Is a great % of wins better than more pips ? | Forex Factory

Options

Permalinks
Print Thread

Similar Threads

A better Editor for mql is available. 100x better than MetaEditor 20 replies

Great trading stories: life changing wins/losses 12 replies

100 Trades more than 75 wins 6 replies

I lose 40 pips and win 6 to 10 pips... 17 replies

Is a great % of wins better than more pips ?

  • #1
  • First Post: Mar 21, 2007 8:24am Mar 21, 2007 8:24am
Hello,

I'm just backtesting a system, but i've found something interesting.

If i set a smaller trailing stop, the overall gain in $ is less, but there is a higher % of winning trades. (This is obvisouly expected if you think about it logically)

SO the question I'm wondering is this...

Is it better to have a higher % of winning trades (ie 75% compared to say 50%) but a smaller take profit or trailing stop (or what is more important, a higher % wins or overall pip numbers?)... [this is using the same lots for each trade]

My thoughts would be that if you have a higher % of winning trades... ie more confident that each trade is a higher chance of winning, then you could increase the number of lots.

Thanks, Anthony
  • #2
  • Mar 21, 2007 9:01am Mar 21, 2007 9:01am
Quoting anthonyrae

My thoughts would be that if you have a higher % of winning trades... ie more confident that each trade is a higher chance of winning, then you could increase the number of lots.
Exactly. If your winning % is good enough, it's a liberating feeling. You can risk a lot more (Be reasonable; Personally I wouldn't go higher than 10% risk, and I only trade 5% risk per trade myself with a winning ratio of 70%). The only thing bad about risking more, is that it is more psychologically taxing when you lose; not to mention you could blow your account quickly if you ran into some tough times.

Regards,

Kevin
  • #3
  • Mar 21, 2007 9:02am Mar 21, 2007 9:02am
its a balance

high winrate = low drawdowns
high payrate = more money

the low drawdowns part is obvious, if you dont lose 10x in a row you wont have huge draw downs

the "more money" part is because if money management. the most you can make on a single trade is Risk * winning amount / losing amount. to tp 5 , sl 10 = R * 5 / 10 vs a 10 tp, sl 5 = R * 10 / 5

so which is better? its in the middle somewhere. you want to make as much money as possible without risking your account.

sorry for not giving an absolute answer but it really depends on how much of a drawdawn you can take
  • #4
  • Mar 21, 2007 9:27am Mar 21, 2007 9:27am
I'm glad someone is thinking about this. I've tested and thought about this for a while now. Is it:
A - Better to have a high winning% with a lower r/r or
B - Better to have a low winning% with a higher r/r?

The answer to me seems rooted in the problem of assymetrical leverage (you lose 50%, but then have to make 100% to get back where you where) which creates a bias against A. Then there is the problem of open equity drawdowns which will be more severe with B and less so with A.

At the end of the day I don't think either is better although with A you would need to make many trades to make decent returns, as each winning trade will only account for a small profit. But with B you can trade less frequently as it only takes one home-run to get a good return.

At the end of the day it comes down to psychological preference. I like to hit homers once in a while rather than trade all day.
  • #5
  • Mar 21, 2007 9:39am Mar 21, 2007 9:39am
F,

I agree (again). This is a very personal preference, although for some reason I tend to think it's easier to work a system that as a low winning %, but a high enough RR to be profitable than it is to find a system with a high winning% but have a high enough RR to be profitable. I suppose the happy medium must be somewhere in the middle (50% RR, with a slightly higher than 1:1 RR).


Quoting faure
I'm glad someone is thinking about this. I've tested and thought about this for a while now. Is it:
A - Better to have a high winning% with a lower r/r or
B - Better to have a low winning% with a higher r/r?

The answer to me seems rooted in the problem of assymetrical leverage (you lose 50%, but then have to make 100% to get back where you where) which creates a bias against A. Then there is the problem of open equity drawdowns which will be more severe with B and less so with A.

