What do you think of Risk/Reward ratio? | Page 3 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Forum Survey: What Risk to Reward Ratio do you use? 3 replies

What you need to know about Risk-Reward Ratio 9 replies

Rationale behind "Risk Per Trade" and "Risk/Reward Ratio"? 8 replies

How do you compute risk:reward ratio? 4 replies

Risk Reward Ratio and Account Risk 30 replies

Attachments: What do you think of Risk/Reward ratio?
Exit Attachments

What do you think of Risk/Reward ratio?

  • #41
  • Dec 12, 2007 8:42am Dec 12, 2007 8:42am
When i started trading i learnt about R/R from many books. At first it makes such perfect sense. And then you try trading for real and you soon learn that it's only a small part of the whole.

No one will ever find a profitable forex system that has a target of 30 pips and a stop loss of 10 pips. Sure you only need to win 1 out of 3 trades to breakeven but there is noise which is random which is very likely to hit your stop and after spreads and commisions your account balance will go down.

If anyone can prove me wrong i would love to here from you.

I believe it's easier to make money with a 30 pip stop and a 10 pip target with the way forex prices usually move. Even when prices break S/R lines and the books say it will move now, it retraces. Often the second break is the one that continues - explain that to me.

IMHO the first thing to decide on is your trading time frame and then build a system to match. Only with trades lasting 2 weeks or more are R:R > 2
appropriate.
Don't kill your dreams - execute them.
  • #42
  • Dec 12, 2007 8:49am Dec 12, 2007 8:49am
1:9 and 1:17 Risk/Reward can be done! It called, Money Management. Most people failed on that aspect.
Trainee at UBS Investment Banking
  • #43
  • Edited 10:46am Dec 12, 2007 9:00am | Edited 10:46am
I prefer profit/loss = 1/(risk/reward). But let us use RR
Quoting Time
... If your risk/reward ratio is 1:1 then your win/loss ratio needs to be 50% to break even
If your risk/reward ratio is 1:2 then your win/loss ratio needs to be 25% to break even
If your risk/reward ratio is 1:3 then your win/loss ratio needs to be 33.3% to break even ...
Let us set w = Average_win_ratio

then w/(1-w) / (risk/reward) > 1
this is rather simple rule of combinatorics

suppose that: Average_Null_ratio << Average_win_ratio

PS: if you watch market there'll be some pyramid scheme... Differences are possible
  • #44
  • Dec 13, 2007 4:20am Dec 13, 2007 4:20am
I really prefer profit/loss = 1/(risk/reward). But let us use RR

I believe a real system at forex could be profitable with RR from 0.1 to 3.

Is there another opinion?
  • #45
  • Dec 13, 2007 4:27am Dec 13, 2007 4:27am
Quoting Time
Ok lemme put it this way.

If your risk/reward ratio is 1:1 then your win/loss ratio needs to be 50% to break even

If your risk/reward ratio is 1:2 then your win/loss ratio needs to be 25% to break even

If your risk/reward ratio is 1:3 then your win/loss ratio needs to be 33.3% to break even

So to simplify things:

If you make 10 trades in a month then 5 need to be winners if you wanna trade 1:1 risk/reward ratio and still break even.

If you make 10 trades in a month then 4 need to be winners if you wanna trade 1:2 risk/reward ratio and still break even.

If you make 10 trades in a month then 3 need to be winners if you wanna trade 1:3 risk/reward ratio and still break even.
Thanks for the great explanation Mr. Obvious, what i ment is you cannot know what the Win/Loss ratio for any system is, you can find out when it was (doing back testing) but that's past, and markets always change. So, to make a long story short, it might be a great ida to have a 1:3 R:R ratio, if you expect more than 33% winnning trades, but since you don't really know the odds of the trade, all the mathematical analysis becomes pretty useless.

1:1 R:R is a good idea, as long as you have ove 50% winning trades
  • #46
  • Edited 5:18am Dec 13, 2007 5:01am | Edited 5:18am
Quoting zek132
... 1:1 R:R is a good idea, as long as you have ove 50% winning trades
Why a sytem cannot make 30% winning trades with P&L > 2.34 if your system make 50% ??
it is strange

you've got simple formula.

