trading with R:R < 1.0

[Dinos]

That is exactly what I have done. I used to try and aim for +2, +3 or whatever risk:reward but came to the conclusion that this is not always possible on the lower time frames.
I

Have you tried tick charts rather than time frame based charts, if your wanting sub +10 pips moves, with the right size tick chart and set up, on EU as I trade, I get 5-7 trades a day, with good consistency, R&R isn't a factor, % accuracy is

 

[DashRiprock]

Have you tried tick charts rather than time frame based charts, if you wanting sub +10 pips moves, with the right size tick chart and set up, on EU as I trade, I get 5-7 trades a day, with good consistency, R&R isn't a factor, % acc:cheesy:uracy is

WOW 2003?!

Dino are you the longest member here? (nobody joke about willies please)???

OK so my original point was that, ceteris paribus, r:r < 1 makes more sense because of all the biases we have as humans. really, I think it does.

but now we are talking about, from very short term ----> very long term, how (i think at least) markets go from mean reversion ------> trending.

if everyone else wants to talk about that then cool beans

 

[jd1888]

Have you tried tick charts rather than time frame based charts, if your wanting sub +10 pips moves, with the right size tick chart and set up, on EU as I trade, I get 5-7 trades a day, with good consistency, R&R isn't a factor, % accuracy is

I haven't tried tick charts as I'm using MT4 and haven't found an EA, script or whatever that can give me tick charts. By EU do you mean EUR/USD?

 

[arabianights]

WOW 2003?!

Dino are you the longest member here? (nobody joke about willies please)???

OK so my original point was that, ceteris paribus, r:r < 1 makes more sense because of all the biases we have as humans. really, I think it does.

but now we are talking about, from very short term ----> very long term, how (i think at least) markets go from mean reversion ------> trending.

if everyone else wants to talk about that then cool beans

Ok I'm getting your point now... better to take large losses and small profits cause it's better for you mentally?

Well... probably... but IF YOU WANT MENTAL HEALTH DO NOT DO THIS GAME

Even with your system you're still seeing a lot offside waiting for it come back. Go and become a civil servant or something. My honest advice to 99% of people.

 

[DashRiprock]

Ok I'm getting your point now... better to take large losses and small profits cause it's better for you mentally?

Well... probably... but IF YOU WANT MENTAL HEALTH DO NOT DO THIS GAME

Even with your system you're still seeing a lot offside waiting for it come back. Go and become a civil servant or something. My honest advice to 99% of people.

well, yeah. with an edge, small profits vs. large losses gets around some tricky problems.

 

[arabianights]

Only if you **** off and don't watch the trade!

But if you do that you're basically doomed. To trade that way you HAVE to monitor the position.

So just as much pain really.

 

[meanreversion]

Dash, interesting post.

I've done extensive research into all the "basic" systems, whether they're trend following, mean reverting or variants thereof.

Like you (and everyone else) I would like to have a high win rate, ergo why not shoot for 1:2 rather than the oft quoted 2:1.

Here's my rough and ready conclusion - if you're trading breakouts or trends, you must have a 2:1 ratio or greater, because that's the very essence of the system.

However, if you're looking for pullbacks to the mean, then you need 1:1 or even less (i.e. 1:2).

Now this is fairly intuitive so far. What I would say is that both from people I've seen make a lot of money and from backtesting, the way to clean up is to operate a win:loss ratio of 3:1 or higher. The fool's gold of 1:100 (e.g. LTCM, Niederhoffer, Cohodas) will only work for so long.

So - run two systems. Have a 3:1 and a 1:2. Then maybe you can make money and appease the psyche.

 

[DashRiprock]

Dash, interesting post.

I've done extensive research into all the "basic" systems, whether they're trend following, mean reverting or variants thereof.

Like you (and everyone else) I would like to have a high win rate, ergo why not shoot for 1:2 rather than the oft quoted 2:1.

Here's my rough and ready conclusion - if you're trading breakouts or trends, you must have a 2:1 ratio or greater, because that's the very essence of the system.