At the end of the day I don't think either is better although with A you would need to make many trades to make decent returns, as each winning trade will only account for a small profit. But with B you can trade less frequently as it only takes one home-run to get a good return.

At the end of the day it comes down to psychological preference. I like to hit homers once in a while rather than trade all day.
  • #6
  • Edited 10:16am Mar 21, 2007 9:50am | Edited 10:16am
There is one more thing to consider. If you decrease your TP and SL, and increase your Lot size, the amount of the spread the broker charges is much more important...

For example: If you have a TP of 100 pips and a lots size of say 1.
with a spread of 5, that is 5 % of the total movement of the trade.

If you have a TP of 10 pips and a lot size of 10. that 5 pip spread becomes
50 % of the trade you will pay to the broker

So in the first case you pay $50.00 assume a usd currency and you keep $950.00
and the second case you pay $500.00 to the broker and you keep $500.00

Edit:
To carry this out though the stop loss is also important.
Lets say you have a 1:1 RR If you place 10 trades, in the first case you would break even.
In the second case, (even though you pay more to your broker) you would have 7 winning trades of $500 or $3500 and you would have 3 losing trades of $500 or -$1500, so you would have a net gain of $2000.

The point is that all four factors are important in the decision of which is best.
  • #7
  • Mar 21, 2007 9:54am Mar 21, 2007 9:54am
Glad you guys are putting the "thinking hat" on. I have been working on this for the past weeks. Comparing, Testing, Testing, more comparison...

I know for fact: The PSYCHOLOGICAL aspect is important. No, lets correct myself. VERY important. It is a matter of:

- Only making a few pips, but you need the nerve to enter again and again and again (this requires lots of trust in your system! And not missing re-entries)

OR. The other way around: low rate of winnings but big returns....

- It is difficult to see your account "go down" 40, 50, 70, and more pips into the red (as I have seen with many systems around here) and you really DON'T KNOW if the return is going to be big.

I would like to ask Itme to "come" over here and give some insights, because I have problems saying "my system is a 5:1" as R:R goes....but you really never know if you can get 5 times your risk back... You could SAY to yourself: "well, I don't care about the 80 or 90 pips in the red, because this is a trade that will return 500...." well, I'm not sure about this.

Maybe I'm only tricking myself with this reasoning????

Anyway, I have been working on a system that would give HIGH FREQUENCY of trades (maybe 5 signals/day per currency pair) with a winning rate of 80% or better. Pips to be made: 5 pips. Stop Loss: 10 Pips.

But I don't know if this is the way to go....

Thanks for opening this great thread!!!

Peter.
  • #8
  • Mar 21, 2007 10:18am Mar 21, 2007 10:18am
Pete,

Comment on your first part about risk/reward...it really is a personal thing. It depends on your risk tolerance. If you can't see your trades go down 80 pips, then you need to know why. If you're trading a standard account (1 lot) then you're looking at an $800 drawdown, on a mini account, that's in the neighborhood of $80, on a micro that's $8...so, perhaps you see where I'm going with that. If you're unsure about being able to see a certain pip-value, it's because you're tradesizes may be too high for your risk aversion in the case of having to hold onto larger losing positions.

To Scott's point, if you take spread into account, then you may be looking at drawdowns differently. I look at it like this, if I'm trading for 4 pips, and the spread on my pair is 4 pips, then I'm look at 50% of my drawdown just from the spread alone. That's an exceedingly high trading cost, in my opinion, but that has to do with MY risk aversion. While no one likes to see larger losses, if you flip it around to the profits, and take a 4 pip profit on a 4-pip spread, I'm looking at a 50% trading cost...if my trade is 96 pips, then it's only 4% trading cost which is still sorta high, but more manageable.