w/(1-w) * P&L > 1

where
P&L = 1/RR
w = Average_win_ratio
  • #47
  • Dec 13, 2007 5:05am Dec 13, 2007 5:05am
Quoting Prof
Why a sytem cannot make 30% winning trades with rr< 2.33 if your system make 50% ??
it is strange
Since the market is more or less random, having a stop loss = tp makes your chances of winning about 50%.. i'm not saying that it's easier or harder, what i'm saying is that having a R:R ratio without W/L ratio or trade expectancy is pointless.
  • #48
  • Dec 13, 2007 5:39am Dec 13, 2007 5:39am
Quoting zek132
Since the market is more or less random...
Yep! Condition of Prifitable Trading
w/(1-w) > RR

where w = Average_win_ratio
  • #49
  • Edited 3:30pm Dec 13, 2007 6:34am | Edited 3:30pm
Quoting zek132
Thanks for the great explanation Mr. Obvious, what i ment is you cannot know what the Win/Loss ratio for any system is, you can find out when it was (doing back testing) but that's past, and markets always change. So, to make a long story short, it might be a great ida to have a 1:3 R:R ratio, if you expect more than 33% winnning trades, but since you don't really know the odds of the trade, all the mathematical analysis becomes pretty useless.

1:1 R:R is a good idea, as long as you have ove 50% winning trades

lol..ok we seem to talk different languages here.

If you use 1:1 or 1:2 or 1:3 risk/reward ratio trades and if you make 10 trades and you look back at those 10 trades then you can calculate your win/loss ratio

if you have 1 winner and 9 loosers then your winning % is 10%
if you have 2 winners and 8 loosers then your winning % is 20%
if you have 3 winners and 7 loosers then your winning % is 30%
if you have 4 winners and 6 loosers then your winning % is 40%
if you have 5 winners and 5 loosers then your winning % is 50%
if you have 6 winners and 4 loosers then your winning % is 60%
if you have 7 winners and 3 loosers then your winning % is 70%
if you have 8 winners and 2 loosers then your winning % is 80%
if you have 9 winners and 1 loosers then your winning % is 90%
if you have 10 winners and 0 loosers then your winning % is 100%

I dont know what is so difficult about that? Maybe i am simplifying it too much. Maybe because i am not into backtesting and all that detail. I only take my trades for a day and decide how well my strategie has worked. Anything is in the past is exactley that....

Currently i am only trading engulfments...so i agree there is not much to backtest...
  • #50
  • Dec 13, 2007 7:19am Dec 13, 2007 7:19am
Quoting Time
if you have 9 winners and 1 loosers then your winning % is 90%
if you have 10 winners and 0 loosers then your winning % is 100%
scalper could win 10 trade and each bring him 0 pips.

if we are not scalpers

9 winners per $100.00 = $900.00
1 loss of $1000.00
______________________________
- $100.00
  • #51
  • Dec 13, 2007 3:28pm Dec 13, 2007 3:28pm
Quoting Prof
scalper could win 10 trade and each bring him 0 pips.

if we are not scalpers

9 winners per $100.00 = $900.00
1 loss of $1000.00
______________________________
- $100.00

Ok i am not sure what to make of this post.......

But i guess what you are trying to say here is that this example is of bad risk/reward trading. And not necessarily have anything to do with winning %
  • #52
  • Dec 14, 2007 6:14am Dec 14, 2007 6:14am
Quoting Time
Ok i am not sure what to make of this post.......

But i guess what you are trying to say here is that this example is of bad risk/reward trading. And not necessarily have anything to do with winning %
I have not said that that example is of bad risk/reward trading. I just say that one had better using winning % together with risk/reward

There is no sence to use the winning % & risk/reward separately
  • #53
  • Dec 15, 2007 5:02pm Dec 15, 2007 5:02pm
Quoting Tiki Trader
One should not be taking trades for no reason... hopefully trades are thought out in advance.... you have a reason for entering the trade, an idea of where it will go, and when to get out... you put your stop at the point where you know you are not willing to go any further with the trade or that at that point, things have changed and the trade idea is no longer valid... if you have put your stop in the right place and used correct MM, then you walk away from the computer with no worries... you either win, or you may lose and come back to trade another idea later...

Things happen and the direction of the market can change at anytime... it is not always that you made a mistake in your thinking about where things were going...

It is always good to try and find a point to get into a trade where you have to risk as little as possible... right now an example of this might be USD/JPY... there is a very good possibility that price would go down before getting 112... yet we are very close to 112 already... so I have gone short and my idea of where price would go is all the way to 109... NOW, it may happen that price continues up... BUT I will have my stop right above 112 and if I lose, I will only lose 40 pips or so, while I stand to gain a lot.... I also did not put so much on the trade, that it will hurt if I lose....

