The 2/1 risk/reward MYTH... | Page 2 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

1% Money Management is a myth, fib is a myth and so on.. 45 replies

What's your TP? Fixed reward vs dynamic reward 5 replies

Low Risk, High Reward Trading Experiment 3,318 replies

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

Risk Reward Ratio and Account Risk 30 replies

Attachments: The 2/1 risk/reward MYTH...
Exit Attachments

The 2/1 risk/reward MYTH...

  • #21
  • Feb 28, 2007 7:25pm Feb 28, 2007 7:25pm
It has already been shown that the only 'edge' necessary to beat this game is money-management - a random entry will work:

" Tom Basso designed a simple, random-entry trading system … We determined the volatility of the market by a 10-day exponential moving average of the average true range. Our initial stop was three times that volatility reading. Once entry occurred by a coin flip, the same three-times-volatility stop was trailed from the close. However, the stop could only move in our favor. Thus, the stop moved closer whenever the markets moved in our favor or whenever volatility shrank. We also used a 1% risk model for our position-sizing system. …
We ran it on 10 markets. And it was always, in each market, either long or short depending upon a coin flip. … It made money 100% of the time when a simple 1% risk money management system was added. … The system had a (trade success) reliability of 38%, which is about average for a trend-following system."

From 'Trade Your Way to Financial Freedom', Van K. Tharp

This is only one example.

Daemien
  • #22
  • Edited 8:42pm Feb 28, 2007 8:29pm | Edited 8:42pm
Quoting Daemien
It has already been shown that the only 'edge' necessary to beat this game is money-management - a random entry will work
" Tom Basso designed a simple, random-entry trading system … We determined the volatility of the market by a 10-day exponential moving average of the average true range. Our initial stop was three times that volatility reading.
Stop right there and re-read the last sentence Daemien: "Our initial stop was three times that volatility reading..."

I am sorry but this is NOT (NOT!!) a random-entry point, because you are trading in sync with the market. You observe the market and you adjust your stop/trailing stop accordingly so again this is not a random trade anymore, far from it. It is like counting the cards at a Blackjack table (like in the Tom Cruise movie "Rain Man") and betting accordingly, your bets are not random anymore.

A purely random entry/exit system will enter a trade randomly and exit the trade after a PRE-determined stop or limit target has been reached. You cannot modify your stop/trailing stop as you trade and call that a random entry system.

In fact this so-called random system you are describing is simply a trend following system in disguise, even though you are using a coin to enter the trades because, once again, your exit point is not random. Keep in mind that both your entry AND exit points must be random.

Please tell that to Tom Basso and Van K. Tharp http://forexfactory.com/images/icons/icon12.gif
  • #23
  • Feb 28, 2007 9:42pm Feb 28, 2007 9:42pm
The point was about the entry, not about trade-management once a position has been taken. And in this case, seeing as how the stop was a trailing one, it would have been as random (or not) as the markets movement. And of course it is a trend following system - they mention that in one of the last sentences.



D.

Quoting FXTerminator
Stop right there and re-read the last sentence Daemien: "Our initial stop was three times that volatility reading..."

I am sorry but this is NOT (NOT!!) a random-entry point, because you are trading in sync with the market. You observe the market and you adjust your stop/trailing stop accordingly so again this is not a random trade anymore, far from it. It is like counting the cards at a Blackjack table (like in the Tom Cruise movie "Rain Man") and betting accordingly, your bets are not random anymore.

A purely random entry/exit system will enter a trade randomly and exit the trade after a PRE-determined stop or limit target has been reached. You cannot modify your stop/trailing stop as you trade and call that a random entry system.

In fact this so-called random system you are describing is simply a trend following system in disguise, even though you are using a coin to enter the trades because, once again, your exit point is not random. Keep in mind that both your entry AND exit points must be random.