Now, your second point about your system. you have a 2:1 risk to reward ratio...which means that you have to win 2 times for every 1 loss just to break even. Therfore you HAVE to have a long-term winning % of > 66.7% in order to break even. If what you're saying is true and you have an 80% win ratio, then this system will be profitable. I'm hoping whatever method you used to determine the winning %, that you factored in the spread, because if you didn't you'll need to factor that in too otherwise, you will be unpleasantly surprised.

Regards,
Michael

Quoting petercross
Glad you guys are putting the "thinking hat" on. I have been working on this for the past weeks. Comparing, Testing, Testing, more comparison...

I know for fact: The PSYCHOLOGICAL aspect is important. No, lets correct myself. VERY important. It is a matter of:

- Only making a few pips, but you need the nerve to enter again and again and again (this requires lots of trust in your system! And not missing re-entries)

OR. The other way around: low rate of winnings but big returns....

- It is difficult to see your account "go down" 40, 50, 70, and more pips into the red (as I have seen with many systems around here) and you really DON'T KNOW if the return is going to be big.

I would like to ask Itme to "come" over here and give some insights, because I have problems saying "my system is a 5:1" as R:R goes....but you really never know if you can get 5 times your risk back... You could SAY to yourself: "well, I don't care about the 80 or 90 pips in the red, because this is a trade that will return 500...." well, I'm not sure about this.

Maybe I'm only tricking myself with this reasoning????

Anyway, I have been working on a system that would give HIGH FREQUENCY of trades (maybe 5 signals/day per currency pair) with a winning rate of 80% or better. Pips to be made: 5 pips. Stop Loss: 10 Pips.

But I don't know if this is the way to go....

Thanks for opening this great thread!!!

Peter.
  • #9
  • Mar 21, 2007 10:39am Mar 21, 2007 10:39am
Robert Prechter, who won the 1984 Trading Championship wrote in "What a Trader Really Needs to Be Successful":

"Requirement #1 : An objectively definable method".

If one can find, create, or somehow acquire this type of trading system, then the psychological factor should be effectively eliminated. All decisions can be made in a logical, premeditated fashion BEFORE the trading program is commenced.

I have been working on the thread topic lately, but haven't finished my analysis. I created two elaborate Excel spreadsheets, but one seems to have disappeared from my hard drive!

If one can accept the Kelly formula, which is only really valid if the losses occur randomly, then the return on capital over the course of a series of objectively defined trades is a function of only average Reward : Risk ratio, percentage of winning trades, and number of trades. The exact curve is what I was working on, when one extremely complex spreadsheet apparently disappeared. It calculated the return on capital based on these variables.

The question is a mathematical question, and can only be answered mathematically. Psychology is a factor in trading, but if the trading system is objectively defined, then it should not enter into the trading decisions. If a person is logical, then the only consideration entering into the design of the trading program is the optimisation of performance. The answer to this lies in the mathematics. There are fairly complex formulas with exponential functions involved, so casual speculation is unlikely to provide the correct solution.

Optimisation of a trading system is a mathematical problem, and can only be solved mathematically, once the basic trading algorithm is established and the system variables are being tweaked. The perfect testing software would incorporate all the financial outcome optimisation formulas I have been working on in order to guide the system developer whose eye is on the ball - optimum return on capital, not psychological appeasement or "scoring pips".

Drawdowns are part of a winning trading system, and as I have explained elsewhere, should be maximized within acceptable risk tolerances, not minimized in the design of the trading system. Win rate, frequency of trades, distribution of losses, and reward / risk ratio are all important variables. How to adjust the trading protocol to juggle them exactly to optimize performance can only be answered by quantitative analysis employing a properly programmed computer. The human mind is unlikely to be able to solve this complex mathematical problem, unless one is a mathematical genius.

More later.
  • #10
  • Mar 21, 2007 5:49pm Mar 21, 2007 5:49pm
hi everyone.. thank you all for the great and insightful replies. It's nice to see that other people have considered this question as well !!!

