Fooled by Taleb

Quote from jem:

Winning wimbledon does not and can not happen by chance.
Try playing a mini tour player or a recently off the tour player.
Many of them can be found teaching tennis. Could you or any body beat them 7 times in a row. Now make each player much better. There are only a few highly skilled people in the word who can win Wimbledon using their highly developed skills at close to their highest potential.


That random things happen may happen along the way like a particular opponent losing early... does not turn a wimbledon victory into something indistinguishable from chance.
More...

Did you understand what dandxg said here? (an important comment) If ever in history a favorite (that is should have won based on skill measurements) seed actually lost to an inferior player, then in that particular case, wimbledon WAS won by chance wasn't it? One negative proves the supposition wrong and infinite cases can't prove the supposition correct. AKA black swan. So why play wimbledon at all then. We could just compare skills and save airfare. And in the trading case, we could just go on the honor system and mail our cheques in to MAESTRO for how much we would have lost betting against him/her!

Trading is like Wimbledon that the best skills will win most of the time, but luck plays a part even in the hands of the best skilled trader in the world.
 
Could you win Wimbledon? If you can't distinguish Wimbledon victories from chance... Why not?

Winning wimbledon takes extreme skill. I suppose an improperly tightened net cord could make a difference and cause an opponent to win an extremely close match... but the person still had to use extreme skill to get to match point against an opponent... and

To get to that final that person had to show extreme skill. Luck may vary the payout... but luck did not get him the opportunity to get to the final. He had to have great skill.




Quote from StarDust9182:

Did you understand what dandxg said here? (an important comment) If ever in history a favorite (that is should have won based on skill measurements) seed actually lost to an inferior player, then in that particular case, wimbledon WAS won by chance wasn't it? One negative proves the supposition wrong and infinite cases can't prove the supposition correct. AKA black swan. So why play wimbledon at all then. We could just compare skills and save airfare. And in the trading case, we could just go on the honor system and mail our cheques in to MAESTRO for how much we would have lost betting against him/her!

Trading is like Wimbledon that the best skills will win most of the time, but luck plays a part even in the hands of the best skilled trader in the world.
More...
 
its not about winning vs. losing trades

its about how much money u lose vs. how much money u win


winning and losing trades do come in a random distribution

u should preserve your capital , the money comes in sooner or later


even if u have a 70% hit rate , doesent mean u win 7 out of 10
trades .. the edge might kick in after the 50st trade.. the question
is would u still have the capital then ?


and if so , would u be able to milk it properly ..?

like the all to well known story of the blackjack card counters ,
they had a proven edge , but they couldnt not say that they win
every hand wich is dealt , they needed to bet small till the edge
kicked in, and if it kicked in , they startet to bet big...
 
Quote from jack hershey:

I have spent the time minimizing luck and eliminating prediction.

Your d (luck) / d (prediction) differential is different than the original d (predictiond) / d (luck). Interchanging the paradigm's independent variable and dependent variables changes all.

For me the future moves into the present and how finely you deal with it mathematically, there is a fixed limit.

The rate of events is the overall paradigm measure.

You changed the order for those who still deal with the two variables.

I feel that one disappears and the other is unnecessary all based on deduction.

Your forward progress will ultimately lead to a new paradigm.

A logicoligist must look at the facts and how they unfold. All markets step towards events. This is observed by looking repeatedly at the least change per combination of variables.

Context and change of context leads to events. The order of events in trends is irreversable It is unilateral at all times.)

So deducing the system of operation of the market does two things:

1. Luck is eliminated, and

2. It becomes very noticable at some point that prediction is NOT a facet of looking repeatedly t the least change per combination of variables.

For most, I appear to be talking in a foreign language. As I read your post, I see you profer a huge step in observing a system. I feel that the problem solution anyone faces comes down to being able to see the finest pieces of a system and being able to always use a complete solution that exists for following the order of least changes in variables that lead to an order of events for all types of trends (there are four types of trends, mathematically speaking).

So good luck in your pursuit of prediction.
More...

Dude, you are speaking my language - congratulations on being totally awesome. I get you.

Don't ever change.
 
Quote from jack hershey:
I have spent the time minimizing luck and eliminating prediction.

Your d (luck) / d (prediction) differential is different than the original d (predictiond) / d (luck). Interchanging the paradigm's independent variable and dependent variables changes all.

For me the future moves into the present and how finely you deal with it mathematically, there is a fixed limit.

The rate of events is the overall paradigm measure.

You changed the order for those who still deal with the two variables.

I feel that one disappears and the other is unnecessary all based on deduction.

Your forward progress will ultimately lead to a new paradigm.

A logicoligist must look at the facts and how they unfold. All markets step towards events. This is observed by looking repeatedly at the least change per combination of variables.

