Probability, Statistics & Trading

[Perrington]

A lot of people appear to take a probability approach to trading. Iam going to question the validity of such a strategy by presenting some basic statistics that more or less everyone is aware of. I recently read Mark Douglas - Trading in the zone... hmmm Am I destined to be a loser because I dont agree with this gentleman's observations ? Ok, a lot of what he describes gets you thinking, and again a lot of what he says is downright confusing to say the least. I will say in his defence that some of his advice is positive in the way it can help you enjoy trading a bit more, and not worry so much. But what of his fundamental principle regarding thinking in probabilities ?

So, we have 2 possible directions, 1 is up the other is down (discounting sideways as you would break even if a market forever went sideways, which it doesn't do anyway) Now, this is a 50/50 chance of you being a winner or a loser based on the 2 possible outcomes. Another interesting figure that is "common knowledge" is that 10% of traders are successful (profitable) and 90% of traders are unsuccessful (crying to their bank managers).

Ok so we have a 50/50 chance of making money and yet only 10% do. Is it me, or do I sense a fundamental problem with regards to thinking in probabilities taking into consideration these statistics ? am I to assume that over the next century 90% will be winners and 10% losers to redress what has occurred over the previous century, or will the same scenario remain true ? I believe the latter to be the case, so therefore, I refute thinking in probabilites as a "required" strategy to becoming one of the 10%.

Is it a better strategy for the aspiring 10 percenter to focus on what appears to be market manipulation of a grand scale, trying to slot into the big player psychology state. Or should one brainwash themselves into believing that the market can be cracked by thinking in probabilities ?

Mark Douglas refers to minor and macro elements, minor being each and every individual tic being purely in the probability category defined by the number of possible variables that are in or about to participate in the market at any one time. Macro referring to a predictable trend occuring when observing these minor tics over a wider time frame. How do we jump from fractal chaos to observable trends if the market truely operates out of probabilities ? Is this not a fundamental contradiction, no matter how you look at it ? Is the more likely scenario being that the markets are manipulated giving the 10/90 ratio, and that most people dont have the capital or self discipline or downright devious mind to stay in the game long enough to figure out what is going on ?

 

[Roberto]

Interesting subject! I agree with many of the individual points you make, but I think probably not the "train of logic" holding them together here ...

A lot of people appear to take a probability approach to trading.

They do indeed. Some of them find it difficult to imagine another way, and although you make many good and interesting points, I'm not convinced that they necessarily argue against the validity of that approach in quite the way you seem to think they do.

I recently read Mark Douglas - Trading in the zone... hmmm Am I destined to be a loser because I dont agree with this gentleman's observations ?

Hope not, because that would cover me as well, I think ... although I find his books so badly and annoyingly written that I can't be certain of this.

Ok so we have a 50/50 chance of making money and yet only 10% do. Is it me, or do I sense a fundamental problem with regards to thinking in probabilities taking into consideration these statistics ?

Honestly? It's you. There's no fundamental problem there at all. Both of those statements could be completely true. (Personally I don't accept that either of them actually _is_ quite true, but that's not relevant to the point I'm making).

You could have something very close to a 50/50 chance of making money from any individual trade and still have a situation in which only 10% of people are _consistently_ successful, simply because people manage to lose a bit more from their bad trades than they win with their good ones. Just because the "direction of the result" is close to 50/50 doesn't mean that its _magnitude_ is, does it? And that's without even factoring in the cost of doing business, which most people delude themselves about as well ...

am I to assume that over the next century 90% will be winners and 10% losers to redress what has occurred over the previous century, or will the same scenario remain true ? I believe the latter to be the case, so therefore, I refute thinking in probabilites as a "required" strategy to becoming one of the 10%.

It's the "therefore" that's the place where the wheel comes off your argument, I think. I agree that the same will remain more or less true (I think it's understated, actually, but never mind that). I'm not sure whether "thinking in probabilities" is a requirement for success. For me, it was. But even if it isn't a requirement, I think there's still maybe a hole in your logic.

should one brainwash themselves into believing that the market can be cracked by thinking in probabilities ?

