The Fallacy of 1 Risk : 2 Reward

Beginner Questions
1

I wrote about the topic of risk and reward recently on my blog and received a lot of interesting commentary. So I thought I would bring the discussion this way and see what people had to say about it.

Recently i’ve started trading an additional method to my Fib Retracement Method posted in the Swing Trading forum on this site. In that thread I talk about appropriate risk : reward. The idea of risking 1 to get 2 being the right (and only way) to make money in a market. However, that idea makes a number of assumptions about trade setups, and trading in general that do not always equate to profitability.

So I decided to delve into the issue a bit more with some backtesting of my live trades taken over the past couple weeks ON MY NON PULLBACK METHOD TRADES. The results were staggering. Over 22 trades using my tighter initial stop placement ended with a net -2.04 pips and a 52% win rate. By simply changing my initial stop to the previous swing high/low my win rate went to 71% on the SAME TRADES and overall profitability increased from a -2 to a +330. I can’t even begin to tell you what a difference that is.

Heres the updated results from my REAL TIME LIVE TRADES with the only difference being stop placement.

Basically my findings supported what I had long suspected. For those of us who don’t trade pullbacks into a trend it is not only improbable, but unprofitable to try and achieve a 1 risk, 2 reward scenario. I had been placing stops at points in a trade that would allow me to keep this “holy grail achievement” that all traders are supposed to use to be successful right? The reality of the situation is this: The 1 risk, 2 reward is a falsely coveted rule of trading for a lot of people. We try and achieve that risk situation because we believe (and are told) it will grow our accounts faster and with less risk. The reality of the situation is the exact opposite. By taking on a bit more initial risk I can stay in my trades regardless of the whipsaw after i’m in them.

In fact my backtesting concluded that I was only flat out wrong about a move where I lost the entire initial stop less than 17% of the time. Those kind of odds make this kind of risk strategy well worth it in my opinion. It would be one thing if I was hitting 40% win rates, but with less than a 1 in 10 chance of taking a max loss on a trade the larger initial risk equals more profit, more of the time.

Pullback traders are able to automatically place stops at the low and by trying to call tops and bottoms to pullbacks can achieve low risk, high reward setups BUT sacrifice a lower win rate as a result. However, we don’t all trade like that, and thus, we need to accept higher initial risk values as long as we have a sound and scalable method that performs higher win rates to offset the higher initial risk.

What i’m suggesting is this:

A relative increase in risk can be balanced IF it helps you achieve a higher win rate AND you aren’t doing anything to preposterous with your risk, i.e. risking 150 pips to make 2. My risk currently sits pretty close to 1:1, but by doing this i’m making a number of realistic assumptions about the market.

  1. I think my setup is valid and should win over 70% of the time.
  2. I don’t expect the market to go on to the moon or drop to 0 during my trade - thus my targets are based off of getting the most profit, the most amount of time.

Take a look at the TRO Never Lose Again thread… Now I don’t like trading like that and I don’t like TRO, however 98% of those people are trading INVERSE risk : reward scenarios… but almost all of them that grasp the concept of what they are doing are making money consistently.

If the idea of risking 1 to get 2 were the only and correct way to do this then they wouldn’t be having success would they?

I’m not suggesting that this is an end all be all. This is simply to start some discussion, hopefully get you thinking PROACTIVELY about your own trading style and possibly revisiting some of the assumptions your trades make about the market and potential outcomes based on your stops and management.

There are 1 billion ways to enter a trade. The profit lies in the management and the stop placement. Which is why we all need to understand that there isn’t simply one way to look at a situation like risk and reward.

Being discussion now! :stuck_out_tongue:

2

Hi daedalus,

Very interesting topic. Is that what I/we’ve been doing…inverse risk…LOL! I think I’d have to agree although I had thought of it a little differently.

I had thought for example that since I had a mental s/l of 10-20 which sometimes went more that 20 when it wiggled back & forth for a time before deciding which way to go… for a profit of 10 or 20 per trade doesn’t sound good. However if a loss of that much only got accepted once in 5 trades, then the OVERALL trades RR would have been 20 for a profit of 50…or 1 : 2.5.

But if what you’re saying about the swing H/L’s is what’s happening, then I’ll pay a little more attention to that because I do want to use some kind of protective s/l on my live account since I had mostly only been using a mental one on demo.

I wonder who came up with that RR rule originally…the brokers!..:smiley:

Good stuff! …and thanks :slight_smile:

3

Could you give examples of the percentage from bid price of the two instances in your chart,
for example if the tight stop is 1% away and the wider stop 3%.

By this, do you mean you don’t trade when the price is moving the direction of a trend?

my backtesting concluded that I was only flat out wrong about a move where I lost the entire initial stop less than 17% of the time.