What petercross and others have mentioned is the psychological factor plays a huge role !! Have you ever been disappointed after a big loss ? (I have recently with sharetrading) but it does decrease your confidence level, and you think - is it going go be profitable this time ?

Then there is the whole thing about getting just 5/10pips a day ... but the spread cuts you a lot ! I have come to realise, that setting larger targets with larger stops is the way to go, simpy because the nature of FX is very voatile, so you have to allow for quite a bit of movement on both sides.

here is an example with something i'm testing out.

same system, just different TS parameters:
using just a TS (no TP) the profit trades is 55% but the amount won is double (say $3000) what i get some one using a TP of say 30pips (say $1500) but the wins ratio is 74% ?

For the higher ration winning settings i would have to double my account risk from 10% to 20%.

itme - looks like you raised something I'll need to look into - the Kelly money management. I have heard of it, but thats about it.

1. just an aside.. can someone tell me where to get go historical data for backtesting?) - I just want to be sure that the system does okay. I am also demoing it but would be more comfortable if I had better historical data.

2. Also.. is tweaking or optimising your stop loss and take profit levels 'cheating' (or i have heard the term curve fitting) or is it just in fact finding the right variables for the system which work in the past, so have a higher probabliilty of working in the future ?

Thanks
  • #11
  • Mar 21, 2007 6:03pm Mar 21, 2007 6:03pm
Quoting anthonyrae
hi everyone.. thank you all for the great and insightful replies. It's nice to see that other people have considered this question as well !!!

What petercross and others have mentioned is the psychological factor plays a huge role !! Have you ever been disappointed after a big loss ? (I have recently with sharetrading) but it does decrease your confidence level, and you think - is it going go be profitable this time ?

Then there is the whole thing about getting just 5/10pips a day ... but the spread cuts you a lot ! I have come to realise, that setting larger targets with larger stops is the way to go, simpy because the nature of FX is very voatile, so you have to allow for quite a bit of movement on both sides.

here is an example with something i'm testing out.

same system, just different TS parameters:
using just a TS (no TP) the profit trades is 55% but the amount won is double (say $3000) what i get some one using a TP of say 30pips (say $1500) but the wins ratio is 74% ?

For the higher ration winning settings i would have to double my account risk from 10% to 20%.

itme - looks like you raised something I'll need to look into - the Kelly money management. I have heard of it, but thats about it.

1. just an aside.. can someone tell me where to get go historical data for backtesting?) - I just want to be sure that the system does okay. I am also demoing it but would be more comfortable if I had better historical data.

2. Also.. is tweaking or optimising your stop loss and take profit levels 'cheating' (or i have heard the term curve fitting) or is it just in fact finding the right variables for the system which work in the past, so have a higher probabliilty of working in the future ?

Thanks
Most people get metatrader data from the Alpari site:
http://old.alpari.org/en/dc/databank

As for optimizing, as long as you have several sets of parameters that get similar results as your most optimal set of parameters, you should be okay.
  • #12
  • Mar 21, 2007 7:02pm Mar 21, 2007 7:02pm
Quoting anthonyrae
...
Is it better to have a higher % of winning trades (ie 75% compared to say 50%) but a smaller take profit or trailing stop (or what is more important, a higher % wins or overall pip numbers?)... [this is using the same lots for each trade]
Thanks, Anthony
Depends on the size of your ulcer or, how much you want to avoid them.
ie. no two Snow Flayx@ [@ is jic the US Gov decides to use it as a Column Title.] {fx traders} are alike.

