51%

[shadowninja]

Just thinking to myself recently. That whole notion of having a system that's got a 51% chance of leaving you with profit (R:R 1:1) is great in theory. And yes, if you're a casino with hundreds of thousands of punters every year, then it's perfect. However, in the real world, in Darren's world (that's me) where mistakes happen or there's a requirement to eat... or ****... and where Sod's law is a reality, 51% isn't enough. I need a system that's much higher so that when I make mistakes or there's a power outage or I need the toilet or get a phone call, my missing a trade in that instance won't mean the difference between profit and loss for the year! Additionally, if it's 51%, how do you really know it isn't 49%? And if it's 51%, you could have a seriously painful drawdown for 5 months.

So, if it's 51%, I'd suggest that you're better off getting a job.

And to be honest, this whole notion of "it's a probability game" is, frankly, nonsense.

 

[donaldduke]

I don't think anyone recommends 51% with a 1:1 ratio in trading. As after slippage, commisions and mistakes you aren't going to turn a profit.

Most wannabe traders are looking for at least a 66% win rate when using a 1:1 ratio.

The slimey snake oil vendors will tell you such systems exist with 70, 80 or even 90%+ win rates. These sorts of high percentages may be possible during hot streaks or certain market conditions, but not in the long run.

It seems to be much easier to try and make your winners bigger than your losers than to try and up your win rate.

eg 50% win rate with 1.5 : 1 win/loss ratio

This way even if your win rate turns out to be 45% you still make a profit.

It is still a probability game.

An automated trader can come very close to the expected win %.

A manual trader needs to factor in error tolerance. So your system might need an ideal win rate of 60% but after mistakes (human errors) you only end up with 50%.

So you still want a bigger win/loss ratio than 1:1, even if your win rate is higher than 50%, especially if you are manual.

As for painful draw downs, the level of pain will be dictated by your position size (5% risk per trade will give you much more painful draw downs than 0.5% risk per trade).

The draw down duration (5 months or whatever) will depend more on how many trades per day or week or month you make. All other things being equal a system that gives 10 trades per week will come out of draw downs 10 times quicker than a system that gives only 1 trade a week.

 

[the hare]

And to be honest, this whole notion of "it's a probability game" is, frankly, nonsense.

Its only nonsense if you just realised you dont actually have an edge

If the edge isnt big enough, then get more capital (its easier to get more money than improve expectancy, and funnily enough, the greater your returns the less likely you are to attract serious capital anyway).

 

[wallstreetwarrior87]

Just thinking to myself recently. That whole notion of having a system that's got a 51% chance of leaving you with profit (R:R 1:1) is great in theory. And yes, if you're a casino with hundreds of thousands of punters every year, then it's perfect. However, in the real world, in Darren's world (that's me) where mistakes happen or there's a requirement to eat... or ****... and where Sod's law is a reality, 51% isn't enough. I need a system that's much higher so that when I make mistakes or there's a power outage or I need the toilet or get a phone call, my missing a trade in that instance won't mean the difference between profit and loss for the year! Additionally, if it's 51%, how do you really know it isn't 49%? And if it's 51%, you could have a seriously painful drawdown for 5 months.

So, if it's 51%, I'd suggest that you're better off getting a job.

And to be honest, this whole notion of "it's a probability game" is, frankly, nonsense.

To me it appears that you are confusing trading with investing for the longer term. When investing money and only monitoring your investment maybe once or twice a month, then yes draw down will occur.

But if we are talking about trading then I would seriously worry or begin to question myself if I was having drawdown for more than 2 days. Yes human error occurs and that can mess a session up, and cost you a few points. But 2 or 3 days of this should conclude that one needs to do more work.

Im not sure why there is a casino analogy, surely you understand that there is more to this than flicking a coin?

If all thats attainable is a tiny edge I would rather do something else with my time and money.

 

[new_trader]

To me it appears that you are confusing trading with investing for the longer term. When investing money and only monitoring your investment maybe once or twice a month, then yes draw down will occur.

But if we are talking about trading then I would seriously worry or begin to question myself if I was having drawdown for more than 2 days. Yes human error occurs and that can mess a session up, and cost you a few points. But 2 or 3 days of this should conclude that one needs to do more work.

