The market is random

dtrader98 · Oct 5, 2009

Quote from wutang:

Care to give any clues, beyond "fat tails", etc? That's an honest question from someone searching for the right direction(s) to pursue.

I actually gave a never before published (to my knowledge) mechanical edge away in the Bar by Bar thread. The purpose was to help Swan Noir, and the argument was to show how commissions swamp out small net positive edges over time. Although, 1) You could increase position size if you can afford it, and swamp out the negative commission bias. 2) It gives you an idea of how a mechanical trader might think.

If anyone were to run a standard 'test' for randomness on the raw data, it would qualify as random, yet the expectations results I showed had a non-random edge over the data points. The idea is that a set of data can be perfectly random from a statistical point of view, but it is possible to process it in such a way, that the overall expectation with large numbers is better than chance. Of course slippage and commissions can and will work against you if you are under-capitalized.

Now that I described it as an edge, I guess everyone will run over to look (even though the original intent was to point out pitfalls). Unfortunately, most here want the fish, but don't want to learn how to fish. I've offered plenty of collaborative opportunities to discover things in the past, and yet, virtually no one wants to do any work to learn. Go figure.

P.S. I'm not a fan of Bar by Bar, for the record (not that it matters much), but I respect the author for publishing a book.

Banff01 · Oct 5, 2009

TheMan · Oct 5, 2009

Quote from college_trad3r:

What I have realized in the past weeks is that the market is truly random. It has been an epiphany. I dreamt about this at first. I was scared: what if the market is truly random? What if all the effort you put in can come unrewarded? I expounded on this idea and came up with the following: that the market is truly random.

What this means is that there is no winning trading strategy.

You might think, how is this possible? Well it is, because the way people make money is 100% by luck. The luck decides whether your strategy will be profitable for a few years, before the market changes. Each year there are different strategies that work, because of luck.
You might think then how did Buffet made his money? Pure luck! He is an outlier of statistics. Just like the 1% out of traders who are consistently profitable, they are profitable by luck! It is only by chance their strategies seem to work. :eek:

Now, is it possible to put luck in your favor? YES IT IS. By testing out alot of different strategies you can come up with one that seems to work. It is your job to exploit it until your luck changes. However finding strategies that work is mostly dependent on luck. You might dabble with price action strategies, and find out they don't work this year, though they might be the best strategy around in the long run.
Trading is more about luck than hard work It is not a meritocracy that people think it is. Most people ignore this cold hard fact.

This is the hard and pure truth i have finally uncovered about the markets. See what you will do about it. I can only give you the tools but you have to do the fishing yourself.

why do all of your threads have to do with failing , the impossible and cant be done?

your entire post is complete bullshit

just because you cannot do it does not mean that it cannot be done

with the attitude you have about trading, i would stop now and give up---- it will save your sanity

bec you will never make it in this business

dtrader98 · Oct 5, 2009

Allegory of the cave theme. And that was an awesome video! Thanks for sharing.

rolextrader · Oct 5, 2009

The harder I work, the luckier I get.

bone · Oct 5, 2009

Wow - how the hell have I managed to make money 17 consecutive years in a row? Random luck I guess.

MandelbrotSet · Oct 5, 2009

Quote from FB123:

Yes, it is. The "secret" (which isn't much of a secret) is understanding the transition between range-bound and trending states. It has been written about in tons of places. It is too complicated to program in an easy way into a computer, but if you can read price action then you can do it. In fact, that is exactly what reading price action IS in many respects ...

+4 (as in GRANDSLAM)

***

'have no idea who this guy is (assuming FB123 is a guy

), but he's sitting here giving away the Keys to the Kingdom ... and they say that there is no more value on ET!

wave · Oct 5, 2009

FB123 hit it on the money.

trend to range, vice versa.

if, then, else and sometimes an elseif.

think overlapping straddles.

net profit is all you care about.

MandelbrotSet · Oct 5, 2009

Quote from wave:

FB123 hit it on the money.

trend to range, vice versa.

if, then, else and sometimes an elseif.

think overlapping straddles.

net profit is all you care about.

btw, great counter-trend call on the ES Thread today wave.

nitro · Oct 5, 2009

Quote from dtrader98:

...

If anyone were to run a standard 'test' for randomness on the raw data, it would qualify as random, yet the expectations results I showed had a non-random edge over the data points. The idea is that a set of data can be perfectly random from a statistical point of view, but it is possible to process it in such a way, that the overall expectation with large numbers is better than chance...

I agree and this is clearly correct, but I don't know why people find this to be some sort of revelation.

This stuff goes under the heading of seperating signal from noise, and what is signal to one, is noise to another. Let me give an example. Take a typical music CD. Say you gave it to someone who had never seen one or does not understand what it is. He may be able to somehow look at the data stored on it, run statistical tests, etc, and he will be completely dumbfounded by it and decide to use it as a freesbie. I come along, put the CD in a CD player, and wham, out comes music that has nearly 100% structure!

The point is that 99% of trading is seperating signal from noise. When you finally have the market equivalent of the "CD player", markets no longer look random.