Win Rate vs R:R

[DionysusToast]

This thread is partly inspired by the ****** 'trendie'. Perhaps coin ****** would be more appropriate... :innocent:

On of the theories that lies behind random entry systems is that you can just get in at any point in time, in any direction and manage your way to profit.

One of the attractions to such a system is that it implies that you don't have to have an opinion the market when you trade. Realistically though - if you don't have an opinion at the time of entry, you really need to get one when you manage the trade. It's not absolution from having an opinion - it's just shifting responsibility.

With a random entry system, you need your winners to be bigger than your losers. You should expect to lose 50% of the time.

So a win rate of 50% is expected and yet at entry time there is no reason for your winners to be larger than your losers because your entry is random.

Anyway - say your starting point is a 50% win rate system. Which direction would you prefer to take that in and why?

1 - Improve the win rate and possibly accept a lower R:R
2 - Improve the R:R and possibly accept a lower win rate
3 - Look to both improve the R:R AND win rate

Obviously number 3 sounds like the best option but the "why" should really explain why you feel it's feasible.

As an aside, it might also be worth considering how your choice would impact the number of opportunities that would occur.

 

[malaguti]

On of the theories that lies behind random entry systems is that you can just get in at any point in time, in any direction and manage your way to profit.

help me out here a bit please, I've got the flu so am a bit slow today..if the market is going down and you randomly choose long, there aint no amount of money/trade management is going to turn that around for you into a profit.
Random entry yes, but this presupposes at least you have the trend right, otherwise like I say, it simply is not going to happen.

 

[DionysusToast]

help me out here a bit please, I've got the flu so am a bit slow today..if the market is going down and you randomly choose long, there aint no amount of money/trade management is going to turn that around for you into a profit.
Random entry yes, but this presupposes at least you have the trend right, otherwise like I say, it simply is not going to happen.

Well - I don't agree with the theory but anyway, no trading method can be judged by a single losing trade can it?

 

[malaguti]

Well - I don't agree with the theory but anyway, no trading method can be judged by a single losing trade can it?

then I would argue that you can't possibly improve a win rate based on my argument above. if its going down you can't turn it back again. so a win rate has to be based on whether your original random entry was lucky enough to be in the direction of the market, and if its 50 50, that is your win rate. it can't be improved upon. thats my logic anyway, which leaves only r/r

 

[DionysusToast]

then I would argue that you can't possibly improve a win rate based on my argument above. if its going down you can't turn it back again. so a win rate has to be based on whether your original random entry was lucky enough to be in the direction of the market, and if its 50 50, that is your win rate. it can't be improved upon. thats my logic anyway, which leaves only r/r

You shouldn't argue with me about it because I think coin toss systems are nonsense. It's just that talk of a coin toss system got me to thinking about what people are trying to achieve...

Hence the question

Anyway - say your starting point is a 50% win rate system. Which direction would you prefer to take that in and why?

 

[alfiemal]

i would say random entry systems are bs aswell, but i would also say that all you need is a basic system to make money with good psychology and money management.

When you say 50% win rate, what does it mean? a 50% win rate at what risk reward, clearly if it was 1:2 then your cleaning up on a 50% win rate...

 

[Liquid validity]

I must have been lucky for the past 2 years then
Yeah, its not the most efficient way.
It is consistent and requires little manual input.
Either way, who cares,
Do whatever floats your boat and works for you.

 

[random12345]

I must have been lucky for the past 2 years then
Yeah, its not the most efficient way.
It is consistent and requires little manual input.
Either way, who cares,
Do whatever floats your boat and works for you.

LV is more jaded than Orwell on his return from Catalonia these days.

A genuine loss to the forum.

 

[Shakone]

The irony is that a large % of the people trading, including some of those that think random entry is BS, are probably trading an entry that is equivalent to a random entry, and they don't even know it.

 

[scholfield]

The irony is that a large % of the people trading, including some of those that think random entry is BS, are probably trading an entry that is equivalent to a random entry, and they don't even know it.

Exactly.
That's kind of what I was getting in my 100 trades thread.
I've done OK in the first 100 trades (i.e I'm up) but i'm not convinced yet that i'm any better than random.

Started another batch of 100 trades yesterday!

 

[Shakone]

I totally agree. I'd estimate that most retail traders have no edge, which is a key reason they fail.

Not just people without edges.

If you wanted to test whether your entry was equivalent to a random entry, what do you think is the best way to do it?

You can't use any sophisticated exit techniques, because that would invalidate the test. You could try just taking fixed stops and targets as Scholfield refers to. You'd have to vary these a bit so that you weren't unknowingly taking advantage of some property of the market.

Alternatively, you could look at the distribution of the price some fixed period of time after entry.

Any other methods?

 

[Atilla]

You shouldn't argue with me about it because I think coin toss systems are nonsense. It's just that talk of a coin toss system got me to thinking about what people are trying to achieve...

Quote:
Anyway - say your starting point is a 50% win rate system. Which direction would you prefer to take that in and why?

Hence the question

Does random entry mean 50% of calls are long and 50% of calls are short?

In which case does it matter which direction the random trades are done in? In fact I would go as far as to say you shouldn't have a reason for going long or short it should be random - as in thought-less !


Unless you are alluding to some indicator like MA to skew the long's v short % calls etc. Then it's not random.