Technically and without taking risk/reward ratios into consideration, I would love to go long this chart... Sorry to contradict your opinion, but this is what I can see from the chart....Remember that everytime the market gives you an excellent risk/reward trade opportunity, it is actually giving you a lower success probability...


Thanks,

Nader
  • #54
  • Dec 16, 2007 2:46pm Dec 16, 2007 2:46pm
And you were correct... and I also changed my mind shortly there after... going long... my first target of 113.00 has already been hit... now lets see if we can make the second target of 114.40... although everyone is telling me no... !!! And as far risk reward ratio, by the time I took the trade, my RR was well better than 1:1 !!!

Check here....

http://www.forexfactory.com/showpost...5&postcount=18

Quoting narafa
Technically and without taking risk/reward ratios into consideration, I would love to go long this chart... Sorry to contradict your opinion, but this is what I can see from the chart....Remember that everytime the market gives you an excellent risk/reward trade opportunity, it is actually giving you a lower success probability...


Thanks,

Nader
  • #55
  • Dec 18, 2007 6:36am Dec 18, 2007 6:36am
Interesting discussion.

IMHO RR in itself is ultimately meaningless. Here's my logic:

Consider a hypothetical simulation where price has an exactly 50/50 chance of rising or falling. The trader who sets his TP at 20 and SL at 10 will, on average, have his SL hit twice as often as his TP, hence his overall expectancy (RR x winrate) = 2 x 0.5 = 1. The trader who sets his TP at 40 and SL at 10, will have his SL hit four times as often, i.e. his expectancy = 4 x 0.25 = 1. No matter what values you set for TP and SL, expectancy will always be 1. The reason for this is that, in this simulation, price has a 50/50 probability of rising or falling. Neither RR nor win rate has any bearing on expectancy.

Now consider a simulation where price has a 75% chance of rising, and 25% chance of falling. The first trader (assuming he was able to forecast this correctly, and went long) will now have his TP hit more often than his SL, because the probability of a rise is 3:1 and his RR is only 2:1. We can do the same math for the second trader. In all cases the expectancy works out at 3, no matter where the TPs and SLs are set. If either trader had gone short, his expectancy is 0.333, i.e. the reciprocal. Again, it's the probabilities related to the bias in the price movement – and one's ability to somehow forecast them correctly – that determines the expectancy, not the RR.

The above relates to entry. What about exit? Again, it's about being able to gauge where the probability of a reversal > 50%, that will improve the potential overall profitability of the trade. For example, it might be possible to show that price has a higher probability than "normal" of reversing at prior S/R. If that can be shown to be true, then exiting (partially or fully) at S/R will, on balance, improve bottom line.

Setting arbitrary TP and SL points – at "X" number of pips – doesn't deliver an edge, unless you have statistical evidence that price is more likely to reverse at "X" than normal. You also have to somehow prove that back-testing results are not merely the result of curve fitting. As a previous pointed out, large samples offer greater statistical confidence: assuming that the oldest data is equally as "valid", in the context of the immediate future, as the most recent (i.e. no obsolescence).

To summarize:
a) If, or whenever, price movement is random (50/50), expectancy is always 1, no matter what system is being used.
b) Conversely, using random entries and exits, expectancy is always 1, no matter how predictably price moves.

Hence, to have an edge, you have to somehow identify occurrences of price behavior where there is a greater probability of a move in one direction than the other, and devise non-random entry and exit rules to exploit it. How do you establish this edge? Trends, S/R, patterns, indicators, fundamentals, market knowledge, whatever are your proven tools of choice. If you do have an edge that's capable of profiting indefinitely, then you have achieved positive expectancy in your forecasting.

Of course none of the above takes costs into consideration: if you have no demonstrable edge, then – everything else being equal – costs will ensure that your account balance bleeds to zero at a rate that's in direct proportion to your position sizes.

What about "cutting losses short and letting profits run"? It will work profitably to whatever extent prices "trend", i.e. the probability of price continuing to move in the same direction exceeds the probability of reversal. In a 50/50 situation, it makes no difference (as explained in the second paragraph). If price moves far enough to overcome costs, and you exit before the profit is eroded by the reversal, you win; otherwise, you lose.

Two further reasons why I believe RR is meaningless:

1. As an earlier poster said, while risk is fixed by setting a stoploss, reward is variable: price could potentially reverse at any point (at or around any prior S/R, Fibo, floor trader pivot; or none of these).