Please tell that to Tom Basso and Van K. Tharp http://forexfactory.com/images/icons/icon12.gif
  • #24
  • Feb 28, 2007 10:39pm Feb 28, 2007 10:39pm
Quote
The point was about the entry, not about trade-management once a position has been taken.
The point was about pure random entry/exit points, please see my previous posts. Your entry is truly random but your exit is not, so your example is not a valid one, from that point of view. The idea of making money in the forex just by flipping a coin is, at best, a childish utopia.

Like I said before, just flip a coin, Head buy (or Tail sell) the euro each day with a profit target of 60 pips and a stoploss set a 20 pips and see if you can make ANY money with that truly random forex system, after 1000 trades!


Quote
And in this case, seeing as how the stop was a trailing one, it would have been as random (or not) as the markets movement.
You forget that the market itself is not random, otherwise we wouldn't be able to make one red cent from it, no matter what system you are using !!
  • #25
  • Feb 28, 2007 10:48pm Feb 28, 2007 10:48pm
Quoting hilmy83
dismissing the 1/2 risk reward ratio is definitely a flawed argument. Trading in forex is very different than playing roulette or playing lotto. The chances of winning is very small with gambling.

But in the financial markets, even though nobody can predict the market, you have enough edge from fundamental and technical analysis to get you on the winning side given that you follow a certain method with consistency. And the fact that each trade is a uniqe opportunity, you haveto make sure you will be statistically ahead by getting the most rewards with your initial risk on each trade.

Because of this I believe that trades that only uses reward to risk of less than 2, could potentialy lead you to ruin or breakeven in the long term. You have to have a system that is consistently 60-70% winning percentage to take advantae of 1 to 1 risk to reward ratio. And from what I read in Van Tharp and other books, most successful traders make money from less than half of their trades (<50%).
I couldn't said it better. Here is the Probability Ruin Matrix. It shows the chances of ruin your account based on R/R and your win to loss ratio. The matrix says it all, see it for yourself :
Attached Image
  • #26
  • Feb 28, 2007 11:03pm Feb 28, 2007 11:03pm
Quoting JimboFX
I couldn't said it better. Here is the Probability Ruin Matrix. It shows the chances of ruin your account based on R/R and your win to loss ratio. The matrix says it all, see it for yourself :
Yes thank you, we are all (I hope) well aware of the relation between RR, winning percentage and probability of ruin. This is basic probability math stuff.

But I will go even further: REGARDLESS of your RR ratio and winning percentage, you can never eliminate the possibility of ruining your forex account period. The risk zero simply does not exist.
  • #27
  • Edited 12:14am Mar 1, 2007 12:12am | Edited 12:14am
I'm with FXTerminator. I hope I don't misunderstand his points though.

I'm not very familiar with all this matrices. But this is what I understand about winning the market.

We must make more than we lose ( how obvious). Lose 100 pips, Win 900 pips. Net gain 800 pips.

It doesn't matter how we do this.
Either Method
A - Lose 100 pips after being wrong 50 times, but be right 5 times and win 900.
B - Lose 100 pips after being wrong 5 times, but be right 50 times and win 900.
(These example looks funny but you get the idea)

To me, R/R calculation is too hypothetical. How can anyone estimates Rewards ? You can set the risk with appropriate Stop loss, but rewards ? No matter how good the math is, if we use numbers from dubious source, the calculation will be useless.

There's a solution to our inability to calculate rewards. The so-called
-cut your losses short (reasonable stop loss, move SL to break-even when profiting, try to reduce Risk)
-ride on your winners as long as possible. (you try to increase Rewards as much as possible)

R/R idea works intuitively, not quantitatively because there's no good number to define Rewards.

Any strategy will work as long as it does the above.
  • #28
  • Mar 1, 2007 4:34pm Mar 1, 2007 4:34pm
Luq,

I think I get what you're saying here, but I look at it differently.

You need to understand your RR when it comes to your personal expectancy. It's not enough to just say, here's my TP and my SL so there's my RR and I'm good to go. No, that's not enough...you have to say, based on my personal history (or system's history), I believe my expectancy is X%, therefore, I need an RR of Y in order to be profitable.

