Probability of trading outcomes | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

My Trading Outcomes ! 8 replies

Charting Trade Outcomes 0 replies

Set and Forget Outcomes! 1 reply

Why not the same outcomes? 4 replies

Attachments: Probability of trading outcomes
Exit Attachments

Probability of trading outcomes

  • #1
  • First Post: Aug 7, 2011 11:10am Aug 7, 2011 11:10am
hy guys. if you have a system which wins 80% of the time and the risk/reward is 1 to 0.8 what is the probability that in 50 trades you're going to break/even. also what is the most probable outcome in 50 trades.
  • #2
  • Aug 7, 2011 11:16am Aug 7, 2011 11:16am
Quoting deltatrade
hy guys. if you have a system which wins 80% of the time and the risk/reward is 1 to 0.8 what is the probability that in 50 trades you're going to break/even. also what is the most probable outcome in 50 trades.
50 trades

40 winners
10 loosers

winners bring 32 units, loosers loose 10...so thats 3,2 to 1 win loss ratio

The most probably outcome is nice profit. What is your strategy based on?
  • #3
  • Aug 7, 2011 11:26am Aug 7, 2011 11:26am
sorry , this is not the type of answer that i'm looking for.
  • #4
  • Aug 7, 2011 11:56am Aug 7, 2011 11:56am
Quoting deltatrade
sorry , this is not the type of answer that i'm looking for.
The probabilty of being in positive balance is way over 99% after 50 trades so i dont see a point in calculating the excat number.
  • #5
  • Aug 7, 2011 12:09pm Aug 7, 2011 12:09pm
Quoting deltatrade
hy guys. if you have a system which wins 80% of the time and the risk/reward is 1 to 0.8 what is the probability that in 50 trades you're going to break/even. also what is the most probable outcome in 50 trades.
be careful about calculating the probabilities. Cause this is then just the AVERAGE probability.

Even if you have a system that is right 80% of the time you might experience a time period, when your next 50 trades will all be losers. These are then statistical outliers.
And then your account will be wiped out.

You need to plan for the unlikely events, not for the likely ones. Not planning for unlikely events is what caused the financial crisis, the nuclear meltdown in japan,... among other things.
  • #6
  • Aug 7, 2011 1:04pm Aug 7, 2011 1:04pm
custos i am aware that bad periods always await. if you look closely at my question i asked what is the probability of going to break even in 50 trades. if you know math maybe you can help.
  • #7
  • Aug 7, 2011 1:23pm Aug 7, 2011 1:23pm
Quoting deltatrade
custos i am aware that bad periods always await. if you look closely at my question i asked what is the probability of going to break even in 50 trades. if you know math maybe you can help.
maybe I am too simple-minded here.
If you take a win/loss ratio of 80% and an R:R of 1:1 then it should be the following.
no matter how many trades you take you have the same probabilities:

80% probability that you end up with a profit, 20% probability that you end up with a loss. Well, not sure how one would calculate the probability of break-even. Does your win/loss ratio include b/e trades? So, are the 80% shared by b/e and win-trades? Or is the 20% probability shared by losses and b/e trades?

Once you know this, you still need to know how likely a b/e trade is. I don't think you can just calculate it from the figures you gave.

Obviously also in 50 trades the most likely outcome will be a profit with 80% probability.
  • #8
  • Aug 7, 2011 1:37pm Aug 7, 2011 1:37pm
Quoting Custos
maybe I am too simple-minded here.
If you take a win/loss ratio of 80% and an R:R of 1:1 then it should be the following.
no matter how many trades you take you have the same probabilities:

80% probability that you end up with a profit, 20% probability that you end up with a loss. Well, not sure how one would calculate the probability of break-even. Does your win/loss ratio include b/e trades? So, are the 80% shared by b/e and win-trades? Or is the 20% probability shared by losses and b/e trades?

Once you know this, you still need to know how...
He said risk is 0,8:1 but for simplicity lets say its 1:1

Than the probability of loosing after 50 trades should be 0,2 * 0,2 * 0,2 ...and so on...50 times....very very low indeed.
  • #9
  • Aug 7, 2011 1:39pm Aug 7, 2011 1:39pm
Quoting reres
He said risk is 0,8:1 but for simplicity lets say its 1:1

Than the probability of loosing after 50 trades should be 0,2 * 0,2 * 0,2 ...and so on...50 times....very very low indeed.
yeah, you are right about the calculation. Now that you wrote it, I remember, haha
  • #10
  • Aug 7, 2011 6:11pm Aug 7, 2011 6:11pm
Quoting reres
He said risk is 0,8:1 but for simplicity lets say its 1:1

