Why does TA not work (for you)?

[Fah Q]

Quote from Fah Q:

you cant test all ta, the combinations of what a trader can put together; the overlap of certain concepts and studies, are nearly infinite. testing can only be done on some of these.

i agree with fah

 

[logic_man]

Quote from marketsurfer:

I disagree. However, If this system is as you state--- why worry about points and fractions? Why not trade Binary Options and make 100% on each and every trade? if you are able to capture 1/2 of every directional move on average-- trading Binary Options should be an easy way to wealth.

In Western logic, it is not enough to simply say "I disagree" as if that were some kind of rebuttal. Not that you owe me an explanation for why you disagree, obviously.

I'll look into binary options. From what I've read, those are open to close-based, no? My trades don't necessarily start at the open and end at the close.

 

[pythontrader]

Quote from Fah Q:

i agree with fah

Good boy. Never disagree with him.

 

[logic_man]

Quote from Fah Q:

i agree with fah

Yes, the math of factorials makes it impossible to test everything.

 

[R. Raskolnikov]

Obviously. Anyone who says they've tested it all is deluding themselves. Too many combos of inputs.

Quote from Fah Q:


you cant test all ta, the combinations of what a trader can put together; the overlap of certain concepts and studies, are nearly infinite. testing can only be done on some of these.

 

[marketsurfer]

Quote from logic_man:

In .

I'll look into binary options. From what I've read, those are open to close-based, no? My trades don't necessarily start at the open and end at the close.

Try www.nadex.com it's a real regulated exchange their binaries are not closed and can be traded in and out of during the time frame. surf

 

[NoDoji]

Quote from SteveH:

Donna,

If you randomly pick trades from the population of valid trade setups from a positive expectancy system, it will not alter the long-term outcome of success. The only way it would be possible to do so (with negative results) is if the person has the subconscious ability (i.e., intuition) to consistently choose a greater majority of bad outcomes which, of course, is not random. The converse would hold as well: another person having the intuitive ability to non-randomly pick more of the winners than losers, exceeding the experienced results of someone taking all of the setups in the same time periods.

For example, no matter how hard anyone tries, s/he cannot pick and choose coin flips (from a fair coin) and alter the long-term probabilities of the experienced distribution results. The randomness has no reliance on the participant's timing of when to decide to choose a side. It is "built-in" to the fair coin itself.

So, there are two ways to neutrilize a person's intuition in taking trades from a positive expectancy system which could negatively impact the results: either take all of the trades OR flip a coin, you take the trade if it's heads, pass it it's tails (or vice verse). Then, you're assured of taking random entries from a winning system and your long-term expectancy results WILL mirror those of someone who took all of the trades.

Hi Steve!

It seems you're comparing two different things here: the random distribution of a 50/50 odds situation (coin toss) and a positive expectancy system (odds greater than 50%).

I would think that if your overall profitability is based on the average win rate of trading all valid setups in a particular time frame being greater than 50%, picking and choosing which trades to take from among valid setups can ruin the edge pretty easily. What if out of 20 setups in a given day, I choose six losing trades and two winning trades, instead of trading all 20 setups? What if I do this day after day? I've spoken with a few struggling traders who look to trade well-defined setups, but they don't trade all of them and they seem to consistently lose money.

TRO posted this neat little study that illustrates the problem with picking and choosing in the framework of a positive expectancy system with totally random distribution:

"Look, for example, at this elegant little experiment. A rat was put in a T-shaped maze with a few morsels of food placed on either the far right or left side of the enclosure. The placement of the food is randomly determined, but the dice is rigged: over the long run, the food was placed on the left side sixty per cent of the time. How did the rat respond? It quickly realized that the left side was more rewarding. As a result, it always went to the left, which resulted in a sixty percent success rate. The rat didn't strive for perfection. It didn't search for a Unified Theory of the T-shaped maze, or try to decipher the disorder. Instead, it accepted the inherent uncertainty of the reward and learned to settle for the best possible alternative.