This is where the bulk of the confusion is.

You really do have to understand your RR in terms of your expactancy in order to know if it's even worth your effort.

I think what's becoming readily apparent is that the bulk of the traders on these forums pull the trigger without really understanding their RR and required expectancy. This would explain why the percentage of profitable traders is so low. My guess is that a lot of traders (including some that are posting to this thread) are just counting pips without truly understanding the mechanics behind their success (or failure).

Let's take an example...

The long-term price trend of the EUR/USD is up (let's say the daily or 4H chart). We reached a top about a month ago around 1.3600. We bottomed out around 1.2900 and are returning to an uptrend. Now, there is some resistance around 1.3200 to 1.3250. What do you do if you were to want to trade the break of 1.3250? Your upside has a max of 350 pips while your downside is also around 350 pips (major top and major bottom). You now have a long-term RR of 1. Now, based on historical data, you see that the break of the 50% retracement from a recent low is > 50%...do you take the trade? Of course you do, because you now know your expectancy versus your RR. Note that this is just an example and isn't backed up by any real facts. Furthermore, you can adjust your RR accordingly, but that would also affect your expectancy.

Let's say you wanted to only take 50 pips of the 250 pip move up. However, your confidence level goes up because you believe you have more than enough room on the downside (.20 RR = 50:250 = 1:5). however, now you know what your expectancy HAS to be...you have to be right 5 times out of 6...or better than 83%...now, historically, even though you had a 50% chance of being right to retest the high, you find that making 50 pips before retesting a low occurs only 75% of the time...do you still take the trade?? See the problem? Without being aware expectancy necessary to risk the trade, you see that being profitable at THIS system would be dumb-luck. Now, does that mean the trade itself would be a bad trade? No, of course not, we only analyzed one or 2 aspects of historical trends to determine if the trade was any good.

So, I still think it's important to know your RR, but not necessarily rely on you RR as the end-all-be-all of money management. You absolutely must have a reasonable expectation (through historics or forward testing) of reaching the proper expectancy to match your expected RR. Otherwise, and I seriously mean this, you're just getting lucky (if you're profitable) or you're doing something right without truly understanding why it's right. Overtime, expectancy can and will change. As vulnerabilities are closed, expectancies will decrease and new vulnerabilities will have to be exploited.

This is why you have to constantly be aware of your expectancy AND your RR. Not just one or the other.

Quoting luqmanz
I'm with FXTerminator. I hope I don't misunderstand his points though.

I'm not very familiar with all this matrices. But this is what I understand about winning the market.

We must make more than we lose ( how obvious). Lose 100 pips, Win 900 pips. Net gain 800 pips.

It doesn't matter how we do this.
Either Method
A - Lose 100 pips after being wrong 50 times, but be right 5 times and win 900.
B - Lose 100 pips after being wrong 5 times, but be right 50 times and win 900.
(These example looks funny but you get the idea)

To me, R/R calculation is too hypothetical. How can anyone estimates Rewards ? You can set the risk with appropriate Stop loss, but rewards ? No matter how good the math is, if we use numbers from dubious source, the calculation will be useless.

There's a solution to our inability to calculate rewards. The so-called
-cut your losses short (reasonable stop loss, move SL to break-even when profiting, try to reduce Risk)
-ride on your winners as long as possible. (you try to increase Rewards as much as possible)

R/R idea works intuitively, not quantitatively because there's no good number to define Rewards.

Any strategy will work as long as it does the above.
  • #29
  • Mar 1, 2007 5:36pm Mar 1, 2007 5:36pm
Quoting mrmikal
What FXT is saying IS correct, there IS a myth that having a Risk/Reward ratio of 2/1 is necessary.
Hello mrmikal,

You are of course right on the money my friend.