I'm certainly not brainwashed about it, but I found it probably the single most significant step in becoming consistently successful. Like many others, I _think_, although one never quite knows for sure, I admit.

How do we jump from fractal chaos to observable trends if the market truely operates out of probabilities ? Is this not a fundamental contradiction, no matter how you look at it ?

Not sure what you mean by "the market truly operates out of probabilities". The market operates how it operates. As Socco is rightly so fond of saying, it doesn't operate for your benefit or for my benefit.

Is the more likely scenario being that the markets are manipulated giving the 10/90 ratio

The markets are manipulated by inside interests, for sure. No question about that. But what does that have to do with producing the "10/90 ratio"? It might be a small part of it in that both are real things that really happen ... but if you're suggesting that one causes the other you'll need some pretty good evidence! IMHO more than 90% of "traders" would be losers even with no market manipulation at all. It's a bit like taking the zero off the roulette wheel. The casino still wins and cleans out all the punters, but a little bit more slowly.

most people dont have the capital or self discipline or downright devious mind to stay in the game long enough to figure out what is going on ?

Again, I agree entirely with the factual parts of your statement, but I don't see how they justify the rest of what you're saying at all ... sorry.

 

[ardhill]

So, we have 2 possible directions, 1 is up the other is down (discounting sideways as you would break even if a market forever went sideways, which it doesn't do anyway) Now, this is a 50/50 chance of you being a winner or a loser based on the 2 possible outcomes. Another interesting figure that is "common knowledge" is that 10% of traders are successful (profitable) and 90% of traders are unsuccessful (crying to their bank managers).

Ok so we have a 50/50 chance of making money and yet only 10% do. Is it me, or do I sense a fundamental problem with regards to thinking in probabilities taking into consideration these statistics ? ...

The 50 / 50 figure lacks something - time scale.

The markets don't go up or go down over a given set period of time, the big problem with most people, in my opinion, is trade management.

Look at the first chart below, this is a 5 Min chart of 'part' of today for MMM.

You can see that the stock opened, and went up - hooray for all those going long.

But then it came down - big time. Great for those who want to go short here.

If you were a swing trader then you may look at the daily chart and decide that it was going down today an take a short position - great you are in the money!

But look at the second chart now, this incorporates the rest of the day. (Actually the market hasn't closed yet as I take this, so things may still change dramatically.)

If you were a swing trader, your winnings are now all gone and you are losing a lot of money

A scalper may be in and out of this trade a number of time in a day, but when does he or she get into the trade and when does he or she get out?

We can correctly predict the direction many times, but you need to know when to get out when the move is over, even if it is over after only a 4 cent move.

Is the more likely scenario being that the markets are manipulated giving the 10/90 ratio, and that most people don't have the capital or self discipline or downright devious mind to stay in the game long enough to figure out what is going on ?

Are markets manipulated? Sure they are, but you don't have to be devious to profit from someone else's manipulation, just be able to judge market action. If I was a scalper trading this stock, I don't need to be able to think up manipulations, just be able to read the market and manage my trade.

When I trade I don't necessarily have to know WHY a market is moving in a certain direction - just be able to recognise it and get in and out at the right time.

 

[Glenn]

You are oversimplifying the statistics.
Your 50/50 ratio only applies to the entry.
So let's say you have a 50% chance of getting the entry right and it does indeed move in that direction.
But for how long and how far ? 1 second or 1 day ? 1 tick or 50 ticks ? Etc.
You will only make a profit if you get your Exit right. Just because you got the entry right does not guarantee anything.
AND
You will only make an overall profit if your total moneys from profitable trades is bigger than that from your losing trades.

Glenn

 

[ducati998]

333333333333333333

 

[TheBramble]

What bollox...

You don't have a 50/50 of being right and if the market goes sideways - you don't break even. Try it.

It's a waste of time even thinking about debating this.

Is it a better strategy for the aspiring 10 percenter to focus on what appears to be market manipulation of a grand scale, trying to slot into the big player psychology state. Or should one brainwash themselves into believing that the market can be cracked by thinking in probabilities ?

Better yet - find yourself something that actually works in the markets rather than debate 'this or that' high-level theoretical nonsense on these boards.

