How to research and verify trading ideas

[Smart Money]

Talon,

I think you're fighting a losing battle. You seem like a stand up guy that knows things and there are those of us who appreciate what you're saying. But many people are thinking there is a shortcut, or they lack the drive or know-how to research. They want to be spoon fed and will be pissed at you if you don't do that for them.

I get what you're saying because I've recently stumbled upon something that is working now, but I fumbled around for a while with a couple hundred trades and then one day looked at things differently.

Thats what I get from you...look at things differently, and do research.

I find it interesting that there are threads here that say T.A. doesn't work. If people really believe that, then why are they here in the first place? Of course it works. How many times have we seen a stock go to the moon and come crashing down? If you've witnessed that kind of behavior (or any kind of behavior), then find a way to capitalize on it. Look for stocks that were run way up, and then monitor them until they fall. Find ways to identify stocks that fell too far and are on the rebound., etc.
I get what you're saying.

There are a lot of us that aren't lazy who just need to be pointed in the right direction.

SM

 

[riskfreetrading]

Quote from talontrading:

How do you so completely miss the point. Get a f*cking pencil and paper. Go back 5 years and find the adds and drops from the index... calculate the returns and see what the strategy would return. This system is only a starting point for the discussion... somehow you are missing the point.

THE POINT OF THE THREAD IS THIS:
Given a trading idea... let's start with this one since it works... given a trading idea... how do you verify it and research it BEFORE you risk your money? That is the question of the day here.

When I first decided to do this trading idea it took an intern less than a week to crank out the numbers and stats on this. It's not complicated.



What do you find complicated? This is by far the most simple trading system I use so I guess I must be really confused about what a profitable system is. I didn't realize they could all be written on a napkin.... Crap... I'm going to have a find a new job.

And please define a "mathematical result"? What do you mean?

Let us address the question of a case (whether applicable to audience here or not). There is a gap in your thinking, which is rooted in confusing causality and correlation.

If you observed a phenomenon and got the statistics, you should not conclude that you have an edge, because an edge rests on a fundamental truth which you still have to discover. A mathematical result is an example of results that establishes the truth.

So the statistics you collect can be used ONLY to reject systems with no edge, they cannot confirm that a system has an edge.

Do you understand it a bit better now?

There is no escape from a theoretical result to establish an edge.

 

[riskfreetrading]

Quote from Smart Money:



Find ways to identify stocks that fell too far and are on the rebound., etc.

There are a lot of us that aren't lazy who just need to be pointed in the right direction.

SM

The guy who takes the opposite trade is also as convinced as you are, also has a system, etc.

It would be naive to think that because you did the analysis, it is the end of it. The opposition also does the analysis, or at least it is safe to assume that.

So all this work on I see something, as if the opposition did not see it is not convincing.

What one needs is something like: I see X,Y,Z, and the opposition did not see it because if they saw it this would be in contradiction with fact "W".

So one needs X, Y, Z and W, with the connecting logic. That is something I can rely one. The rest is just luck of the draw.

 

[talontrading]

There are many gaps in my thinking, but I assure you that issues of correlation vs causation and epistemology in general occupy a lot of my time... and in fact this thread was meant to address some of those issues.

I don't need a statistics 101 lesson... and anyway you are mistaken. in our testing structure every pattern, every system, every idea submitted for statistical analysis is first assumed to be random. then what we look for is evidence that is "inconsistent with randomness"... so when we find something that we are reasonably sure isn't random, that's a potential edge.

Do I understand it a bit better now? This has been my source of income for over a decade and I spend many hours every month doing statistical analysis so I think I understood it quite well to begin with.

I think by "theoretical result to establish an edge" you must mean a theoretical backtest? Like I make a rule set (let's say sell every 20 bar high) and then apply it to data to see what the result "would have been." Is that what you mean? If so... you're wrong... a test like that is only one of many and a potentially misleading one at that.

I'm a little surprised by the arrogance and condescension already... I know I am a real jerk and I certainly can be arrogant and confrontational (at best lol), but I'm thinking this experiment is going to work. Absolutely no reason I should deal with your attitude.... especially when your post is a little confused.

BTW... this system i put at the beginning of this thread I chose precisely because it DOES have a fundamental truth and an easy economic explanation for why it should work. Open your mind and think about it.

Quote from riskfreetrading:

Let us address the question of a case (whether applicable to audience here or not). There is a gap in your thinking, which is rooted in confusing causality and correlation.