Context and change of context leads to events. The order of events in trends is irreversable It is unilateral at all times.)

So deducing the system of operation of the market does two things:

1. Luck is eliminated, and

2. It becomes very noticable at some point that prediction is NOT a facet of looking repeatedly t the least change per combination of variables.

For most, I appear to be talking in a foreign language. As I read your post, I see you profer a huge step in observing a system. I feel that the problem solution anyone faces comes down to being able to see the finest pieces of a system and being able to always use a complete solution that exists for following the order of least changes in variables that lead to an order of events for all types of trends (there are four types of trends, mathematically speaking).

So good luck in your pursuit of prediction.
More...
Does... Not... Compute...

Brain... Exploding... Must... Hold on... To... Sanity... Please... Help... To... Stop... Cannot...
 
Perhaps jack having proactively gone out of his way to minimize luck, would care to donate the chunk of luck that he's not using, I will quite happily take it off his hands, hey I'll even pay a generous rate for it.

I've spent the last 10 years learning to capitalize on good luck, but then I would wouldn't I

:p
 
I would argue that the concept of luck doesn't really apply in markets and trading, given that markets are always changing and that each real-time setup has a virtually unique set of conditions. The mirror-image of this would be the concept of skill in a game of pure chance with a fixed set of conditions, say electronic roulette.

One would need to drill right down to the exact impetus behind each trade a trader places to determine whether his results were truly "lucky" (aka a temporary anomaly that masked an actual lack of edge) or instead born of skill/information/edge. But, take any "true" edge that eventually deteriorates over time, and take a trader who was the very first to discover this edge yet cannot change his method to adapt to market shifts over time -- his results be virtually indistinguishable from a temporarily "lucky" trader who was just betting on hunches and had a good streak. Both will end up broke over the long run. LTCM ended up broke yet were their initial successes pure luck? And what about all those retired market "wizards"? The concept of luck just doesn't apply in practice. In this light, it's not surprising that Livermore once said no man can ever beat the general market over time - perhaps he meant that eventually the very thing which grants a trader success in one period will be the exact source of his downfall later on? What place does "luck" have there?
 
Quote from jem:

how could I resist.


2. of course I would desire to see your answer... you are the respected poster I was referencing. (a few respected posters have already contributed)

More...

Ok, I will contribute.

Let’s do a mental experiment. I would flip a coin and decide whether I will be Long or Short one contract of ES for 1 point. If I lose 1 point or win 1 point I’ll exit and flip my coin again. So, assuming my coin is fair and I pay no commissions then I would expect a natural “drift” of my equity line of D = sqrroot(2NG/pi), where N is the number of flips and G - is the ES point value. If after N flips the absolute value of my gain/loss is less or equal to D then I cannot consider my success (failure) anything but a natural drift of a random walk. If after P trials of N coin tosses my average absolute gain/loss is greater than D then my confidence of me playing better (or worse) than a normally distributed random walk is directly proportional to P. That is all to it.

Cheers,

MAESTRO

P.S.

You can easily modify this experiment to answer your own questions. You can modulate the walk with your average gain/loss, you can add commissions etc. However, the experiment itself is a universal measuring stick to assess the “skew” of your personal win/losses distribution compared to a normally distributed random walk.
 
Quote from MAESTRO:

Ok, I will contribute.

Let’s do a mental experiment. I would flip a coin and decide whether I will be Long or Short one contract of ES for 1 point. If I lose 1 point or win 1 point I’ll exit and flip my coin again. So, assuming my coin is fair and I pay no commissions then I would expect a natural “drift” of my equity line of D = sqrroot(2NG/pi), where N is the number of flips and G - is the ES point value. If after N flips the absolute value of my gain/loss is less or equal to D then I cannot consider my success (failure) anything but a natural drift of a random walk. If after P trials of N coin tosses my average absolute gain/loss is greater than D then my confidence of me playing better (or worse) than a normally distributed random walk is directly proportional to P. That is all to it.

Cheers,

MAESTRO

P.S.

You can easily modify this experiment to answer your own questions. You can modulate the walk with your average gain/loss, you can add commissions etc. However, the experiment itself is a universal measuring stick to assess the “skew” of your personal win/losses distribution compared to a normally distributed random walk.
More...

It doesn't matter how many times you flip a coin or what the small drift is, the expectation of heads is still 50/50.
A trader has some advantages, although small they can be significant. For instance
1.when the bet is made
2. how large the bet is
3. when the exit is made

yet we can safely assume the " House " will still be ahead if a number of people play. The lucky and clever ones will probably be up. The unlucky and average Joes down
 
You must log in or register to reply here.
Back
Top