money management to make a lot of profit with only 60% win rate - page 2

New comment
 

You know in reality you are already in a big loss for all the time you wasted.

I can remember very well, back in the early day's when i started to trade i didn't have as much money as i now have.

I was just started trading 0.01 lots a time and things moved slowly, and after a while i went to 0.05 Lot's.

That took the longest, that first part.

It was difficult but you learn fast.

You have wasted so much time and money.

If you think you got something then put it to market it is the only way.

Nobody here can tell you the end result.

Only the feedback from market will tell you right from wrong.

No backtest is capable of that too.

 
Your idea might work with no leverage. But as long as you are buying on margin in any trade the amount of capital that is at risk is always your entire account. 

Using only 1% of your margin as opposed to 50% only means that the probability that you get a margin call is 1% and not 50% 

That's why you need to always start out with small lots and work your way up as has been mentioned above. I think it was Buffet who said markets are mechanism of transferring money from the impatient to the patient.
 
Zee Zhou Ma:

If you are wrong at yr guesses, will you cut loss or hold?

How do u decide how much to trade at each trade, assuming u know the win probability? 

Are you trading the same direction in this sequence, or different, sometimes bullish, and sometimes bearish?

i think eventually this idea will evolve into martingale, grid strategy or something hybrid. Not really silly, but u need to keep improving on the model.

if you assume this model works, you have to assume each new trade is dependent somehow from the past trades, because if you use a coin toss for example, where each new event is independent, you can see clearly this doesn't work, i tried an you can't use these blocks for coin toss, but for trading on backtests, it does look good, anyway, if you have a good idea to improve it, i''ll be happy to see it
 

"It is possible to keep losing 4 hrs straight because you still think the trend has not changed. 10 consecutive losses are possuble, beyond your statistical boxes."

Yes that is the basic concept of this system, i have several strategy blocks, and in none of them i have more than x miss trades, in the example it was 2 trades. If i have for example 10 blocks of these on my backtest, i think it's very plausible to say that in the future the proportion of not more than 2 miss trades will happen each block.


60% win rate - 35 trades

o x x o | o o x o | o x o o | x x o o| x o o o | x o o x | o o x o|  x o o x | x o o

here you can see that for each block it was not more than 2 miss trades, so i assume that in the future it will remain.

But that's the problem, if each event is independent, those blocks does not work. If you get a coin flip

50% ( so close to 60%)- 35 trades. I used random.org coin flip site

 heads are win=0:

0xx0x0x0xx0x00xxx0000xxxxxxxx0x0x00 - that was the result on a coin flip for 35 trades.

As you can see i can't use favorably those blocks. So that's the reason i regret this  topic, the idea is not really useful, only if you compare trading with something else than traditional probabilities

 
Stephen Njuki:
Your idea might work with no leverage. But as long as you are buying on margin in any trade the amount of capital that is at risk is always your entire account. 

Using only 1% of your margin as opposed to 50% only means that the probability that you get a margin call is 1% and not 50% 

That's why you need to always start out with small lots and work your way up as has been mentioned above. I think it was Buffet who said markets are mechanism of transferring money from the impatient to the patient.
I agree.

Time is another important factor here. Coz even with a tight SL and a reward to risk ratio of 3:1 you could spend a long time getting a meaningful yield.
 

So as far as I have read it depends on the maximum size of capital one is going to risk at a 'promising' (2 out of 4) trade.

Of course there cannot be a real promising trade without knowing the background (chart), but if you want to test it..: A bigger step would be to have it coded as a whole dashboard: you get a signal for each pair when, as you said, a promising trade will appear (O - O - x ->signal)

You use the EA, but in this case you let it take bigger leverage (or Only take trades at all) once 2 or 3 consecutive trades in one block were losers.

And letting it automatically trade, with low balance, would show you how it goes.


One more thing: What trading and time have shown me, backtests show to a high degree the real win/lose ratio for the future.

 
Andreas Bauer:

Of course there cannot be a real promising trade without knowing the background (chart),

There can be a promising trade even without a chart.

bool direction=MathRand()%2;//0=short,1=long
 
Andreas Bauer:

So as far as I have read it depends on the maximum size of capital one is going to risk at a 'promising' (2 out of 4) trade.

Of course there cannot be a real promising trade without knowing the background (chart), but if you want to test it..: A bigger step would be to have it coded as a whole dashboard: you get a signal for each pair when, as you said, a promising trade will appear (O - O - x ->signal)

You use the EA, but in this case you let it take bigger leverage (or Only take trades at all) once 2 or 3 consecutive trades in one block were losers.

And letting it automatically trade, with low balance, would show you how it goes.


One more thing: What trading and time have shown me, backtests show to a high degree the real win/lose ratio for the future.

i see you really understood the concept, but to code a whole dashboard seems very difficult, because for each scenario, it would be another type of trading


So if my strategy the blocks were 4, so 2 maximum loss each block, so i calculate the worst case scenario, which is:

x x o o             

x o x o               

x o o x              

o x x o

o x o x 

o o x x

( that's the worst case scenario, for 2 consecutive miss trades each block)

So based on that knowledge, i can make profit:

investment 1000, trades invest 50% of the risky capital

x |x o o  -                2 trade - 500             3 trade - 1000 (recover)     4 trade (1500) - profit: 500

x |o x o -                 2 trade - 1000           3 trade 500                      4 trade (you know it will be a win, because you don't have 3 wrong trades each block) - profit: +0 undefined if you reinvest more than initial investment

x |o o x  -                2 trade - 1000           3 trade 1500                   4 trade ( you don't make the 4 trade, because the 2 wrong trade may happen here) - profit: 500

o |x x o  - here you don't trade, because the 1 trade was a win

o |x o x  -here you don't trade, because the 1 trade was a win

o |o x x-


So i disagree that you can't make money fast, i think it's the opposite, you can make money fast because some trades you have almost certain that you system will win.

 

How can you agree or disagree to something you haven't tried ?

That is called a fatality.

 

This i called soft martginale and is reasonable if 1st position size is low and the number of steps is restricted. Profitability is not higher than using linear position sizes.

123
New comment
Reason: