71.2 % Win Rate

[Sergio77]

Quote from Ballpowder:

I see some people talk about the 2% rule of capital. Does that mean 50 different trades going? I could not keep track of that! 71% win rate would be amazing

It mean 2% of current bankroll risked per trade per system. if you have 20K for example and you trade two systems you can allocate 10K each and risk $200 per trade.

Quote from Ballpowder:

I have been using my BIC method. Balls Intuition Confidence. It will prob cost me some cash soon tho...

This is the best system I try to master too. When you trade like that it is very important to know how biased is the market in any direction. The methodology in this blog is useful I think (you do not need the specific software to follow it, you can develop yours along the same lines of analysis. VB+excel can do the trick)

 

[bwolinsky]

Quote from Huskeez:

Hey guys,

I ran a few backtests a while ago (but with limited flexibility in terms of parameters), and was able to find a strategy with a pretty high W/L % (based over 10 years od DOW data)


Now, do you guys think its possible to use the strategy which gave the W/L % and apply a reward/risk ratio of 3:1

(if I could interpret this into the backtest I would but unfortunately my skills are not quite there yet)

I will begin paper trading this , and see how I get on , but was pretty interested to see if this is how some other traders evolve their system.

Cheers Benji

The answer is no, Benji.

At a 3x, 45% Drawdown becomes:

(1-0.45)^3-1=83.36% drawdown, of which no possible way to even get to the featured 1:1 would remain...

And, yes, returning even 100x or 10,000% is probably not happening either due to the path dependence description in my answer below.

Quote from Swan Noir:

I urge you not to think of a $3 reward based on $1 of risk as a small reward ... it is not; and conceptualizing it that way might lead you to believe it is easy to acheive. You must also keep in mind that a test implies you are testing something and that something by definition must have parameters that have meaning to a working trader ... a guy at a screen trying to wring a profit from his days work. In order for a test to have meaning you must have the risk/reward predefinined. As NoDoji pointed out earlier a 50% trade when executed for a three point reward may well be a 40% trade if you shoot for four points. On the other side of the equation a stop change will also drasticaly impact your win %. Make sure the equation is complete in that it has all the elements needed to justify calling it a test. Right now you have an indicator you think gives you good entries.

Benji, this is a 2 profit factor with a 1:1 r:r and 50% win assumption regardless of past performance.

You'll need to leave that return as being path dependent, and very much so you can never do (1+140.xx)^3 which will leave almost certainly less than even the 140 since the drawdown's path dependence prevents it.

This numeric happens at greater than 40% drawdown where any typical return being so magnified and not annualizing greater than the percent drawdown has magnified trades such that during the greatest compounded period equity curves will be prevented from even making retracement to a 1x return.

If the return was already magnified by 3x this is as great as this DOW system will do.

In the future, recognize only APR's at least 1.5 to 2 times greater than the DD can be leveraged without worrying about drawdown path dependence.

 

[tradingjournals]

Quote from bwolinsky:

The answer is no, Benji.

At a 3x, 45% Drawdown becomes:

(1-0.45)^3-1=83.36% drawdown, of which no possible way to even get to the featured 1:1 would remain...

Could someone explain the calculations? Shouldn't 3x leverage & 45% drawdown lead to zero equity?

 

[Ghost of Cutten]

Quote from dbphoenix:

Nor does one have to concern himself with "positive expectancy" if he has a reasonably high win rate and profit to loss ratio.

Expectancy is derived entirely from win rate and profit to loss ratio. If the latter 2 are positive then the expectancy is positive also, by definition.

 

[tradingjournals]

Quote from Ghost of Cutten:

Expectancy is derived entirely from win rate and profit to loss ratio. If the latter 2 are positive then the expectancy is positive also, by definition.

Are you sure that is what you wanted to state? A win rate is always positive, and a ratio is always positive, yet expectancy can be negative.

 

[bashatrader]

Quote from tradingjournals:

Are you sure that is what you wanted to state? A win rate is always positive, and a ratio is always positive, yet expectancy can be negative.

You are correct. Check out this as I am lazy to right down the equations in detail but basically Expectancy = wR - (1-w) where w is the win rate and R is the average win to average loss ratio. So it can be negative if wR < 1-w or R < (1-w)/w

This is first grade stuff

 

[bashatrader]

Quote from Ghost of Cutten:

Expectancy is derived entirely from win rate and profit to loss ratio. If the latter 2 are positive then the expectancy is positive also, by definition.

Expectancy can be negative because if R < (1-w)/w. See above. I'm sure you wanted to say something else.

 

[murray t turtle]

Quote from Huskeez:

Hey guys,

I ran a few backtests a while ago (but with limited flexibility in terms of parameters), and was able to find a strategy with a pretty high W/L % (based over 10 years od DOW data)


Now, do you guys think its possible to use the strategy which gave the W/L % and apply a reward/risk ratio of 3:1

(if I could interpret this into the backtest I would but unfortunately my skills are not quite there yet)

I will begin paper trading this , and see how I get on , but was pretty interested to see if this is how some other traders evolve their system.

Cheers Benji

===================
Ben;
its murray t turtle[nickname]

1%] Best not to risk 10%;
early, 1% +/. Yes i know some experienced do more %

20] I dont do private mail much; i may make an exception for a Christian college woman, but that local,LOL, but true

58 or 60%] Futures magazine had a real impressive futures fund named'' 60% Drawdown'' LOL, but true. Available free for traders/investors....

10 years candlechart ]See my derivative/trend comments, gold futures section.For educational purposes only.....

murray t turtle

Also DOW, DIA are a good trend study, not that is as helpful as S+P,SPY.

Also make sure to study ALL 1990-1999/20008/9 trends, 1928,1929 trends, 1930 trends -1933

Wisdom is profitable to direct
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

 

[trader198]

there are three ways to solve a problem.

1. rational

2. intuitive

3. mixed

I found the worst is way 1 to do trading. the more you tested, the more you believe that is an edge.

trading is not science. it is an art. do not waste time on those backtests.

the market sometimes has patterns, but sometimes it does not.

even we have the most advanced equipment, we still do not when earthquake will come.

do not play statistic game, you will get burned.