These days it seems that you cannot open a technical analysis book without reading garbage like this :
"By the way, only take trades with a good risk/reward ratio, at least 2/1 or better, I personnally never trade if the RR ratio is less than that"

What kind of nonsense is that ? Because the trade offers a 2/1 RR it is automatically a "good" trade ??? Since when ???? What if that "good" 3/1 trade has only a 5% chance of success you triple idiot ?

Quote
Having a 2/1 RR is just some arbitrary number that a lot of traders stick by for whatever reason.
That's precisely the 2/1 RR myth I was trying to debunk in that thread.

Quote
That's why TP and SL aren't your true measurement of Risk Reward.
Right again. And yet, many traders still fall for that trap.
  • #30
  • Mar 2, 2007 6:01am Mar 2, 2007 6:01am
Quoting FXTerminator
So in conclusion it does NOT matter if you are trading with a 2/1 or 2000/1 risk/reward, the ONLY WAY to make money in the forex (or futures, stocks, etc...) is to have a STATISTICAL EDGE.

In other words your entry and exit points must BEAT pure random entry/exit points.
in my experince, there are other factors deciding wheter you make money or not...

what do you mean by "STATISTICAL EDGE"

maybe you meant somthing like
The only way to win at futures trading is for you to be larger (have more positions) when you are right and less positions when you are wrong.

http://members.aon.at/tips/moneyMan1.htm


btw..in this context, nobody ever has been talking about random EXIT points....random ENTRY points, yes, but random EXIT points...this did not serve as any usefull reference
  • #31
  • Mar 2, 2007 6:34am Mar 2, 2007 6:34am
Quoting semar
In my opinion when you trade, you don't trade agains your broker(like in a casino),the broker is an intermediator, and the spread is theyr winning(profit you can say). Brokers don;t wan't you to lose money, they wan't to have a big number of transactions, so they can take profit from the spread. Forex traders don't loose money agains brookers(like a casino player in a casino), they loose it against other traders.
I'm wrong or I didn't understand what you were saying?
brokers do take the other side of your trade
  • #32
  • Mar 2, 2007 6:57am Mar 2, 2007 6:57am
If you get an average of n/1 RR Ratio then you need average odds of success to be at least 1/(n+1). That's all that matters, if you satisfy this condition then you will make money in the long term.
  • #33
  • Mar 2, 2007 7:00am Mar 2, 2007 7:00am
Quoting 17121712
If you get an average of n/1 RR Ratio then you need average odds of success to be 1/(n+1). That's all that matters, if you satisfy this condition then you will make money in the long term.
Well that's it isn't it. The entire thread boils down to that simple expression.

LOL... You are right, nothing else matters.
  • #34
  • Mar 2, 2007 7:53am Mar 2, 2007 7:53am
btw, random trading is not such a bad thing..

---------------------------
The extremeness of the outcome in this story seems to support an apparent phenomenon that I’ve observed many times over the years, but for which I have no hard evidence: The majority of people trade worse than a purely random trader would. — William Eckhardt, interviewed by Jack Schwager in The New Market Wizards, Conversations with America’s Top Traders
  • #35
  • Mar 2, 2007 10:08am Mar 2, 2007 10:08am
Quoting peter100
what do you mean by "STATISTICAL EDGE"
Hi Peter,

Frankly, I am very surprised you asked that question, because if your trading system does not give you a statistical edge, even a small one, then the only one who is going to get rich in the long run is your forex (or futures, stock) broker, via the spread ! In other words your entry/exit trading points are just random, giving you NO mathematical advantage so you are only making your broker richer everyday, via the spread.

Ok, here it is.

Let's say you go to Las Vegas and after many, many days of observation you notice that some numbers at a particular roulette table come up more often than others. This is called "clocking" the wheel by the way. The idea is that over time, some roulette wheels do become un-balanced and biased to a certain area of the wheel.

Now, that's your statistical edge right there!

Just bet on the numbers of that specific unbalanced area and you could break the bank every day. Now our entry (bet) is NOT random anymore because we now have a definitive statistical edge that will allow us to make money.