However, if you're looking for pullbacks to the mean, then you need 1:1 or even less (i.e. 1:2).

Now this is fairly intuitive so far. What I would say is that both from people I've seen make a lot of money and from backtesting, the way to clean up is to operate a win:loss ratio of 3:1 or higher. The fool's gold of 1:100 (e.g. LTCM, Niederhoffer, Cohodas) will only work for so long.

So - run two systems. Have a 3:1 and a 1:2. Then maybe you can make money and appease the psyche.

hi meanreversion

well heres the thing (I don't know about niedhoffer, but)

LTCM blew up cos they were going for 1:100 but they only had 90 to risk. like they just kept it on and on and on because it always comes back. well probably alot of people lose money this way, and so they should. Im not saying run your losses forever (like LTCM), just give them more room than your wins.

As for howard... well i guess its sort of the same, except that I am assuming that you have an edge before you start. Howard doesnt have an edge, so he's doomed. But if you DO have an edge, changing your r:r to less than 1 shouldn't negate it (should it?)

I mean if you have an edge, you have an edge. The R:R is just maths.

 

[Jack o'Clubs]

I'm celebrating for you :hic:

Also it worth saying... not for the benefit of us sensible people, but for the morons who will read this and blow up... IF YOU'RE PAYING A SPREAD YOU SHOULD NOT BE TRADING THE TIMEFRAMES WHERE THERE IS STRONG MEAN REVERSION, EXCEPT IN CERTAIN CIRCUMSTANCES WHICH YOU WILL KNOW ABOUT IF YOU ARE GOOD ENOUGH TO DO IT, AND THUS WON'T HAVE BOTHERED READING THIS FAR. IF YOU'RE STILL READING THIS FOR GOD'S SAKE DON'T DO IT.

They'll probably just think you mean they need to trade with worldspreads or something.

This thread reminds me of the old saw a long term position is a short term position gone wrong.

 

[Jack o'Clubs]

hi meanreversion

I mean if you have an edge, you have an edge. The R:R is just maths.

Good comment.

 

[DashRiprock]

They'll probably just think you mean they need to trade with worldspreads or something.

This thread reminds me of the old saw a long term position is a short term position gone wrong.

Hi Jack of Clubs

well yes, "a long term position is a short term position gone wrong" or more one that I know "I'm not averaging a loser, I'm scaling in to a winner" is not really what I am saying.

Im saying have stops and keep them. don't change them, don't double up, or anything like that. But who says stops have to be smaller than targets? It makes more sense the other way around I think.

(as well are worldspreads just another spreadbetter? I dsont know much about spreadbetting :/)

 

[wackypete2]

I mean if you have an edge, you have an edge. The R:R is just maths.

True. You can't change a negative expectancy to a positive one using only money management (math).

Peter

 

[robster970]

Its v interesting you say with more frequency, because in my eyes (and only my eyes ok im not some wizard) the more frequently / shorter time of typical trade, the more attractive it becomes to have r:r < 1

I agree. I came to realise this a few months ago when I asked Pete some advice in a PM associated with trading reversals. He hit the nail on the head when he suggsted I was holding out for too much of a move. After thinking about this some more, the penny finally dropped that this was not just specific to the nature of the reversal but it was also associated with the timeframe too. This was probably the point that tipped me from breakeven to profitability and it's stood me in good stead since.

So as I said before, I'm looking at improving frequency by doing that RTH-High/Low **** that I told you about in another conversation (or was that a PM?) by doing it in correlated markets to ES (i.e. NQ and YM). My average R:R is 1.62 currently. Will it come down in time? Probably. Will it go down sub 1? Maybe.

Interesting thread though dude.

 

[DashRiprock]

True. You can't change a negative expectancy to a positive one using only money management (math).

Peter

and changing your R:R doesn't mean your edge is now balls*


* but as a bit of pedantry, "signal" frequency and commissions come into play here, but lets ignore these for sake of good discussion.

