Lose/Win Percentage for a Good Trading System

[pkfryer]

One such system might be to cut your losses, quickly (a small stoploss - rigidly adhered to) and let your profits run (a trailing stoploss - equally rigidly adhered to).

I think for day trading it works to have extremely tight stop losses, but this is difficult to accomplish with short term trading. In order to cancel out market noise of a trend or oscillation it is important to place stop losses sufficiently far outside the average daily range to not be hit unnecessarily. The only problem with this approach is that if the trade turns against you you have a lot more to lose, which means you can only trade smaller sizes to adhere to money management rules, therefore reducing profit considerably on winning trades.

The complication is that an otherwise winning trade could be stopped out at a loss unnecessarily if the stop loss is too close, meaning that you would have to have several times more winners than losers... or have winners that can go sufficiently far to outbalance the losing positions.

If we need to let winning positions ride in order to be profitable, we would have to increase the distance that stop losses are placed away from entry, bizzarly enough making it necessary for the system to be more accurate.

Maybe this is why 95% of traders lose, and why many traders have the notion that stop losses are for losers! (I dont think this, but I can imagine some people coming to this conclusion)

Have I got this wrong?

 

[TheBramble]

You haven't got it wrong PK, but it is a tradeoff.

Lots of (very) small losses with a few good wins - or - a lower number of larger losses with a greater number of lesser (on average) wins.

Many traders start off with too big (or no!) stoploss in place - hoping to catch the money when the trade turns back in their favour.

Frankly, if you have a system which identifies entry criteria which you are comfortable with - and the criteria are met - the trade should very quickly prove itself to you as a winner - or you kill it.

That's just my opinion. I have tried both extremes and it's all down to personal comfort.

 

[Matt321]

They say this system trades in any market every day(whether market is trending or consolidating) / I haven't tried it but the broad outline is as follows.
Ave price =( 2 x close) + high + low) / 4

The long trigger =(Average price x 2) -low
The short trigger= Average price x2) - High

If one position crosses the other trigger the positition is closed
The stats show the long trades winners 277 losers 211
short trades winners 299 losers 204

Stop losses should be minium of 3%

 

[growltiger]

Not true my friend. Those who have a 70%+ method just keep it quiet. Wouldn't you?

Mathematically, it is easy to create a 70% method, but without being very profitable. All you have to do is cut your winners very short. In fact, a system that sets a target stop that is less than one volatility measure away from the entry, and a stop loss that is three times that distance, can almost be guaranteed to generate a high percentage of wins and very likely an overall loss. Whether the system is actually profitable or not will depend on the actual behaviour of the market, of course, but the natural volatility of the market will ensure that the profit target is hit much more frequently than the stop loss (and if prices are more or less normallly distributed, this can be guaranteed). In system optimisation, it not infrequently seems to happen that the the "solution" is to have no stop loss combined with a tightish profit target; I ran one such system for the S&P for a while, but while the expectancy was reasonable, the drawdowns were not (unfortunate, but consistent with the backtesting statistics). And, of course, very tight targets and stops mean you trade a lot, so commissions are high relative to the bare expectancy, and slippage may be a problem too, relative to a tight target.

 

[Hooya]

a 50/50 trading system really is the best you can hope for. anyone who claims to have invented a system 70+ is fibbing!

two things to think about..70% strategy is possible but would likely come from an none mechanical system. and secondly a 70% wining strike rate doesnt necessarily mean you make money...it all comes down to how much you win when you win and ahow much you lose when you lose.

 

[dc2000]

a 50/50 trading system really is the best you can hope for. anyone who claims to have invented a system 70+ is fibbing!

Hmmm how about 89% win rate, av win 140, av loss 23.25 for last 12 months to April.
A myth, a legend or ?

 

[JonnyT]

If its for real then its fabulous.

JonnyT

 

[growltiger]

Expectancy

PKF,

Take a look at the attached spreadsheet, it should help to explain the points about money management the others are making?