It exactly the same principle with forex trading.

Let's say you decide to buy the euro/usd at $1.3100, your goal is to reach $1.3175 unless you are stopped by a 25 pip loss.
Sure, but why on earth did you choose THAT specific $1.3100 entry point and not $1.3029 or $1.3177 for example?

Simple, because you strongly believe that

1/ $1.3100 is NOT a random entry point. In other words you believe buying at this specific price level will give you an EDGE.

2/ Your 3/1 risk/reward trade (75 pip take profit - 25 pip stop loss) has more than 25% chance of success.

Sure, but why do you believe that?

Because your (hopefully) backtested forex system tells you that by consistently buying and selling in this fashion you will make money in the long run. Your system gives you an edge. Your system beats random entry/exit points.

So, again, trade because you have a fully backtested forex system that gives you a mathematical edge, and NOT simply because the trade gives you an attractive (??) good risk/reward ratio like many newbies do.

Remember, without a definitive mathematical edge, you are doomed to failure.

I hope this clarify the subject.
  • #36
  • Mar 2, 2007 10:20am Mar 2, 2007 10:20am
Quoting 17121712
If you get an average of n/1 RR Ratio then you need average odds of success to be at least 1/(n+1). That's all that matters, if you satisfy this condition then you will make money in the long term.
Absolutely. And just to break even you might add. To make money you need MORE than 1/(n+1).

But that "average odds of success" itself can only be known through (back)testing, so we are back to square one.
  • #37
  • Mar 2, 2007 10:25am Mar 2, 2007 10:25am
This thread literally is arguing with itself.

On one breath, we say that Forex isn't like flipping a coin. You can't say that if you take 10 trades, 5 will be good, and 5 will be bad for a 50/50 ratio. Forex is much more than that and your chances increase as long as you know what you are doing and can read the market since it's not random.

And then, on the other breath,
we're tabling and talking about win/loss ratios. If you have a 1/3 you need 66% of winning trades etc to break even and get into profit.
Are you always going to win 6 trades and lose 3 consistantly? Nope.

It seems to me, outside of hypothetical questions and answers, that we started off talking about how these stats are a myth outside of random games, and then we validate the myth by going back to the same discussion on what risk/reward works.

If what I'm doing nets me 1 win, and 9 losses, the system didn't get me that one win, that was luck and coincidence. You're just doing something wrong, plain and simple.
  • #38
  • Mar 2, 2007 11:10am Mar 2, 2007 11:10am
Tec,

You're probably right on some level. This thread IS trying to define the difference between random entries and what it takes to win.

I think where you may be wrong is where the thread has gone after discussing the initial points. The truth absolutely is, that in your time as a trader, given a certain average RR you absolutely have to have a certain win/loss ratio.

To your example, given that your RR is 1/3, during your time as a trader, you better have won more than 66% of the time in order to have been profitable during that time. There is no dispute on that fact. Whether you lost your first 33 times only to win the last 67 times you traded, or whatever...you absolutely had to have a certain winning percentage given your RR in order to profitable.

So, the absolute facts are in order to profitable during any period of time, you have to have a winning ratio that allows your RR to become profitable and vice versa.

Where there is a gray area is how you determine your winning percentage. The winning percentage can be based on an infinite number of variables, so it's impossible to say with any sort of certainty what your actual winning percentage will be for a certain forward-looking period. You could be lucky and have a higher winning percentage than required or you could get unlucky and get wiped out before you came into a streak that would even out your winning percentage.

Given this information, having an RR alone is not only irrelevant, but not useful for determining whether or not you'll be profitable. In other words, pulling out some arbitrary RR doesn't make you profitable...it's your winning percentage along with your RR that gives you profitability.

While this could be construed as a rather cynical look at trading, it isn't entirely a lost cause. Statistical analysis based on historical research can give you more confidence to pursue a certain strategy. with that said, it doesn't necessarily give you a winning edge. This is where it requires more flexibility and creativity in order to make money in the market.