This is something that should be noted in your journal.{btw the following is totally fictitious. pls don't ask for the system that provided it 'cause I dont have it.}
"Pulled 200 P/C on the Long GBPUSD Trade opened Mar 16, closed today {Mar 21} but felt like the sky was falling when it moved 100P against me."
  • #13
  • Mar 22, 2007 9:04am Mar 22, 2007 9:04am
Quoting mrmikal
F,

I agree (again). This is a very personal preference, although for some reason I tend to think it's easier to work a system that has a low winning %, but a high enough RR to be profitable than it is to find a system with a high winning% but have a high enough RR to be profitable. I suppose the happy medium must be somewhere in the middle (50% RR, with a slightly higher than 1:1 RR).
It's definetly a personal choice but the answer to the question is way more subjective than what you think; what is a happy medium? To you a high r/r might be 1:5 but to me a high r/r is 1:20. Conversely a low r/r for you might be 1:0.1 and for someone else it might be 1:1. It's all about what kind of d/d can you handle (the large sudden d/d you get with a high winning % or the smaller losses culminating in a overall larger d/d you get with a system that has a high r/r). The same goes for winning %.

I'd say the best situation would be a system with a high winning % and a high r/r. :
  • #14
  • Mar 22, 2007 8:48pm Mar 22, 2007 8:48pm
Is a great % of wins better than more pips ?

I enjoy the benifits of both.
  • #15
  • Mar 22, 2007 11:40pm Mar 22, 2007 11:40pm
Quoting fx.rhino
Is a great % of wins better than more pips ?

I enjoy the benifits of both.
Rhino, if you don't mind sharing, what means do you employ to enjoy the benefits of both an abundance of pips and high win %?
  • #16
  • Mar 23, 2007 12:01am Mar 23, 2007 12:01am
Don't worry about the spread cost when you trade. You enter and exit trades based on technicals, psychology, and fundamentals. I could care less how much spread a broker makes on my trade. What goes in my pocket is most important. I often trade 50 to 100 lots so the broker is making $1,500 to $3000 on my trade with a 3 pip spread. I focus on making money for myself, not how much my broker is making.
  • #17
  • Mar 23, 2007 1:38am Mar 23, 2007 1:38am
I see all of the regular experts debating this topic...very interesting points made by all. One thing hit me though and I though I might add to this conversation...

I use IBFX as my broker... currently im only trading a mini lot... but IBFX offers something called a platinum Rewards card.

Basically, The points are calculated off of standard lot sizes....for every ten mini lots would equal one standard lot...a round lot trade is buying or selling a position until closing the postion...in and out of the trade is a round lot trade...

You would need to open and close ten 1.00 lots in a mini account, so at .10 lot size you would times that by ten, so 100 round lot trades would equal 50 reward points...

The place to redeem these points is here:
http://www.zionsbank.com/connect_one_awards2.jsp#travel

Ok now on to my point...By closing more lots...I.E. more trades... the broker gets more pips (due to the spread) and rewards us with travel points. So, wouldnt the correct answer be.. the system with the highest winning % AND the one that closes the most round lots? Then we could rack up travel points too...

Thoughts?

-Matt
  • #18
  • Mar 23, 2007 10:46pm Mar 23, 2007 10:46pm
Quoting Rolandlc33
Don't worry about the spread cost when you trade. You enter and exit trades based on technicals, psychology, and fundamentals. I could care less how much spread a broker makes on my trade. What goes in my pocket is most important. I often trade 50 to 100 lots so the broker is making $1,500 to $3000 on my trade with a 3 pip spread. I focus on making money for myself, not how much my broker is making.

Well said, who cares how much the broker makes, of course that is the cost of doing business. What goes into my own account at the end of the day is the most important thing.
  • #19
  • Last Post: Mar 24, 2007 10:04am Mar 24, 2007 10:04am
Quoting juxta14
Well said, who cares how much the broker makes, of course that is the cost of doing business. What goes into my own account at the end of the day is the most important thing.
It's naive to ignore spread costs and when you're trading a method that has a high number of entries, like trading short term with high leverage. What you seem to be saying is, "I don't care how much my broker makes as long as I make money". You could end paying your broker a huge amount in spread costs. I don't think you can afford to just ignore that.
0 traders viewing now