Sounds like you have man flue too

 

[Shakone]

I'm not sure I understand the question... if you use a truly random entry method, the entry will be just that, random. Exits etc are a different topic. Or are you asking how to prove whether your system has edge, or not?

No it's not about proving an edge, because that involves entry and exit. I'm suggesting that many entry techniques are equivalent to just entering at random. The fact that the trader has a reason behind it makes him believe it's not random, which is fine, but it doesn't mean it's any better than a coin flip. You can't look at the overall edge or the results and conclude that the entry is a good one just because you're winning. It may well be that it's how you're managing the trade that is making you money. And in that case, your entry may as well be random.

So, the question is, what is the best way to determine whether the entry that you have is better than a coinflip?

 

[Atilla]

Ok, now I understand your question. So you can chose a set exit criteria and run a large sample of random entries to generate your random results. Then run your system's entries through that large trade sample, with the same exit criteria (testing software would obviously be useful here). Finally compare the two sets of stats.

I understood the Risk:Reward to be fixed from point of entry.

Is exit point R:R variable?

 

[Forexmospherian]

No it's not about proving an edge, because that involves entry and exit. I'm suggesting that many entry techniques are equivalent to just entering at random. The fact that the trader has a reason behind it makes him believe it's not random, which is fine, but it doesn't mean it's any better than a coin flip. You can't look at the overall edge or the results and conclude that the entry is a good one just because you're winning. It may well be that it's how you're managing the trade that is making you money. And in that case, your entry may as well be random.

So, the question is, what is the best way to determine whether the entry that you have is better than a coinflip?

I think it is down to that word "efficiency"

A brilliant efficient trade is one that need minimum stop - ie if the spread is 2 pips then the stop needs to be only 3 pips.

Then the trade will carry on in a direction that is above the average ATR of the hourly charts - ie say ATR on GU at peak times on the hour chart is 28 pips - if the trade made you 33 +pips or more - that is efficient.

So end result on this example is an efficient entry trade that wins and gives a RR of 10 - excellent stuff and as you guess fairly rare.

If you did the same exercise with say "coin tosses" the results would be far more inconsistent and not efficient - ie you will need larger stops for the trade to work and your wins would not produce RR's of 10 - or even 2 or 3 on average.

So efficient trades have the "edge" in terms of results and random entries would not be in the same ball park the majority of the time

Regards

F

 

[Atilla]

The Risk:Reward ratio is not fixed at entry. You have no guarantee of the results for any given trade.

I recognise there is no guarantee of the results for any given trade and I doubt any one could otherwise we wouldn't be having this debate.

You saying no Stop and Exit is set when a trade is placed?

How is the R:R then subsequently calculated or managed?

When is Stop/Limit and what are the R:R ratios?


I understood the question to be one of random entries subject to R:R ratios set & predetermined in advanced.

 

[scholfield]

So, the question is, what is the best way to determine whether the entry that you have is better than a coinflip?

I don't know.
Have a random entry system running along the method in my journal, but with the same 20stop / 30target, with 1000 trades, and see if my method is better?
Not sure....

 

[Shakone]

I don't know.
Have a random entry system running along the method in my journal, but with the same 20stop / 30target, with 1000 trades, and see if my method is better?
Not sure....

It's tricky as you know. For example if you have a larger target than stop, you may inadvertently be capitalising on the market's tendency to have fat-tailed distribution, and your entry wasn't the reason for success. If you run a coin flip alongside your entry and compare then (the previous problem aside) your entries may be capitalising on some time feature of the market, but yes it's a decent attempt at testing it.

It's not an easy problem to solve.

I tried to start a thread on similar issues before. I don't believe that coin flip entry for direction is the best that can be done, because I believe in trend, and successfully recognising that trend alone can change it from a 50/50 to a winning proposition without the need for some amazing entry system. However, I don't dismiss it, because I think it makes you ask the right sort of questions.

 

[Atilla]

No, I'm not saying "no Stop is set when a trade is placed?", but rather that only determines your trade risk (assuming the market doesn't gap through the stop). I am saying a set exit doesn't determine your reward.

fwiw - having played about with fixed R:R I would add that they would really need to be used with other indicators such as ATR, ADX or MA lagging indicators to produce an edge. Otherwise I can see no value in a random entry with subsequent managed or fixed exit levels.

 

[DionysusToast]

I think it is down to that word "efficiency"

A brilliant efficient trade is one that need minimum stop - ie if the spread is 2 pips then the stop needs to be only 3 pips.

Then the trade will carry on in a direction that is above the average ATR of the hourly charts - ie say ATR on GU at peak times on the hour chart is 28 pips - if the trade made you 33 +pips or more - that is efficient.

So end result on this example is an efficient entry trade that wins and gives a RR of 10 - excellent stuff and as you guess fairly rare.

If you did the same exercise with say "coin tosses" the results would be far more inconsistent and not efficient - ie you will need larger stops for the trade to work and your wins would not produce RR's of 10 - or even 2 or 3 on average.

So efficient trades have the "edge" in terms of results and random entries would not be in the same ball park the majority of the time

Regards

F

I think you have 2 ways to approach directional trading.

I do not think you can consistently be onside with just 2-3 ticks risk, make 10x your risk on those trades and have a high win rate.

I think if you look at areas where risk is defined at a point very close to your entry, you will have to tolerate a higher rate of losses.

Of course, that does presume a valid reason for having a trade on and the failure point so close.