HTH

Cheers

Mayfly

The spreadsheet is very interesting. But a simpler way may be to apply the expectancy formula that was mentioned by others in this thread. Because there are three independent variables (stop loss, target, win%) the area of interest needs to be looked at in three dimensions. The clearest way to do this is to run sensitivities of the capital growth to: (a) different combinations of gain and loss percentages on winning and losing trades respectively, holding constant the win/loss ratio; and (b) different combinations of gain percentage with different win ratios, holding the stop loss percentage constant.

It becomes clear, as people have said, that the different combinations make available whole families of winning and losing styles. And the winning styles definitely include styles where a majority of trades are losses, but also completely different styles where the trader aims for a high win ratio and accepts lower levels of profit. Taking 20 trades as the basis for comparison of the compounded returns, someone who set a stop loss of 10% and made 2*risk would do quite well on the basis that 40% of the trades were winners: before costs, the overall return would be 49%. The same return would be achieved, with the same stop loss, by someone who aimed to make 1*risk on 60% of his trades.


The analysis does not, of course, take any account of the influence of the chosen combinations of stop and target on the probability of winning. That is another story.

 

[pkfryer]

Hmmm how about 89% win rate, av win 140, av loss 23.25 for last 12 months to April.
A myth, a legend or ?

Whose success rate is that? sounds fantastic.

growltiger - very interesting. An optimising parameter for a system could also include size of trade related to win percentage. (Is this related to optimal f?) Would this encourage losers to gamble more the worst they are doing with the premise that by so doing they will perform better in the long run. Or have I got the wrong end of the stick?

I'm also wondering whether having a fixed profit target and exiting trades whenever this is achieved without trying to chase the 'perfect trade' has merit. Grab lots of small profits and restrict yourself to even smaller losses. The amount of times I've been in a marginally profitable trade waiting for it to do something interesting only for it to suddenly reverse and hit my stop. If I had merely taking the small profit each time I would be out ahead. Is this an approach that is worth any consideration or does it make becoming profitable very difficult. As winning trades are cut short in there tracks.

 

[TheBramble]

Hmmm how about 89% win rate, av win 140, av loss 23.25 for last 12 months to April.
A myth, a legend or ?

You haven't mentioned drawdown, but if the other figures are genuine, I calculate it wasn't more than 150.

As someone else once said, "Come on Sir, spill the beans!".

 

[Uncle]

It is the bottom line that counts- i.e. profit - the number of wins/losses are irrelevant. e.g. a few big losses and lots of small profits are counter productive. Conversely a few big wins can outweigh many small losses. The name of the game is money management..............STOP LOSS STOP LOSS STOP LOSS!!

John

 

[TheBramble]

Formula for Drawdown?

Playing around with some stats in relation to this thread and I have a possible correlation between Probability of a Win and Max Drawdown (in terms of average consecutive number of losses) which lends itself to a formula.

I don't keep stats on my trades (as mentioned above), but if anyone else could confirm/reject my hypothesis based on their own data:-

In any trading system, where Pw is the probability of a Win (and is expressed 0.01-1.00)

Average Number of Consecutive Losses = 20*(1-Pw).

Obviously, if you're using a fixed stoploss and if this formula is reasonably valid, it's possible to determine your max drawdown to within one std.dev.

 

[BBB]

I take it these are mechanical systems you are talking about. As a 'manual' trader, I'd say 70% is fairly good for a competent trader, but then I have the luxury of managing my trade to the way the market is behaving, rather than a one-size-fits-all approach - where I would be throwing myself to the mercy of 'the law of averages' at best.

 

[dc2000]

Hi Peeps a few more stats for ya.

For 2003 - 04
max drawdown was in march this year at 141 the usual is circa 50 points
max consecutive losses 1

For Jan 2002 - April 2003
win ratio 86%
Av win 237 points
Av loss 50.05 points
max consecutive losses 3 in 2002

yesterdays trade +176 points

 

[Matt321]

Tony I re-posted the formula below with some comments and updates ,using the data below would
your formula work out like so.

