradeable and meets the system criteria of independent traders as well.

For research and testing purposes only. Trade at your own risk. Trade risk capital only.

We have been chasing this trade. After being stopped out twice, we feel quite confidence that we are very close to the final stop of the trend. The current price is 1.6983. The top looks like 1.6985 - 1.6995. A short here with a stop at 1.7005 looks like a good, solid low risk, high reward trade.


Revision to TS#67 (Sell EUR/AUD) either exit or raise stop


Monday 5/21/2006 21:30 ET

Revision to TS#67 (Sell EUR/AUD) either exit or raise stop and re-enter higher up

After the sell trade signal was issued, EUR/AUD moved down about 60 pips to 1.6816, then headed up again. We were incorrect in calling the top of this upwards trend. AUD/USD is still descending more strongly than EUR/USD, which is moving up EUR/AUD past our stop. The current price of EUR/AUD is 1.6898. We expect it to rise even higher. For those in this trade, we recommend either exiting it, or raising the stop to 1.6954.

For those not already in this trade, we recommending selling EUR/AUD at 1.6930 with the stop set at 1.6954

Targets are the same as before. Target #4 is 1.6400, offering a 520 pip profit.

Confidence level: B

Hello ITME,

I want to say it's my own mistake that caused big loss.
I traded w/o stop loss and i used too many lots .
I followed the thread of mini king winner on this forum and located company called Phincorp , so far so good .
It costs quite a bit for the subscription though but it worths for the money.
Thanks for helping me to learn the worthy lesson in forex ;
nerver trade w/o stop loss and never trade with so many lots.

Bye and good luck for your research

I am glad you found a good analyst service. Phincorp is using some aspects of our system. They are very, very good. You have found a way to riches, I believe. Patience, discipline, clear logic and caution are so important in this endeavor.

EUR/AUD re-entry opportunity TS#67 This is finally it! We are quite sure that 1.7023 is the final top of this trend. A small stop at 1.7036 should protect this trade, which should be good for several hundred pips of profit.

Confidence: very high

______________________

We have been chasing this trade. After being stopped out twice, we feel quite confidence that we are very close to the final stop of the trend. The current price is 1.6983. The top looks like 1.6985 - 1.6995. A short here with a stop at 1.7005 looks like a good, solid low risk, high reward trade.

mmmmm....something contrarian seems to have occured with EUR/AUD

Well, this has been quite a fiasco! EUR/AUD keeps surpassing our entry targets. It has now taken a parabolic rise path and reached a high of 1.7056. This point (or localised region, to be more accurate), according to our analysis, is a point of extremely high likelihood of completion of upwards travel. There should be one more small movement upwards to about 1.7055 to 1.7060. A stop of 1.7076 is advised.