2. RR is dynamic. It is always changing, once the trade has been entered: by every price movement, and/or whenever the trader moves his SL or TP.

OK, that's my take on the underlying mathematical theory. Theory being the operative word.

David
  • #56
  • Dec 18, 2007 6:50am Dec 18, 2007 6:50am
Why not create simple risk/reward math?

This is my simple formula:

Reward to Risk Ratio = Winning Trade/Stoploss

Lets say your ratio here is 16. That means if you win 1 trade you must fail 16 trades to get a breakeven.

simple...
Trainee at UBS Investment Banking
  • #57
  • Dec 18, 2007 7:00am Dec 18, 2007 7:00am
Quoting hanover

Two further reasons why I believe RR is meaningless:

1. As an earlier poster said, while risk is fixed by setting a stoploss, reward is variable: price could potentially reverse at any point (at or around any prior S/R, Fibo, floor trader pivot; or none of these).

2. RR is dynamic. It is always changing, once the trade has been entered: by every price movement, and/or whenever the trader moves his SL or TP.

OK, that's my take on the underlying mathematical theory. Theory being the operative word.

David
For a discretionary trader is the RR maybe meaningless. I dont know. I am not sure.

I am a mechanical trader that trades with rules. I dont pick a trade if my rules dont let me trade for example.
That is why RR for me very important. RM and MM is a part (80%) of my trading system. It works very well.
Trainee at UBS Investment Banking
  • #58
  • Dec 18, 2007 11:11am Dec 18, 2007 11:11am
As I see it, after years and years of testing and tryings, the real fallacy of the RR concept lies in the fact that it dictates that you MUST have a take profit for your trades. While I thing that it is mandatory to define the stop loss (mental, physical or whatever) before you even start a trade, having a fixed take profit is not good for me. One of my few trading rules I never break is:
NEVER, NEVER, NEVER close a position when the market is going in my favor.
On the other side, I also like not to lose money at least on trades that started well, so I do move my stop losses to a little more than break even when price moves a reasonable amount of pips in my favor (30 - 60 depending on the system in use). All of this makes RR totally non-sense for me.
.Ubeee.
  • #59
  • Dec 18, 2007 3:41pm Dec 18, 2007 3:41pm
Quoting SABTrader
When i started trading i learnt about R/R from many books. At first it makes such perfect sense. And then you try trading for real and you soon learn that it's only a small part of the whole.

No one will ever find a profitable forex system that has a target of 30 pips and a stop loss of 10 pips. Sure you only need to win 1 out of 3 trades to breakeven but there is noise which is random which is very likely to hit your stop and after spreads and commisions your account balance will go down.

If anyone can prove me wrong i would love to here from you.

I believe it's easier to make money with a 30 pip stop and a 10 pip target with the way forex prices usually move. Even when prices break S/R lines and the books say it will move now, it retraces. Often the second break is the one that continues - explain that to me.

IMHO the first thing to decide on is your trading time frame and then build a system to match. Only with trades lasting 2 weeks or more are R:R > 2
appropriate.
yes i agree. It would be prett stupid to aim for 30 pips while risking 10 pips. There is a better way to trade 1:3 R/R. Risk 50 pips in order to make 150 pips. Or risk 100 pips in order to make 300 pips. Obviously the more pips you risk the bigger the time frame on which you would trade since most currencies dont move 300 pips a day.
  • #60
  • Dec 18, 2007 3:50pm Dec 18, 2007 3:50pm
Quoting ubeee
As I see it, after years and years of testing and tryings, the real fallacy of the RR concept lies in the fact that it dictates that you MUST have a take profit for your trades. While I thing that it is mandatory to define the stop loss (mental, physical or whatever) before you even start a trade, having a fixed take profit is not good for me. One of my few trading rules I never break is:
NEVER, NEVER, NEVER close a position when the market is going in my favor.
On the other side, I also like not to lose money at least on trades that started well, so I do move my stop losses to a little more than break even when price moves a reasonable amount of pips in my favor (30 - 60 depending on the system in use). All of this makes RR totally non-sense for me.
Interesting comment you made. But i have a question. If you never, never, never close a position when the market goes in your favor, then do you just leave the trade to open until the market reverses and hit your stop loss? Or do you wait until your winning trade turns into a loosing trade and then bail out? I assume you dont do that. So obviously you do take profit at same stage otherwise you will never be profitable in forex

Surely you must take profit at same stage because NOBODY knows when the trend is over or a correction is about to start?
0 traders viewing now