 

[robster970]

exactly. few understand that which is why so many say you can't trade off small TF's. If you take countertrend trades off 1m or 5m charts then trying to get RR > 1 is ruin.

Peter

This is exactly what Pete told me that made the difference.

 

[Mao]

there is popular saying in trading "cut your losers short, let your winners run".

I'm saying - balls to that, have small consistent winners and get stung every now and again.

Now first off, if you think about the maths side for a moment, it shouldn't really matter what your R:R is, as long as your risk as a sensible portion of your whole account. what really matters is that you have a positive expectancy for every trade you are making. So, say, comparing two strategies of the same expectancy, the one with r:r that it less than one isn't any better or worse than the one that the r:r is bigger than one.

Size of risk vs size of account comes into it as well but well just say that we can treat them the same. I'm not ignoring it, i'm just saying lets assume that the stategies already have taken that into consideration.

Now, the interesting bit: People say "one of the secrets to trading is to let your winners run and to cut your losers fast".

BUT, we know from experience that most people do the opposite - it's called the Disposition effect.

AND we also know that people naturally choose smaller losses over pro rata gains (Loss Aversion)

We also know that alot of people don't make it as traders because they can't manage to unlearn all the bad habits in thinking like this that everyone picks up in life.

So, what I'm saying is, why not make them work FOR you rather than AGAINST you?!

If you have an edge in your strategy, howver simple or complicated, you should be able to get the same expectancy from each trade by mixing up your r:r to less than one (please dont anybody start that money management, OK? if you think money management is relevant here go straight to jail and do not collect 200 pounds).

For the same expectancy, having r:r less than one makes waaaay more sense if you think about our natural biases?! collecting winners makes you feel good, having a p&l graph that is mostly upward sloping (I mean slow upline then sharp downline and the slow upline again, instead of shallow downline and big upline) makes you more likely to stick to your rules, re-enforces your discipline, and most of the time you are making money that makes you think you are good trader so you stick to your rules and so on and so on.

Alot of traders have problems because they go for r:r bigger than one, and the strings of small lossess causes them all sorts of problems in the mind like lack of discipline, changing rules and stuff.

So why doesn't everyone trade with a r:r that is less than 1?

just idea for talking, thats all

If your trading for RR less than 1, then a positive expectancy is contingent on having a high win rate - the win rate required could be very high
depending on how much less than 1 the RR is. If a very high win rate can be achieved fair enough, easier said than done though

 

[DashRiprock]

If your trading for RR less than 1, then a positive expectancy is contingent on having a high win rate

yes

- the win rate required could be very high
depending on how much less than 1 the RR is

very is a subjective term, see comments below

If a very high win rate can be achieved fair enough, easier said than done though

But its no harder than improving a lower win rate form zero expectancy to the concomitant positive expectancy.

It does, however, ally itself to our natural biases as humans.

 

[Mao]

yes



very is a subjective term, see comments below



But its no harder than improving a lower win rate form zero expectancy to the concomitant positive expectancy.

It does, however, ally itself to our natural biases as humans.

Wowee...what did you say ? Speak English man.

 

[Mao]

yes

very is a subjective term, see comments below .

Well its not - Im sure theres a formula to work out what win rate needs to be based on a certain RR to gain positive expectancy; the lower the RR the higher the WR required; could be 70% plus....

It does, however, ally itself to our natural biases as humans.

So you're suggesting a strategy that most noobs start off with , the need to be right rather than maximising gains and cutting losses. isnt the whole point of learning to trade is learning to reverse this inbuilt human intuition..

Obviously there is a need to understand the timescales / lifecycle of the relevant move traded...

 

[DashRiprock]

Well I started the thread under the impression that everyone would discount the "number of win x size of win VS number of losses x size of loss" argument as a given really.

Its about, all other things being equal, how the "common knowledge" of running winners and cutting losers goes against our behavioural bias, and trading in a way that is parallel to our biases rather than against them.