Thus where it hits the stop less than 1 time out of 6, and moves upward 5 times out of 6.

4

Great thread D, particularly timely because I’ve just been analysing my most recent trades I did for exactly the same reason.

When I first started forex, I backtested loads of strategies and found that the SL is a bell curve (on any given strategy, didn’t matter what it was). Starting from far far away, the SL doesn’t save you money, then slowly rises as you close it in as it starts saving you money, till you get to a peak and there’s a sharp drop-off as it gets even closer where it starts hurting your profits. Which led me to the rule I still stick with now - if in doubt, widen it out (I know, but it rhymes, alright?!). But then, as you rightly say that skews your R:R.

Fast forward to now, and the main conclusions I’ve made are similar to yours:
a) If you’re trading reversals, you have probably identified a fairly decent R:R. However, if price doesn’t reverse where you’ve expected to, it’s actually more profitable to not close the trade at 1R, but consider it a losing trade, widen out your stop and look for the best place to reverse as close to break even as possible. Reason being, price is already overextended if you’ve done your analysis right, and should offer you an opportunity to get out at a better price
b) If you’re trading breakouts, the R:R thing doesn’t make sense, because you’re looking to capitalise on momentum. If you’re right about the momentum, you should be able to make binary bets (SLs same as TPs) and do better than random coin tosses. Nick B’s method is based on 1:1 all the way through, and though I don’t trade it, I can recognise it’s a corker.
c) If you’re trading inverse risk, you need a far higher win rate to sustain it.

In short, the golden rule about R:R isn’t golden at all. It’s so utterly dependant on the probability of success (expectancy) of your strategy that everyone should question it in relation to their own trading.

5

Great point triphop,

In short, the golden rule about R:R isn’t golden at all. It’s so utterly dependant on the probability of success (expectancy) of your strategy that everyone should question it in relation to their own trading.

The maxim “Let your profits run & cut your losses short” is also
brought to mind.

We can also classify each trading stategy with a different risk
attached to it, ie countertrend trading is high risk, therefore
should we still be looking at 1:2 ratio, when trading like this?

Found this on my travels,
Determining Risk And The Risk Pyramid

6

By this, do you mean you don’t trade when the price is moving the direction of a trend?

No. Think of it like, trading a breakout, rather than a pullback. Its two different setups at two different extremes of a price swing, but if we are still placing the stop at the swing low/high of the swing then one trade will have a low risk (the pullback) as you try and constantly catch the top or bottom and a lower win rate but are getting in the trade closer to that stop and thus having a lower risk, higher reward setup.

Now take the breakout. You’ll be getting into the trade at the top of the swing and it will have a high risk value if the same stop is placed at the previous low, but if the expectancy of success is higher, it should offset it.

Thus where it hits the stop less than 1 time out of 6, and moves upward 5 times out of 6.

Yes.

Here is a chart example of what i’m talking about. Same swing, two different entry points, the same initial stop. One is lower risk, one is higher risk, one has lower win rates, the other higher win rates… BOTH WORK.

Thanks for the great discussion guys! Great to get feedback and input from you all!

7

Good discussion…

I tend to agree R:R is somewhat of a fallacy. I use the same stop all the time, every time and it basically works out to intially a 1:1 ratio, but according to my strict trade management rules, moving the stop at a particular point, taking partial profits at a particular point as well.

I would say I always enter on what you are defining a pullback entry.

By the time the trade got to the breakout point (you dont know its a break-out as yet), in my opinion I would have missed the trade.

Depending on what kind of trader your are, widening your stop is more guessing as opposed to having a sound and tested trading technique.

Perhaps this is where knowing the currency pair(s) you trade very well as you can learn where to put your stops from experience based on teh general behaviour of the currency pair.

I note you are trading 8-9 different pairs, I’m not sure that you can apply a standard trading method across a wide number of pairs.

Cheers

8

Hey daedalus,

Nice post and nice work.

At some point in my ‘career’ I also ‘toyed around’ with the different ‘golden rule’ ‘risk / reward’ ratios and they almost always ended up costing money. You THINK you’re ‘playing it safe’ but in reality: those supposedly ‘safe losses’ add up eventually.

I’m not going to go through my ‘usual Wilder routine’ here but I thought you may be interested to know that with Wilder’s ‘stuff’ the stop loss (actually they’re always a stop and reverse not just a stop but nevertheless) is ALWAYS the previous high swing point or previous low swing point and, with some of the trading systems, there ARE no initial stops at all i.e. in some cases your stops (stops and reverses) are only placed AFTER THE CLOSE and / or when a signal is given AFTER THE CLOSE (on forex that’s after the close of the current bar).