In any trading system, where Pw is the probability of a Win (and is expressed 0.01-1.00)

Average Number of Consecutive Losses = 20*(1-Pw).
20 (100% -55%)
20 (45%) = 9
I have probably made a pigs ear of it,as I don't fully understand it.

Trades
1 -4
2 4
3 6
4 -3
5 -2
6 4
7 5
8 3
9 -5
10 4
11 6
12 -5
13 -4
14 -5
15 4
16 -2
17 4
18 6
19 4
20 -5


Wins 11
Losses 9
Ratio 55


Average win 4.55
Average loss 3.88

Pw (Aw) - Pl (Al)

=0.55 x (4.55) - 0.45(3.88)
=2.50 - 1.75 = 0.75

Update:
If the last number turns out to be(ie -0.75) a minus then your system is losing,although the one above is positive.
There is no backtesting system I know of, you just get a calculator and your broker's statement.
The ratio is really a %,The way I did it was divide 20 trades into a 100= 5 then 11wins X 5 = 55 then 9 x 5 = 45 so thats your ratio or % of winners55%/losers45%.
You can also juggle figures around and play the "what if game" ie: if I had a closer stop
losses it would mean more losses but smaller ones etc the different combinations are endless,but it can help your tweak your system up.Try swopping the losers for winners
so you have 45% winners 55% losers the answer should be if i haven't got it wrong -2.15
bad news.

 

[TheBramble]

Average Number of Consecutive Losses = 20*(1-Pw) [where Pw is expressed 0.01-1.00]

So, if your trading system has a win rate of 65% the calculation will be:-

20 * (1 - .65) = 7

So trading a system with a 65% win rate you must expect on average to get strings of 7 consecutive losses.

You will get strings of less and greater numbers of consecutive losses, but this is the average you can expect .

Not only does this make it (slightly) more bearable when it occurs, it also means you can manage your position size accordingly.

Of course, the beloved 'Black Swan' may well float into the picture at any moment so it still makes sense to keep your max risk per trade within minuscule limits in relation to your trading capital.

BTW - This formula when applied to DC2000's trading systems stats (come on man, we all want to know...!) indicates he can expect on average, strings of just 3 consecutive losses.

 

[donaldduke]

So trading a system with a 65% win rate you must expect on average to get strings of 7 consecutive losses.

You will get strings of less and greater numbers of consecutive losses, but this is the average you can expect .

What does average mean??

I think on average if you have a 65% system 7 consecutive losses although certainly possible
are quite rare. Occuring less than once per 1000 trades.

 

[theknifemac]

I think that what Tony is looking at explains why people look at the percentage of wins as it is a proxy for the drawdowns that can be expected. Given two systems with the same outright profit I would rather take the one with the higher ratio as it would be easier to live with. Making a profit from a small number of winners might make it very difficult to trade through a drawdown waiting for the big winner.

Stew

 

[Matt321]

That must be quite hard to take more than 7 consecutive losses in a row when it happens.

Looking back over some the posts I was wondering what happen to "3rddawn" and "Jonny Ts" Forex trading system he was using.Is he laying on a beach somewhere sunning himself or did he go broke.

 

[TheBramble]

What does average mean??

I think on average if you have a 65% system 7 consecutive losses although certainly possible
are quite rare. Occuring less than once per 1000 trades.

DD - average wasn't the best term to use, but I'm struggling with how best to describe this.

In any trading run, there could of course be an infinite number of consecutive losses. Mathematically improbable, but possible.

By 'average' I meant that the most likely maximum number of consecutive losses you would encounter (with a trading system with a 65% win rate) would be 7.

You are correct - a relatively rare event, as it would occur approximately once every 1554 trades, but as you approach this number of trades - the probability of this sequence being produced increases.

But that wasn't my point.

My point was that this formula could be used as a basis for calculating expectation of maximum drawdown (expected consecutive losses) for any trading system with a known win rate.

In DC2000's system with a win rate of 89%, the most liekly number of consecutive losses is 3 and he can expect then to occur every 751 trades.