Than the probability of loosing after 50 trades should be 0,2 * 0,2 * 0,2 ...and so on...50 times....very very low indeed.
I'm not a maths guy, but isn't that the probability of not having a win after 50 trades?

i think he wants to know what the proabality of breaking even after 50 trades?
  • #11
  • Aug 7, 2011 6:49pm Aug 7, 2011 6:49pm
Quoting Custos
yeah, you are right about the calculation. Now that you wrote it, I remember, haha
You forget one thing, guys - connected and unconnected probabilities. Just like a roulette, your probabilities are always 50:50
  • #12
  • Aug 7, 2011 7:30pm Aug 7, 2011 7:30pm
Quoting reres
50 trades

40 winners
10 loosers

winners bring 32 units, loosers loose 10...so thats 3,2 to 1 win loss ratio
http://loseloose.com/
  • #13
  • Aug 7, 2011 7:46pm Aug 7, 2011 7:46pm
The expectancy of his "system" is very high (by real-life trading standards).
The math is not simple.

The chance of only a break-even result is very low after fifty trades
because of the high expectancy of this "system".

Likewise, the chance of a break-even or BETTER total gain after fifty trades is very high.

You could find an equity curve simulator (google) and input his values.
The answers you seek are related to the probability distribution of many "generated"
possible equity curves of 50 trades based upon his inputs.

I was involved in a different thread awhile ago-
http://www.forexfactory.com/showthread.php?t=229266

which asked a similar question

Quoting kk007
For example, if his prob. winning 1 unit is 0.6 and losing 1 unit is 0.4, what is his probability of having ever losing more than or equal to 10 units after 30 trials?
The exact math is not simple, but a reasonable estimation can be discovered.
  • #14
  • Aug 7, 2011 8:00pm Aug 7, 2011 8:00pm
No way to know - Random walk theory. Also, how do you calculate brokers shannigans? Also, you have swap/rollover.
  • #15
  • Edited 10:10pm Aug 7, 2011 9:47pm | Edited 10:10pm
Quoting Intu
The expectancy of his "system" is very high (by real-life trading standards).
The math is not simple.

The chance of only a break-even result is very low after fifty trades
because of the high expectancy of this "system".

Likewise, the chance of a break-even or BETTER total gain after fifty trades is very high.

You could find an equity curve simulator (google) and input his values.
The answers you seek are related to the probability distribution of many "generated"
possible equity curves of 50 trades based upon his inputs.

I was...
thanks for the response. this is what i was looking for. it will be nice if someone actually computed the numbers.
  • #16
  • Aug 8, 2011 12:11am Aug 8, 2011 12:11am
WinTrades + LossTrades = 50 -------- (1)
0.8 * WinTrades - 1 * LossTrades >= 0 -------- (2)

Solve (1) & (2), therefore WinTrades >= 27.7, which means you need minimum 28 win trades in order to breakeven or profitable.

Put this formula "=1-BINOMDIST(27,50,0.8,TRUE)" into excel then you will get the discrete probability for becoming profitable or at least breakeven after 50 trades. In your case the answer is 99.997%

Please bear in mind that we assume that each trade outcome is independent of the previous trades, which means there is no tendency for win trade followed by another win trade, or a loss trade followed by another loss trade.
  • #17
  • Aug 8, 2011 1:35am Aug 8, 2011 1:35am
i'm glad that i don't have to learn probs just for a problem.
  • #18
  • Aug 8, 2011 1:48am Aug 8, 2011 1:48am
Quoting havaiana
I'm not a maths guy, but isn't that the probability of not having a win after 50 trades?

i think he wants to know what the proabality of breaking even after 50 trades?
Yes you are correct, but from this we can estimate that probability is extremly low.

Anyone with 80% win ratio strategy should NOT be even asking whats the probability of winning after large sample of trades. He should be trading instead
  • #19
  • Aug 8, 2011 2:32am Aug 8, 2011 2:32am
Quoting reres
Yes you are correct, but from this we can estimate that probability is extremly low.

Anyone with 80% win ratio strategy should NOT be even asking whats the probability of winning after large sample of trades. He should be trading instead
How large is large?
  • #20
  • Aug 8, 2011 2:41am Aug 8, 2011 2:41am
Quoting reres
Anyone with 80% win ratio strategy should NOT be even asking whats the probability of winning after large sample of trades. He should be trading instead
Its either stupidity or insanity that would lead somone to trade without asking that question.
0 traders viewing now