Market is being 100% random and being able to 100% predict the movement of price are 2 totally different things.

Now if you are asking if we can 100% predict the exact movement of prices, well, obviously not! That's as stupid as saying we will reach parity, or is it?

You see, the market is not 100% random because it involves human behavior. Individually, you cannot predict what the next person is going to do, he may hit the buy or sell button, yes this is random. But when you pull away from the noise level, prices are more than likely to stop or at least slow down in areas of support and resistance because of collective human behavior. Is it going to obey it 100% of the time? No. But the probability is much higher that you see prices head into zones of previous buying or selling, and prices will seem to stall before making its next decision. Also, macroeconomic forces will not let currency get out of balance because they need to purchase in baskets of currencies.

There is also a great study of the above with market profile and reversion to the mean.

Also, it is not 100% random because we all know the relationship between crude and the dollar. You cannot have 100% random when you have outside forces with direct relationship to its movement. 100% random means that the prices just move completely by itself without any relationship or disregard for past price movements, then that is just not true.

__________________
Life is a Gamble. You win some. You lose some. And life goes on.

Mr.Marketz;

if someone is making money from a method that suggests the market is random then its very difficult to convince him/her otherwise, and the opposite is true if someone is making money from a method that suggest the market goes through cycles then it will be difficult to convince him/her otherwise.

at some stage of my trading I was 100% convinced that the market is random, at a later stage I was 100% convinced that the market goes through cycles .

People stand for what they believe, it as a religious, and I don't blame them as long as the method makes money for them ....

but even random methods are subject to cycles , and they are biased by the trader or the rules used in the method

Scott is an FF member, he had some personal notes on his web site about this topic you can check them out

FOREX Statistical Research Center/Beyond the Random Walk - A New Market Paradigm

__________________

Hi Zupcon,

I do not realy understand the connection between your results of exams to the persistance of moves.

But I try to understand as good as possible and I will try to give you an aswer as close as possible to the tendency of persistance.

If we take a so called very strong move or strong cycle or strong trend.

The first impression whe have is that the majority of time the price is moving up. That was what I believed for a very long time.
But how unbelieveble this is the opposite is true. Less then 20% of the time the price is making a new high to its final destination.
That means that 80% of the time it is giving back from a high or just moving randomly. or not in the direction that is contributing to the persistance of the final destination.

To proof this lets place the data from a strong move or trend on a chart and on a drawdown chart. The 1st chart shows a clearl and very strong trend up because price goes from 500 to 3000 (that is a destination that is 600% higher then its starting point over only 300bars).

If we look to the 2nd chart that shows the drawdown or giving back from a previous high then you see clearly that there is a lot more action bellow the zero line then empty spaces (empty spaces is where the price is making new highs or moving to its final destination).

In the 3rd collom where you see a 1 that is a new high. In total there are 53 bars that lead to a new high of the 298 bars. That is only 17%!.

So even that this move is so strong in a certain direction (600% increase), 83% of the time price is just moving randomly around in a way that it has nothing to do with the move or contribution on the persistance or direction of that move.

Then it is easy to understand that if I tryed system on that persistance rule that they leaded to B/E results.

Not to mension if one would do tests on prices or markets where you have a lot less then a deviation of 600%.

So I have the tendency to say that even in very strong moves there is no such thing as the persistance rule.

I hope this aswer is in line with what you ment.

friendly regards...iGoR.

PS. You can take this spreadsheet and place whatever kind of move in it and you will these same results or even worse.

Attached Files

persistence.zip (185.6 KB, 18 views)

__________________
Success comes with knowledge. Knowledge comes with experience. Experience comes with time and hard work... (iGoR)
MT Intelligence - Stats for fxigor1

Hi Igor

Thanks for the spreadsheet, its quite an interesting analysis. If you want to look at this on a bar by bar basis, then of course you are quite right, the probability of the next bar being an up bar is not affected by the trend, nor is the run of consecutive up bars, or down bars affected by trend.

Its the SIZE of the overall moves thats important. If the market where truly random, very large moves (and very small moves) would happen very infrequently, but these large moves happen much more frequently than you might expect by random chance. Therefore, its not random chance thats sustaining these moves, its crowd psychology, and the aggregated behaviour of market participants. Regardless of the timeframe you choose, or the resolution of the time slice you take, the distribution in the size of any moves are always biased to the upside, and in that respect its like a weighted dice

Can we just agree that the market/price is mostly random? Not 100%.

In the market, very few things are 100%, certainly not randomness.

This I will agree on.

__________________
Life is a Gamble. You win some. You lose some. And life goes on.

Minime,

I know that we not have been on the best of terms in the past but If you allow me to give some proof that is based on facts and not on what we believe or based on how much money a system makes or what we read on some forum or internet articles.

The proof I'm going to give is based on 100% mathematical facts and not based what profits that I make or on what I learned or what I think or what I feel.

You can do with that proof whatever you want but I hope that you would think about it and look around to confirm it and maybe you want.