If you observed a phenomenon and got the statistics, you should not conclude that you have an edge, because an edge rests on a fundamental truth which you still have to discover. A mathematical result is an example of results that establishes the truth.

So the statistics you collect can be used ONLY to reject systems with no edge, it cannot confirm that a system as an edge.

Do you understand it a bit better now?

There is no escape from a theoretical result to establish an edge.

 

[talontrading]

wrong on many counts.... wrong in general and wrong on specifics as it applies to this system.

i was sitting at dinner with some trader friends and someone threw out the question "why should we make money?" i mean... think about it... are we smarter than anyone else in the market? no. do we work harder? maybe... but people who are better capitalized than we are can hire 50 of us and they will work harder. are we big, smart money? no way... the biggest trader there still trades way under $100M... are we luckier? maybe to some extent but it doesn't explain longevity and consistency of results.

so... why?

well i think the answer is because we play a different game. you are wrong to assume someone takes the other side of your trade. someone may be buying a ton of stock for a pension fund and it doesnt matter at all to them that you buy it several times and sell it to them .25 higher... but that's maybe how you make your living.

so... your logic is not bad but it's incomplete. you are correct to assume there is competition but there can be overwhelming factors that make the issues you raise here irrelevant.

such is the case with this add/drops system btw.

Quote from riskfreetrading:

The guy who takes the opposite trade is also as convinced as you are, also has a system, etc.

It would be naive to think that because you did the analysis, it is the end of it. The opposition also does the analysis, or at least it is safe to assume that.

So all this work on I see something, as if the opposition did not see it is not convincing.

What one needs is something like: I see X,Y,Z, and the opposition did not see it because if they saw it this would be in contradiction with fact "W".

So one needs X, Y, Z and W, with the connecting logic. That is something I can rely one. The rest is just luck of the draw.

 

[wutang]

Talon, I've seen several potentially interesting threads die because the OP got dragged into arguments with people that derailed the thread and caused intelligent contributors to leave. Please don't let that happen. Ignore the pompous crap and continue the thread. Some of us are interested.

 

[lynx]

You mentioned that you thought you were ignored when you posted your S&P500 add/drop system. I just wanted to let you know that I was paying attention, and I'm sure many others were too.

I'm going to back test it soon and see how it works. I think it looks very promising because it doesn't appeal to a lot of people.

For example, "day traders" aren't going to like it because they can't trade it every day. Part timers won't like it because it can't be traded on the particular days when they happen to be free. And I suspect that the companies being added/dropped don't have enough volume to be effectively traded by institutional traders.

I think that it's plausible that this edge persists because the method is not practical or attractive for a large percentage of market participants.

I've done a ton of research on indicator based systems and like you I have a really hard time finding any that are worthwhile. So I'm looking forward to this thread and am hoping to pick up some alternate perspectives.

 

[sosueme]

In answer to your question ...
"How to research and verify trading ideas"

Trade the idea, preferably using more than 75% of your account.

If your account balance increases then you have verified your trading idea is good.

If your account balance decreases then you have verified your trading idea is bad.

See trading does not need to be complicated.

 

[Kassz007]

Quote from lynx:

You mentioned that you thought you were ignored when you posted your S&P500 add/drop system. I just wanted to let you know that I was paying attention, and I'm sure many others were too.

I'm going to back test it soon and see how it works. I think it looks very promising because it doesn't appeal to a lot of people.

For example, "day traders" aren't going to like it because they can't trade it every day. Part timers won't like it because it can't be traded on the particular days when they happen to be free. And I suspect that the companies being added/dropped don't have enough volume to be effectively traded by institutional traders.

I think that it's plausible that this edge persists because the method is not practical or attractive for a large percentage of market participants.

I've done a ton of research on indicator based systems and like you I have a really hard time finding any that are worthwhile. So I'm looking forward to this thread and am hoping to pick up some alternate perspectives.

I think you are correct. The more people that are disinterested in this system, the more interesting it becomes!

 

[Dacamic]

Quote from talontrading:
... i think the answer is because we play a different game. you are wrong to assume someone takes the other side of your trade. someone may be buying a ton of stock for a pension fund and it doesnt matter at all to them that you buy it several times and sell it to them .25 higher... but that's maybe how you make your living.

I believe this is a valid assumption and worth keeping in mind.