Looking for "edge" ideas...

I read the attached PDF, thank you! I would venture to guess that this is a document you wrote! I have read many of your posts and this attached PDF certainly sounds like a summary of the message you are always trying to deliver.

I am trying to do exactly what you say, find a price pattern or market behavior that offers a level of predictability. I just wonder if it can be as simple as reacting to an opening gap, and how I can go about testing for this. Manually testing ideas would be exhaustive, and minor adjustments would cause me to have to start all over again, and hence why software would help immensely. But I fully appreciate that doing this manually will teach me more.

Is it a good start to look for very simple patterns such as what I outline with gaps or double bottoms? Can a system be built with just double bottoms or is this too simplistic? When you only need to be right 50% of the time, it seems like a lucrative proposition, but I doubt that I could have stumbled onto something so easy. Perhaps it works given the right values for stop and target?

 

Forget the patterns. Patterns are just that, patterns. They won't make you money if you don't know the structure behind the pattern. Not every pattern works, and you need to understand the structure so you can cherry pick them accordingly. Same goes with gaps. Therefore, your system will lose without this knowledge.

 

Thanks Slope Trader... for outlining your process. This is exactly how I imagined it to work, but I don't think I have ever come across any posts that outline it in such great detail. All I read is people say retracements in an obvious bull trend are excellent entries, but I think everyone just thinks this because of what we have been told versus actually going through 50 charts to prove this.

I realize your edge might be secretive, but if you can give any details, I would love to see what you are looking for. I guess what I'm asking is that I don't need to reinvent the wheel, so just wondering where to start as double bottoms or gaps seems far too simple of a proposition for a winning strategy. Wouldn't the automated trading system that people have developed already play this? I mean if a stop gaps up, and 50% of the time it retraces back to the previous day's close rather than shooting up even higher, wouldn't this make for a great and simple system? Would I find that its too random... that the choppiness of opening after a gap would stop me out of positions far too soon?

 

Learn what an edge is – Douglas’ definition

Then do what these good folks suggest – back test/ research/ hunt & peck / whatever – the crap out of it – till you find repeating behavior (won’t be exactly repeating..., but close is close enough) - and study it till you can identify it as it's being created

Then get on sim and start looking for that which you can now identify…, while also figuring out how to exploit it... AND.... keep your risk small (last two must go together like peas and carrots)

RN

 

Thanks xelite777, but unfortunately, as of now, I wouldn't quite know what to do with any of it! LOL.... But my gosh, how sweet it would be to just develop an automated system that makes lots of trades a day, wins half of them, and profits twice as much as it loses, and just let it run all day long!

 

If you aren't looking for a pattern, then what is it that you are looking for when trying to find something that repeats? How do you express what you see on the screen if you can't call it a pattern?

Put another way, isn't a retracement just a pattern? If I was to visually describe it, it would be a series of higher highs and higher lows, with the retracement being the few red bars after a bunch of green bars, but with the red bars stopping before it falls all the way to the bottom of where the trend started. Can't you think of a retracement as a pattern in this way?

 

Thanks eurusdzn... I have read lots of his posts as well, but his work is far above my level. I know nothing about programming, nor statistics, so I would be quite lost. But I certainly fully believe in his approach.

 

1. Yes, it can be as simple as reacting to an opening gap, assuming you're talking about stocks (there are no gaps in futures, except from Friday afternoon to Sunday evening). And, yes, manually testing this would be exhaustive. And after futzing with it for hours and days and weeks, you might decide that you want something that shows up nearly every day, not just once in a while. Be wary, however, of being in a hurry. If you hurry, whatever you're doing will take two to three times longer at minimum.

2. I've been criticized for referring people to my threads, but repeating this stuff can get to be a drag. However, there are only three strategies -- reversals, breakouts, and retracements -- and what you choose will depend on a number of factors including how often they come up, how easy they are for you to define, what your risk tolerance is, how well you understand the auction market, how well you are able to detect changes in the balance between demand and supply, how easily you become disoriented with regard to up and down, whether you trade intraday or interday, etc. Yes, you can focus on double bottoms, which are a form of reversal. But if you trade only those which are most likely to generate a profit that will make the waiting worthwhile, then the waits will likely be lengthy. You won't make money trading crap, but neither will you make money if you don't trade at all.

As for being right 50% of the time, I suggest you go for much more than that. If half your trades are losers, you will not likely last very long. And if you set "targets", you will in effect be cutting your profits short, and doing so with such a tepid win rate will be wearing before long.

If you're a beginner, you're in a good position as you have so little to unlearn. Just make sure to determine for yourself what is or is not true among all that you read and that you are told, remembering that you have no way of knowing whether or not whoever is offering the advice is successful or not. And why would you want to take advice from some mystery trader without verifying for yourself that it is true?

I should also clarify with regard to your reference to "price pattern or market behavior" that this is not either/or but rather two ways of saying the same thing. If you don't understand how the behavior of traders is creating the pattern, you will not only interpret it incorrectly but play it wrong.

 

Excellent stuff NoDoji.. thank-you so much!

It makes me wonder though that with so many filters, are you truly then taking each setup as you should? To backtest double bottoms is one thing, but to backtest a double bottom that is far enough away from a resistance level and has adequate outside bars and... and... and.... well this just sounds like cherry picking now which would invalidate a thorough statistical analysis of how often double bottoms work... or am I wrong?

I can certainly understand how its much better to take trades that have a higher chance of success, but are you not now trading in the moment based on your years of experience versus trading a system with well defined rules?

 

Yes, you are correct. You're being too literal though. I'm saying that without understanding why the pattern works, you can not cherry pick them accordingly. For example, double bottoms have no positive expectancy alone. There are other factors that make some work and some not. You need to be able to understand why some fail and some don't.

For example, lets talk about bull flags. We all know bull flags can be a great pullback opportunity in an uptrend. But do you know WHY the bull flag works. Do you know why the W pattern works ? Or the head and shoulders ?