This is why being married to one system will never really lead you to riches indefinitely...and why some systems work for some people while those same systems fail for others...statistically speaking, some people just got lucky.

What you CAN use to become profitable is the understanding of market dynamics and self-fulfilling tactics...There is a certain nature to the markets because it's based on the decisions of humans. Humans tend to think in and act on patterns. The key is to finding the edge first and exploiting it for as long as you can before that exploit goes away. This gives you that statistical edge that will coincide with your RR that will turn you profitable.

Quoting techno79
This thread literally is arguing with itself.

On one breath, we say that Forex isn't like flipping a coin. You can't say that if you take 10 trades, 5 will be good, and 5 will be bad for a 50/50 ratio. Forex is much more than that and your chances increase as long as you know what you are doing and can read the market since it's not random.

And then, on the other breath,
we're tabling and talking about win/loss ratios. If you have a 1/3 you need 66% of winning trades etc to break even and get into profit.
Are you always going to win 6 trades and lose 3 consistantly? Nope.

It seems to me, outside of hypothetical questions and answers, that we started off talking about how these stats are a myth outside of random games, and then we validate the myth by going back to the same discussion on what risk/reward works.

If what I'm doing nets me 1 win, and 9 losses, the system didn't get me that one win, that was luck and coincidence. You're just doing something wrong, plain and simple.
  • #39
  • Edited 12:18pm Mar 2, 2007 11:40am | Edited 12:18pm
Quoting mrmikal
The truth absolutely is, that in your time as a trader, given a certain average RR you absolutely have to have a certain win/loss ratio.
In theory yes Mrmikal, in real life that should not concern us at all, when your forex system gives you a buy/sell signal, a limit and a stop loss then just place your bet (trade), the value of the win/loss ratio is absolutely of no interest from that point on. If your system tells you to risk 10 pips to make 100 pips on a particular trade or risk 100 pips to make 10 pips, so be it, just take the trade, we don't care about the value of the RR ratio and the winning percentage anymore.

That's why limiting yourself (I am not talking about you mrmikal) to trades with a 2/1 RR ratio or better is a total nonsense.

Why?

Because, again, your FX system has ALREADY established that betting (trading) that way IS profitable in the long run. So whether the trade has a 3/1 RR or 5/6 RR is irrelevant, moneywise.

In fact the ONLY variable that should interest us is the size of the average trade (win AND loss) of your tested system. The average trade is your true mathematical expectancy.
  • #40
  • Mar 2, 2007 10:58pm Mar 2, 2007 10:58pm
The reason the thread is having so many problems is because everyone persues their money in the FOREX in a different way.

There are the mechanical traders who will place a 25 pip stop looking for 50, 75 or even 100 pips with no middle ground. The trader places a stoploss then sets a target k(SP) away from entry, where k is the ratio. Either the stop will be hit or the target. Therefore, this gives a set RR ratio, independant of trades taken on. With this the trader needs to aim for his trade success to be sufficient for him to be profitable.

I feel you can't be as strict when it comes to the market. I trade when I feel an oppertunity has presented itself. I set a stop and exit the position when I feel favorable movements are too risky to wait for. Since I can't guess with any certainty either of the two variables I simply aim as high as possible with both. I then review my performance. If I satisfy the condition I mentioned earlier then great. If I don't then I am doing something wrong and my method needs altering.

I am sure someone has said this already, but, flipping a coin gives you 50/50 chance of going long or short but trade success is defined by "closure of position in profit before price touches stoploss/(closure in loss)". This event is dependant on the movement of the market, the coin flip makes no difference (obviously). Whether an entry was a random entry or was considered carefully over tons of analysis is completely irrelevent. You need to manage your trades properly, understand that attempting to set either of the variables is counter-productive (I think this is FXT's point). You simply need to keep these numbers in your mind.
0 traders viewing now