You're so far off the mark you're just wasting your time.

 

[Roberto]

What bollox... You don't have a 50/50 of being right and if the market goes sideways - you don't break even. Try it. It's a waste of time even thinking about debating this. Better yet - find yourself something that actually works in the markets rather than debate 'this or that' high-level theoretical nonsense on these boards. You're so far off the mark you're just wasting your time.

I'm going to have to start submitting my posts to you for precis, Brambly ... looks like we both said something pretty similar but I took my usual long-winded 9 or 10 paragraphs over it ...

 

[Perrington]

The Bramble, I do like a straight talker lol and thanks for all the other replies, ive found them interesting.

I dont feel that iam wasting my time investigating such topics as I genuinely have a healthy interest in all aspects of trading and just about anything related, however important they are with regards to actually making a profit. To me, finding an edge is equivalent to climbing inside the mind of the big players and anticipating their sentiment towards making money by manipulating through timing and volume, I feel basing a strategy on probabilities is bollox, thats exactly my original point I feel systems that produce x profit over y timeframe based on entering on a friday afternoon if the dow is up on the day and closing two weeks later etc is complete crap. What relevance does this kind of backtested strategy have on anything that happens in the future ?

For me the market is driven by sentiment, its all about psychology and trends that manisfest temporary group psychology along with periodic fundamentals such as bank annoucements for fx and corporate earnings for stocks in terms of finding some direction. The subject regarding probabilities and having to incorporate them into your way of thinking in order to become successful is irrelevant to me up until this point in my trading evolution. However iam interested in listening to other peoples viewpoints with regards to this and how thinking in probabilities helps them achieve consistant success.

 

[ZigZag]

Perrington
Im not sure if you understand the difference between an event(your example 50% being a winner) and the probability of an event. Your premise would be correct for the outcome of a coin toss- a random event.
Would you consider a football match a 50-50 bet ? Would the probability be 50-50 for Chelsea v Reading?
A frequency/event table is a prerequisite to calculate probabilities.

 

[Perrington]

I base my 50/50 upon the stage of your position at the point of exit. I discount break even, because if you continuously found a probability set-up that broke even you would be doing something extremely bizzare. The percentage of break even trades is negligible to a certain extent, but must also be accounted for, especially if thats what you regularly achieve with your set-ups.

The frequency/event probability set up can ofcourse be increased based on historical performance. My question is, what has this historical performance got to do with a market that is in a constant state of flux, bearing in mind the types of set-ups that are commonly used when defining a probability driven strategy ?

I agree, probability can bring profit. Does this form of creating strategies based on probability set-ups form the basis for intelligent investing ? Personally I dont think it does, does this matter ? depends on how much faith and confidence you have in randomness providing you with a so called edge. I feel that such strategies are quite high risk, based on the nature of ongoing flux, even though its been backtested to be giving a slight advantage.

Who are the people to benefit from probabilities ? The casino's the bookmaker's etc. How does this translate into trading ? the big players with the big balances ofcourse. So wealth is a key factor in determining how long you can continue with probability type set-ups as the bigger fish can go that step further than the little man. So is it prudent for the little man to adopt a probability strategy ? Personally, I think there are better ways to make the market an enjoyable experience, such as intermarket correlation signals. Commodities inverse to bonds etc, and this infact is one purpose of this thread. Iam attempting to gauge what the guys on this forum think about probabilities, and if they do infact use this as their main form of a trading strategy.

Another aspect of my thread is also to highlight the fact that the markets are virtually completely manipulated, this is my theory for why the ratio of winners to losers is roughly 10/90. Possibly this infact may be a reason why people choose probability set-ups because they dont want to do the hard work of intermaket analysis. Its much easier to test random probability set-ups as it essentially does not take any thorough scrutinisation of the interaction between all markets. However, does that mean that probability is the best way to trade, I dont think it is. Depends what your content with and how thirsty you are for knowledge at the end of the day.

