Probability in Trading

[the hare]

Whats a good trading site?

To be honest, you are probably better off verifying this stuff by testing.

At the most basic level, you have 3 variables. You want to increase win rate and the size of the average win, and you want to decrease the size of the average loss.

Fixed stops and targets add an additional constraint to achieving 2 of those objectives !

Something as simple as trading with the trend will improve win rate if done correctly (identifying trend objectively isnt as easy as everyone makes out)

 

[pbylina]

To be honest, you are probably better off verifying this stuff by testing.

At the most basic level, you have 3 variables. You want to increase win rate and the size of the average win, and you want to decrease the size of the average loss.

Fixed stops and targets add an additional constraint to achieving 2 of those objectives !

Something as simple as trading with the trend will improve win rate if done correctly (identifying trend objectively isnt as easy as everyone makes out)

Thank you. This is what I was looking for. To increase win rate I should go with the trend. Even it only gets me to 55% win rate I will still be making money correct? I just have to accept the fact that there will be drawdowns right? I read a good article here. Psychology: Thinking in terms of Probability & Expectation

 

[the hare]

IMO this is a very dangerous way of thinking about trading.

I'm interested in why you think its dangerous

I'd argue that its the completely unrealistic expectation that every trade is going to go as planned, and most days are going to be profitable that really screws up peoples chances of being successful.

I've said it a few times previously, but more than 80% of my time as a trader is spent losing money or treading water. The last time I looked I was only making new equity highs around 18% of the time.

Even with such appalling statistics, I can guarentee that most people posting here simply would not believe my annual returns, and thats trading at very low leverage by most peoples standards.

I'd argue that understanding the practical realities of the role of probability and varience is one of the fundemental first steps.

 

[6am]

Anyone care to tell me why I'm talking sh1t? I'm fairly certain I am.

Here is my attempt I am sure you could use the tree you draw to calculate probabilities but there is another way.

Let TP - take profit in ticks, Ptp - probability of taking profit, SL - stop loss in ticks, Psl - probability of stop loss.

Ptp + Psl = 1
EV = TP * Ptp - SL * Psl = 0

TP * Ptp - SL * (1 - Ptp) = 0

Ptp(TP +SL) = SL

Ptp = SL / (TP + SL)
Psl = TP / (TP + SL)

Example 1.
A 3 tick stop loss and 2 tick TP
Ptp = 3 / (3+2) = 60%

Example 2.
A 8 tick stop loss and 2 tick TP
Ptp = 8 / (8+2) = 80%

 

[scose-no-doubt]

I see what you're saying. Your way considers only the absolute values where I'm assessing the movement of price. My beef is whether the EV is applicable.

 

[pbylina]

Here is my attempt I am sure you could use the tree you draw to calculate probabilities but there is another way.

Let TP - take profit in ticks, Ptp - probability of taking profit, SL - stop loss in ticks, Psl - probability of stop loss.

Ptp + Psl = 1
EV = TP * Ptp - SL * Psl = 0

TP * Ptp - SL * (1 - Ptp) = 0

Ptp(TP +SL) = SL

Ptp = SL / (TP + SL)
Psl = TP / (TP + SL)

Example 1.
A 3 tick stop loss and 2 tick TP
Ptp = 3 / (3+2) = 60%

Example 2.
A 8 tick stop loss and 2 tick TP
Ptp = 8 / (8+2) = 80%

So obviously I would go with the 80%.

But something doesnt make sense. Now if I add an extra edge such as tape reading, does the 80% go up to 90%?

 

[Lord Flasheart]

I've briefly skipped over last posts as been out but what is suggested cannot work in the real world. By the time you've included spreads and slippage,you will need over 55% to show a profit. casinos win on 50.75 %. in order to compensate for this you have to look beyond a simple mathematical solution,If there was one id have found it by now.My staking plan is pretty much known by many, and all I do is add in to it trades that I think will produce a positive expectancy over time.Some are 15 pip stops,some up to 40 pip stop,but the important thing is that I believe I have an edge,and apply exact staking rules according to the chart in front of me. what im trying to say is that most of what I do is mechanical in my mind but doesnt look like that in pratice. Im probably way of thread now so will shut up and put the rum away.

 

[pbylina]

I've briefly skipped over last posts as been out but what is suggested cannot work in the real world. By the time you've included spreads and slippage,you will need over 55% to show a profit. casinos win on 50.75 %. in order to compensate for this you have to look beyond a simple mathematical solution,If there was one id have found it by now.My staking plan is pretty much known by many, and all I do is add in to it trades that I think will produce a positive expectancy over time.Some are 15 pip stops,some up to 40 pip stop,but the important thing is that I believe I have an edge,and apply exact staking rules according to the chart in front of me. what im trying to say is that most of what I do is mechanical in my mind but doesnt look like that in pratice. Im probably way of thread now so will shut up and put the rum away.

No, it is a good post and about the subject.

What is your staking plan? Adding contracts to winning trades? Isnt it called scaling in? Does it not work if your trading 1 contract? Does someone need atleast 3 contracts to do it?

Whats your edge?

 

[Lord Flasheart]

No, it is a good post and about the subject.

What is your staking plan? Adding contracts to winning trades? Isnt it called scaling in? Does it not work if your trading 1 contract? Does someone need atleast 3 contracts to do it?

Whats your edge?

About 75% of my trades I take 1/2 off at the first point after the market has gone my way but slowed down.I often then move the stop to b/e. it would be virtually impossible to workout what my true r/r is or what % of an edge I have