Im not sure why there is a casino analogy, surely you understand that there is more to this than flicking a coin?

If all thats attainable is a tiny edge I would rather do something else with my time and money.

Discretionary traders who rely on skill and judgement would see a long drawdown as a problem and conclude that more work needs to be done. But the majority of traders in this forum are 'system' traders with simple threshold based rules so for them, drawdowns are accepted and considered normal.

 

[wallstreetwarrior87]

I don't think anyone recommends 51% with a 1:1 ratio in trading. As after slippage, commisions and mistakes you aren't going to turn a profit.

Most wannabe traders are looking for at least a 66% win rate when using a 1:1 ratio.

The slimey snake oil vendors will tell you such systems exist with 70, 80 or even 90%+ win rates. These sorts of high percentages may be possible during hot streaks or certain market conditions, but not in the long run.

It seems to be much easier to try and make your winners bigger than your losers than to try and up your win rate.

eg 50% win rate with 1.5 : 1 win/loss ratio

This way even if your win rate turns out to be 45% you still make a profit.

It is still a probability game.

Maybe the problem is that trading is being treated like a game to most people, rather than a business?

I agree that probability is a massive part, due to the psychological nature of trading, but even when letting winners run at 51% strike rate you have to ask why only correct in your call 51% of the time?

I have highlighted your comment above:

You have shown that you understand that it is possible to achieve a high strike rate, you can not be aggressive in the market unless you have a high strike rate, so we need to work out how to achieve this high strike rate (ie create a large enough edge to make trading worthwhile).

Trading is not about taking punts, its about allocating capital into the markets during periods we have a grasp on. Obviously there is a lot more to it, but that is the end result, and when one is competent in that, they can achieve a high strike rate in the long run.

Afterall, a trader should only trade when they know the conditions, if not they may have better odds going to the casino.

 

[bbmac]

If you accept that a Trading edge is just a set of repeating circumstances that suggests that price has x % of going one way over the over based on historical precedent, then trading is all about probability and allying/optimising your money management to it. I argue strongly that having a high probability trading edge is easier to deal with than a lower probability edge although as DonaldDuke argues above some will find it easier hanhing on to the winning trades than upping their strike rate, but on balance, and particularly for inexperienced traders still developing the necessary psychological framework to trade their edge, I maintain that a higher strike rate trading edge is prefereable for all the reasons I have outlined in other parts of this forum.

If you don't accept that trading is probability based then what is it ?

G/L

 

[shadowninja]

Im not sure why there is a casino analogy, surely you understand that there is more to this than flicking a coin?

If all thats attainable is a tiny edge I would rather do something else with my time and money.

I think you're confusing what I think with what others say. The points of my post is that I'm stating it isn't a flicking coin so thanks for reading my post carefully. The 51% probability thing is what I hear a lot of people talking about when having an edge. And your conclusion is the same as what I wrote.

 

[shadowninja]

It seems to be much easier to try and make your winners bigger than your losers than to try and up your win rate.

eg 50% win rate with 1.5 : 1 win/loss ratio

This way even if your win rate turns out to be 45% you still make a profit.

It is still a probability game.

An automated trader can come very close to the expected win %.

A manual trader needs to factor in error tolerance. So your system might need an ideal win rate of 60% but after mistakes (human errors) you only end up with 50%.

So you still want a bigger win/loss ratio than 1:1, even if your win rate is higher than 50%, especially if you are manual.

The problem is that this is still theoretical stuff out of books and does not reflect how you can grow an account in the real world. It works great for systems traders like yourself but not for discretionary traders like me who don't go based on any signal as such but a degree of feel/experience.

In fact, it's also the stuff that vendors like to spout. Are you sure you're not a snake oil vendor?

 

[wallstreetwarrior87]

I think you're confusing what I think with what others say. The points of my post is that I'm stating it isn't a flicking coin so thanks for reading my post carefully. The 51% probability thing is what I hear a lot of people talking about when having an edge. And your conclusion is the same as what I wrote.

Yes sorry for misquoting you, I thought that were comparing casinos with the stockmarket, and talking about small edges.