 

[ZDO]

If you study Mark Douglas' older work you'll find he uses the words 'probability thinking' to encourage a trader to frame 'success' off following his rules across a statistically significant series of trades instead of worrying, desiring, hoping, fearing, stressing, whatever about 'this' particular trade. Armed with the requisite positive expectation - the trader can simply relax, not fear, not get caught in the equity ticks of this trade and thereby avoid the perceptual narrowings that set up the scared money decisions that take traders out of the game. His 'probability thinking' is ultimately about attitude, not thinking about odds or probabilities. hth

 

[Perrington]

I can agree with this aspect of Douglas's work. The one postitive I did gain from this book was a relaxed approach to trading, which is nice I must admit. However, I dont believe its possible to induce a state of probability thinking, without fundamentally applying this within your trading strategy. Self discipline does exactly the same thing as Douglas's rather drawn out approach.

 

[ZDO]

What do you mean by "However, I dont believe its possible to induce a state of probability thinking, without fundamentally applying this within your trading strategy." ??

 

[Perrington]

Your strategy is the sum total of your awareness in conjuntion with your psychological mental structure's is it not ? Or the best way you can express your knowledge.

 

[ZDO]

now I'm really confused... could you please rephrase that whole thought?
thx

 

[Perrington]

If you think in probabilities, you are going to express that thought within your strategy to a greater or lesser degree.

Douglas says it is a "requirement" to think in probabilities to be consistently successful, I disagree. He uses probabilities to create an elaborate set of frameworks to negate the impact of fear based decision's. I say you can do exactly the same thing purely by being disciplined with yourself.

 

[ZDO]

re: "you can do exactly the same thing purely by being disciplined with yourself."
Yes. Douglas might reply - all but a very few traders are so 'psychologically damaged' that they'll need a gradient path to get them to 'self - disiplined'. Getting to 'probability thinking' / 'this trade doesn't matter across the long haul' is one of his steps on that path. Are you saying 'what does he know?'? Are you saying being his student doesn't change ones' probabilities of 'winning' at all - that those who were going to succeed would have anyway and those that were going to fail would have anyway whether they were ever exposed to him or not ??

re: "Iam attempting to gauge what the guys on this forum think about probabilities, and if they do infact use this as their main form of a trading strategy." Very few traders are actually 'quants' . Very few know their 'odds' in any particular perceived setup, hardly any know their developing 'odds' after the setup has passed and as their results unfold. Many 'test', but few test the right things comprehensively enough to get real results across time remotely similar to the tests . Most do 'numberless' brain 'statistics' that include 'flashbacks', 'distorted thinking', 'random emotions', and 'reptilian inner charts' as factors in an ever changing algorithm... and the results are pure dumb serendipity or synaptic apocalypses ...that's not much of a starting point for unfettered disipline is it? Hope I didn't just kill this post ! Later...

 

[RUDEBOY]

Probability...? Market probability...? individual psychological probability...? Crowd/herd probability...? Or maybe a permutation of all or some or one, or other probability factors that have been missed here? This may sound like a bold statement, but, i personally believe we should not get it wrong that much, if at all, but we do, why? Discipline, who's discipline? How can an unknown be disciplined (money management can only take you so far), money management is good, but what probability is it based on? Even good money management has to be based on something! We could turn ourselves inside out on this one, there is no black and white. There is nothing on the markets that nobody else has seen, i think that maybe, and i do mean maybe, that we question ourselves that much, that the obviousness becomes the unobviousness. We are all guilty of this this i would guess?

 

[Perrington]

ZDO

Now its my turn to be confused, I aint got the slightest idea what your saying, Good thread this aint it But seriously, you do make some interesting points, inparticular your second point. As far as Douglas is concerned, I think anyone who is as psychologically distraught as you imply would probably have given up. Iam sorry, but I dont think trading in the zone is the answer to be quite frank about it, buts its still worth the read.

RUDEBOY

We probably could go round and round, there certainly doesn't seem to be many people who use intermarket correlation like myself here, and Iam intriqued as to why that is. The CRB inverse to commodities and bonds positive to stocks are excellent signal generators.

 

[RUDEBOY]

It's hard for me to talk about IMC with you, without sounding nitpicky, if you know what i mean? If i was to carry on talking about probabilty it would be about probabilty within a single market, price action only and no correlation, as i feel this is fundamental (no indicators or signals).