Why we are taught to set risk/reward ratio larger than 1

[mishak]

Risking $1000 to Get ... Part 1/6

This is the part of answers to questions posed here by overload

Many of us have learned that we should trade with risk/reward ratio 1:3 or 1:2 or at least 1:1.5
This money management rule means that before taking a position on the market (forex or stocks etc.) we should analysize what kind of profit (reward) we are expecting to achieve and compare it to the amount of money we are ready to lose (risk).

For example, say EUR/USD rate is 1.1600 and we wish to buy because we expect the Euro will go up to 1.1675 - this is our target.
If however we are unlucky and the Euro will go down, we are ready to recognize our failure when the price reaches 1.1575.
In this example our reward is 75 pips and risk is 25 pips. So, we are trading with risk/reward ration 1:3.
(btw, we should set T/P (take profit) order at 1.1675 and S/L (stop loss) order at 1.1575 in this example).

Why we are taught to set risk/reward ratio larger than 1
Because when we (unexperienced at that time) started to trade forex, most of us had the slightest idea where the EUR/USD will go from 1.1600. And we were at the same position after the first, second, third (you name it) deal.
Our instructor had expected this. And he wished that we stayed in the market longer to get our experience (or to pay him more monthly fees, or spreads - please no offence intended).

So, he told us:
if you trade with risk/reward ratio 1:3, you can afford to lose three deals of every four and still break even.
Look, say you lose three times of 25 pips for -75 pips summary loss . However when you'll win your fourth deal for +75 pips , you still keep all your money -75+75=0.

Then he continued:
if you perform better, say lose two deals of every three, then you will make profits.
Look, say you lose two times of 25 pips for summary -50 loss, than win the third one +75. Your total balance is -50+75= +25 pips - good profit

When you get experienced, you may drop risk/reward ratio to 1:1,5. In this case you can still lose 3 deals of 5 (60%) and break even !
Anyhow, please keep risk/reward above 1.

Was he right ?
some people doubt

to be continued ... here

©MK 2003

What do you think ? Please add you comments

[vikingtraider]

to overload:

My suggestion to beginners is to take signals as soon as possible. While risk reward is definitely important, I think it's not the key concern.

The reason is once signals are given, it is very easy to identify when you are wrong. What this means is your risk is very small...

In the example you gave.. if a signal was generated at 1.1685, I would not take the trade if it was trading at 1.17 presonally. While 1:3 is still good, I always look for higher reward/risk ratios because I realise that I AM WRONG MOST OF THE TIME!!

My trading statistics show me I am wrong about 55-60% of the time, yet I am profitable. Obviously my wins have to be huge.

For example, I trade the aud/usd pair a lot, and my average losing trade is a loss of 16 pips, which is apparently really low.

Just something to think about.. not necessarily something that suits your trading style

[Truville]

I’m not sure I get this.

Yes,1:3 is good, but 1:4 is better, and 1:8 is really great! Why not 1:8.354? How about 32:98.321?

The risk/reward ratio you are describing here has no basis in reality. It’s just an arbitrary number. Nice and round so it’s easy to calculate, and not too high because that would seem odd.

Let me suggest something. Instead of letting a signal make your entry then deciding arbitrarily on how much you are willing to risk/make; how about letting a signal make your entry, AND your limit, AND your stop?

For example, if you use MACD for entry, why not let it ride until MACD says otherwise? Let it tell you what to do!

Stating from the beginning:
“I will risk $1000, or $100, or $231.45”
“I will make $350, $34.21, or $12.99”

Is no better than guessing.

[vikingtraider]

Truville - i agree.. my trading style is very close to what you say when profits are running... but what happens if you generate a buy signal at 1.17 and the target is at 1.175 and stop is at 1.165?

I woldn't take that trade personally..

you need to be able to identify you are wrong very soon...

but definitely agree with your profit riding style..

[rezo_s]

What happens if you don't get a "cut loss/close the deal signal" if it goes against you...lets say 100pips or 200 pips...? well, little overdone, but what I mean is:

You should define how much are you willing to risk...and if not exact target, then at least estimated. You should know how much to trade for.
One cannot define what kind of money management to apply to his/her trading style until s/he has got test period of the system. Even if its fully automated system, r:r ratio should be one of the filters (just as vikingtraider suggested) to choose whether to trade or not specific signal.
If you have a very volatile trading (for ex.) on Euro, lets say its in a range of 200 pips for last week or so, and you get a buy signal (even a very strong one) very close to upper border of the range - what you do? will you trade the signal? You know that there was a very volatile trading recently, and the price may bounce back from that upper border just like it did during last week...what you do? will you follow the signal? on the one hand, you got a buy signal, on the other hand you know your first target (level) is very close to current price (lets say 20 pips), and the range was 200 pips, so the level where the price bounced back from recently (the recent strong support area) is 180 pips away...will you trade?
if the buy signal occurs near the lower border of the range (pattern), than it is a good trade, because stop may be put below the range border...
if you wait for signal to give you exit signal when you are in trade already...woow, man it is way too dangerous approach. So what you suggest? not setting the stop order? is that what you say?
You should define the stop and target before the trade, and it REALLY is best for trader ratio to be >1. Sure, during the trade, you may get an exit signal (cut losses or take profits before estimated stop or target), but initially those levels should be at least estimated.

Why >1? I think this is well explained by mishak, I also added some comments in "part 2" of this post (see link in mishaks' post).
I don't believe there are many traders having 100% of winning trades. I doubt there are many with 80%. I believe 60% is pretty good, but it also depends what is the avg gain in those 60% winners and what is the avg loss in rest 40% losing trades. Based on this and some more data (which can be obtained only after test of system - and preferably on real account), one can set filter on which trades to take when there is a signal, and which should be passes.

Just my 2 c,

Rezo

[rezo_s]

I took some time to illustrate the example I wrote about:

[mishak]

Hi Truville

You are absolutely right.
This is a good approach and some hedge funds use it. In messages it is refered sometimes as "black box" or "model" trading.
Unfortunately it is not available for everyone:

• some novices have no idea about TA signals. They are just looking for $$$.
• it is difficult to design the proper model signal even for advanced traders. For example, it should not give you two consecutive Buy signals, etc...
• you need some kind of trading robot which will follow real prices and compute signal for you while you are playing golf. And here come problems with compatibility to your broker's platform etc...
  Of course many problems can be solved, but not on the Beginners Forum.

Some novices have a little undersanding of money management (please no offence intended). So, I've started this thread to bring different opinions on risk:reward into collision

[Truville]

Rezo,

Your illustration and explanation proves my point exactly. It’s one thing to look at a chart, determine probabilities, and decide whether to enter or not. But it is quite another to simply choose a 3 to 1 profit to risk ratio. The former is taking clues from the market. The latter is simple guesswork.

You stated “You should define how much are you willing to risk”. I agree. But shouldn’t that number be a function of the market, and not some random number that seems “sufficiently high?”