As we came to the same conclusion, surely the next question or set of questions must be:

What is a strike rate? Does a target always have to be hit?
How do we get a high probability strike rate? (i Mean 75% + with r:r at least 1:1)
Is it possible to get a high strike rate? Because if we dont think it is possible then there is no need to go any further.
What do I personally need to do to get a high strike rate?
If I know what to do and Im not doing it, why am I not doing it?

Then you will know why so many fail to attain a high strike rate, then:

Finally, why be bothered with what others say/do?

If you are a discretionary trader with a high strike rate, then power cuts etc are not really an issue.

As new_trader pointed out, maybe this discussion is not relevant when some traders are system based.

 

[shadowninja]

Replies in bold:

As we came to the same conclusion, surely the next question or set of questions must be:

What is a strike rate? Does a target always have to be hit?
This is precisely the problem with saying 2:1 or whatever. What if you only see 1.8? Does a trader accept a loss for the sake of 3 pips??

How do we get a high probability strike rate? (i Mean 75% + with r:r at least 1:1)
Is it possible to get a high strike rate? Because if we dont think it is possible then there is no need to go any further.
And this is also an issue - if we accept the world isn't perfect, then we need to make sure how we trade caters for Acts of God.

What do I personally need to do to get a high strike rate?
If I know what to do and Im not doing it, why am I not doing it?
Then you will know why so many fail to attain a high strike rate, then:
Finally, why be bothered with what others say/do?
Indeed. It's just something that's been on my mind (as I wrote initially) and thought others ought to consider it.

If you are a discretionary trader with a high strike rate, then power cuts etc are not really an issue.
That's exactly it - a trader needs a high strike rate if they are manually taking trades.

As new_trader pointed out, maybe this discussion is not relevant when some traders are system based.
Yep. But then can the manual system traders guarantee to be around for every trade? What if they trade FX?





As many of you know, I've been dabbling in sports trading for several months and the anti-51% edge thought comes as a result of observing very efficient markets.

Brace yourself for the next outrageous statement: Anyone who claims the financial markets are efficient doesn't have a clue.

 

[donaldduke]

How do we get a high probability strike rate? (i Mean 75% + with r:r at least 1:1)

You wont get a win rate of 75%+ while using fixed 1:1 stops, i mean real ones you actually hold with your broker.

And so you will have to take some large losses every now and then.

 

[arabianights]

more to point 51% rate on 1:1 system is never going to be anything anyone with a brain is goign to trade in practice as there simply isn't that certainty...

 

[shadowninja]

You wont get a win rate of 75%+ while using fixed 1:1 stops, i mean real ones you actually hold with your broker.

And so you will have to take some large losses every now and then.

Indeedy.

 

[shadowninja]

more to point 51% rate on 1:1 system is never going to be anything anyone with a brain is goign to trade in practice as there simply isn't that certainty...

Ha! That's also been gnawing at my mind, too. Have to adapt one's trading style to suit the market so just how do you measure probability? It goes up and down depending on how quickly I adapt.

 

[tar]

There is a difference between strike rate and real odds , ie : if you trade system X and your strike rate is 62% this doesnt mean the real odds for this system is also 62% , it could be 49% but you managed to achieve 62% in the period in question .

 

[arabianights]

Ha! That's also been gnawing at my mind, too. Have to adapt one's trading style to suit the market so just how do you measure probability? It goes up and down depending on how quickly I adapt.

You could look at the series of returns and just use standard statistical analysis, null hypothethis being that system is not 51%+ reliable...

(and cause of that you are not goign to get a big enough sample size so you can't do it, in pracitce...)

 

[shadowninja]

There is a difference between strike rate and real odds , ie : if you trade system X and your strike rate is 62% this doesnt mean the real odds for this system is also 62% , it could be 49% but you managed to achieve 62% in the period in question .

That's the thing. if the real odds are that bad, expect some lean times ahead especially in Darren World where things never go according to plan.

 

[shadowninja]

You could look at the series of returns and just use standard statistical analysis, null hypothethis being that system is not 51%+ reliable...

(and cause of that you are not goign to get a big enough sample size so you can't do it, in pracitce...)

Yeah... I don't trade enough as I'm more a swinger. It'd work fine for my sports trading (100 matched bets yesterday!).

 

[arabianights]

how is that going? are you ready to start a journal here about it?

(ignore the ph challenge thread, that tldr)