Also: in one of the systems (oddly enough the only one that does INDEED used stops and ALSO oddly enough the LEAST profitable of them all BECAUSE of the stops) the stops eqaute to roughly a 3:1 risk / reward ratio. This particular system ‘falls down’, as I mentioned, BECAUSE of the stops i.e. you make five excellent trades and be stopped out twice and you’re down again BECAUSE sometimes the stop is JUST near a swing point or a pivot and, particularly with pivots, depending on which pivot it is, you’re normally going to get a bounce there anyway so you get stopped out and the trade goes back in your direction to WAY past your TP.

Personally: I also make ‘liberal use’ of a percentage of the ATR e.g. 10% of the value of ATR(14) above the high or below the low of the bar where a / the stop loss order or stop and reverse prder is to be placed and this works well for me (on the daily charts).

I know of some people who use multipliers of ATR as well but using that method (I believe) is going from the sublime to the ridiculous.

The SHORT version: I agee with your findings TOTALLY!!!

Regards,

Dale. (forexbrokersonline.net).

9

"…Depending on what kind of trader your are, widening your stop is more guessing as opposed to having a sound and tested trading technique. "

In general yes Cdawg - but it depends on how you approach it.
I trade mostly reversal trades (so in Daed’s pic, it’s the first time it hits that line, one step before the pullback and the breakout), and then sometimes pullbacks in a big trend.
I wouldn’t expect it go more than 20-30 pips say on EURUSD beyond entry point before I classify it as a losing trade. If it does go past that point however, I’m looking to rescue it and close as close to b/e as possible. The wiggle room is usually twice that of where the SL would be if the trade had worked out (40-60 pips), but this is decided [B][I]before [/I][/B]entry in the trade. So the widening of the SL is more mental than practical, if that makes sense.
If the trade works out of course, SL is tightened to the original level.

10

I think you are analyzing a very worthwhile topic, however a sample size of 20-30 trades is not at all significant. I would submit that you will need to analyze 200-300 trades before you start to even begin drawing conclusions.

11

I’ve done MANY X’s that amount in backtesting on every market imaginable from tbills, all the eminis, exotics, crude, gold, silver, grains… and some tradeable equities AAPL, RIMM, etc…

I just posted up a sample from my live trades I had taken in the FX markets in the past month because thats the content we are focused on in this forum. :slight_smile:

And for that matter, I’ve found the EXACT same results, sometimes moreso in other markets. The wider initials do wonder for results.

But I appreciate what you’re saying and you’re correct! It needs to hold true over a very large spectrum of trades, and over a long period of time. I went back 5 years on each market to make sure the same was true regardless of market volatility.

Cheers!

12

Fair enough, objection withdrawn. In your initial post you stated that you backtested a couple weeks trading (about 30 trades) and the results were “staggering.” From reading your last post I can see that you are well aware 2 weeks data wouldn’t be enough evidence to make such a proclamation. I appreciate you bringing your research to the forum or everyone’s consideration.

Wax

13

With regard to the actual topic, I have been using ATR thus far successfully in a strategy I have been backtesting and forwardtesting. The results aren’t sufficient to get too excited yet, but using ATR to determine stop levels works for me conceptually. I trying to implement the Turtle strategy, but adapted for forex, and shorter timeframes.

14

I was obsessed with finding a strategy that had a reward:risk of 2:1 and since I have lowered my expectations to trading a minimum RR of 1:1 my results have improved drastically and I am much more confident as a result.

15

The whole point of trading is to make money. How many times have you been in a trade with a RR of 2:1 where price has gotten to half way of your target only to reverse and you’re either left with nothing because you moved your stop to break even or worse, price hit your stop?! I’ve realised in my time that I am good at spotting trends but that I shouldn’t ask price to do too much to hit a profit target or go somewhere that it hasn’t been before. For me time is also just as important as price. Get in, make your money, and get out.

16

Having experimented and backtested with various R:R ratios, all I have to say is, the smaller your reward, the more sensitive you are to slippage, negative swap and bad spells. As a general rule, I’d recommend a reward of at least 0.5 relative to risk. The actual optimal R:R ratio will depend on your strategy.

17

Contrary to this thread I use 2:1 reward:risk with no problem, just saying.

18

I cannot say thank you enough for this interesting thread… Someone else, a few months ago, posted a whole series of statistics that showed how they had observed, over hundreds of trades, how going from 1:1 to 1: 2 Risk-Reward ratio did not improve results, but when you went to a 1:3 Risk-Reward ratio then things really started improving considerably.

I wish I could find the thread, because it had a wealth of stats showing a serious study…

19

You mean this one… 301 Moved Permanently

20

Thank you, SweetPip, yes, that was it!!