The experiment was then repeated with Yale undergraduates. Unlike the rat, their swollen brains stubbornly searched for the elusive pattern that determined the placement of the reward. They made predictions and then tried to learn from their prediction errors. The problem was that there was nothing to predict: the randomness was real. Because the students refused to settle for a 60 percent success rate, they ended up with a 52 percent success rate. Although most of the students were convinced they were making progress towards identifying the underlying algorithm, they were actually being outsmarted by a rat."

P.S. If the trading system is based on a win rate close to 50% and money management is the edge, then picking and choosing probably wouldn't matter.

 

[logic_man]

Quote from NoDoji:

Hi Steve!

It seems you're comparing two different things here: the random distribution of a 50/50 odds situation (coin toss) and a positive expectancy system (odds greater than 50%).

I would think that if your overall profitability is based on the average win rate of trading all valid setups in a particular time frame being greater than 50%, picking and choosing which trades to take from among valid setups can ruin the edge pretty easily. What if out of 20 setups in a given day, I choose six losing trades and two winning trades, instead of trading all 20 setups? What if I do this day after day? I've spoken with a few struggling traders who look to trade well-defined setups, but they don't trade all of them and they seem to consistently lose money.

TRO posted this neat little study that illustrates the problem with picking and choosing in the framework of a positive expectancy system with totally random distribution:

"Look, for example, at this elegant little experiment. A rat was put in a T-shaped maze with a few morsels of food placed on either the far right or left side of the enclosure. The placement of the food is randomly determined, but the dice is rigged: over the long run, the food was placed on the left side sixty per cent of the time. How did the rat respond? It quickly realized that the left side was more rewarding. As a result, it always went to the left, which resulted in a sixty percent success rate. The rat didn't strive for perfection. It didn't search for a Unified Theory of the T-shaped maze, or try to decipher the disorder. Instead, it accepted the inherent uncertainty of the reward and learned to settle for the best possible alternative.

The experiment was then repeated with Yale undergraduates. Unlike the rat, their swollen brains stubbornly searched for the elusive pattern that determined the placement of the reward. They made predictions and then tried to learn from their prediction errors. The problem was that there was nothing to predict: the randomness was real. Because the students refused to settle for a 60 percent success rate, they ended up with a 52 percent success rate. Although most of the students were convinced they were making progress towards identifying the underlying algorithm, they were actually being outsmarted by a rat."

P.S. If the trading system is based on a win rate close to 50% and money management is the edge, then picking and choosing probably wouldn't matter.

I think there is a difference between that study and Steve's point, which is that in the study, the students didn't randomly choose to not participate in some trials and they didn't apply a random selection process like a coin flip to determine which way they went when they did choose.

Let's say you have 100 trade signals over the next 100 days, but you flip a coin to only end up taking 50 of them. Those 50 should, within some margin of error, have the same win percentage as the 100 would have. That's with all things being equal, but as the number of trades approached the thousands, I agree with Steve that the expectancy should remain consistent even if you only take 50% of those thousands of trades. There are a couple of things, at least, which could change the expectancy over the short run, but over the long run, I don't think it would change.

 

[hoodooman]

A quote from an old movie seem appropriate today and that is "get thee behind me Satan"

I reduced the size of my que today in order to improve my computer's performance.

I still had 26 signals and 25 were wins using the same two indicators. Here is the score so far since the thread was started.

DATE MKOP T= W/L WIN%
8-2 UP 1.5 27/3 90%
8-3 UP 1.5 18/1 95%
8-6 DWN 1.0 18/1 95%
8-7 UP 1.5 35/8 81%
8-8 DWN 1.5 47/1 98%
8-9 DWN 1.0 24/2 92%
8-10 DWN 1.0 25/1 96%

How do these numbers compare to Surf"s chimpanzee picks

 

[jack hershey]

Quote from marketsurfer:

I know for a fact it will not stay consistent over time. No fixed system is ever successful over a long enough time frame. It will need to be tweaked and altered to the changing market.

54 years for me.The technological advances have made information more easily handled.

I liked the copier and the PC when they showed up.