I know that you are quite "busy" with EA's.
Thats good because they give mathematical FACTS no assumptions.
All these EA's give 3 very important figures that most people don't realize.
The hitratio and the avg profit and avg loss. Those 3 numbers have a very important connection to each other that give proof on the discussion that we have here on this topic.

Look in every single rapport that you have or that you find here on this forum or on any other forum (if you do not have enough examples you can go to this site and look into the reports http://championship.mql4.com/2008/users ) (lookonly into the systems that have enough trades--EA's that have only 10-20 trades don't proof anything)

You will find ONLY 2 possibilitys:
1) Or that the hitrate is smaller then 50% but an avg winning trade that is bigger then the avg loosing trade.
2) Or a hitrate that is bigger then 50% but automaticly an avg winning trade that is smaller then an avg loosing trade.

That is not by coinsidence. On the contrary it proofs a very important element in trading and randomnes of the markets.

The 1st option proofs that their system has a hitrate less then randomnes (50%) but with the help of a solid MM plan they cut the losses short and let the winners run ( avg winner bigger then avg. losser). That means that their MM plan makes that they can be profitable with a system that scores less then a flip a coin strategy.

The 2nd option option proofs the opposite. They have hitrates more then 50%but only in a synthetic way.
If you investigate the stoploss they use and the T/P they use you will notice something very intersting. They use a much smaller T/P then the S/L they have (the oposite then the 1st option)
This way they make a synthetical hitrate not a natural hitrate.
Ex. They place you with a blindfold in a room that is 10 meters wide and they place you at 2 meters from a wall and they say you can start to move or walk. The chance that you will bounce against the wall that is 2 meters away from you is bigger then touching the wall that is 8 meters away from you. Or lets say that we flip a coin for price to go up or to go down. Heads is up so we go long. We place a S/L of 100pips and a T/P of only 20pips. Whatever price does or moves, it is logic that in most cases the price will hit the T/P sooner then the S/L.
If you do that on every trade then logic again that the hitratio will be bigger then 50%. But that is not due to the reliability of the signals but because of the MM plan that you force the hitrate into more then 50%.

Only if one can come up with an EA or tool that he can place on at least 10 different markets (because there are 100's of markets) and important the values and parameters need to stay the same at all time,and if on these 10 different markets the hitrate is bigger then 50% and with a S/L that is equal or smaller then the T/P ONLY then he proofs that his EA or tool can beat the "randomnes" of the market.
Or he proofs that his system or tool in all conditions has a a natural hitrate that is bigger then 50% and this way proofs that markets are not random.

I do not doubt the profits that they make or you make or others. But it is the only scientific way to give proof of beating the randomnes of the market.

I hope you see my post as a help and not me desperatly trying to convince you.
We sit here all together on this forum and a lot of people say what they feel and make claims based on assumptions or on what they read on the internet.
But proof in a scientific or mathematical way is still the most reliable way even if it is against what we feel or the profits we make.

friendly regards...iGoR

__________________
Success comes with knowledge. Knowledge comes with experience. Experience comes with time and hard work... (iGoR)
MT Intelligence - Stats for fxigor1

MiniMe

Can you give more details on how you determine your cycles and how you determine the dominate cycle out of all the cycles that may exist. Any other comments on cycles would be of interest, as well.

Keit

Randomness and the run of numbers;

I lived in LasVagas for a while and got to like BJ and the craps table when playing craps I devised a dealer win ratio which I called DAWF (dealer adjusted win factor). It was really good and keep you off tables were the dice were cold. Before everone starts telling me that the game of craps is a negitive game, I know and I also know which bets on the table that had the least odds again me. All I did was to test if the dice were running in favor of the players or the house.

Now let me tell you about one night at the Horse Shoe down town (at the time they toke over the Mint). I was playing at a table that was full and loud and the players were winning. Next to me was an old salty cowboy did not get his name even though we talked. I was using my DAWF keeping tract of it in my head. When it turned negitive I left the table headed to the bar for a drink I was ahead and done playing for the night. A few minutes later the older cowboy appeared next to me and order himself a drink too.

In talking we both agreed that the table turned cold, he asked my how I figured it out. When I was finished explaining my count, he said that that was interesting but too complicated for him. He then said some thing I later checked and verified that it worked. He said all he does is to see were the chips are. If more of the people at the table had there chips on the rail the table was hot as the players needed both their hand to place bets. If on the other hand most of the people heald on to their chips in hand the table was cold. They know they should leave but they just cant. So simple.


The run of numbers happens even in negitive out come games with random throws of fair dices. Now, price presistance not every bar will traval in the same direction but a run of bars in one direction happens even in random events.

Keit

edit: By the way, in BJ, I used a 3 point count with a side count of aces. Specized in double deck games only.

How do you explain "trend"

__________________
"A successful speculator bases no moves on what supposedly will happen but reacts instead to what does happen."

Max Gunther