Just show me one mechanical technical analysis that has been profitably backtested?

I can show you a method of trading that is based on statistics and money management that goes on streaks of a 100% win rate but if you force me to rely solely on technical analysis I'd probably lose on balance. Now don't get me wrong, I draw support and resistance lines all day long but I always have a plan for when they fail, and they fail often.

It's my opinion, and my experience, that you can't trade successfully if you can't manage your account. Putting on a trade with the same size on each trade and managing risk with a stop is destined to fail.
 
This looks pretty good but to call it the most profitable ever I think you need a win rate above 90%.

Quote from bwolinsky:

The most profitable mechanical system ever written is below. It's called SuperBands
More...
 
Quote from crgarcia:

One, just one.

One that has worked profitably (backtesting) for many, many years.
More...

So in plain English, you are asking for a profitable system for free.

Clever!
 
Quote from Bob111:

heh.. how do you know,if it's most profitable EVER?
i have no problem,if you say-most profitable posted on public forum..
More...
I thought about adding that. Most profitable publicly available script. Hear that, Jack? Most profitable in the public domain? And I didn't even need a printer.LOL!
 
Quote from ProfitTakgFool:

This looks pretty good but to call it the most profitable ever I think you need a win rate above 90%.
More...

That's where most people go wrong, too.
 
Quote from Pekelo:

So in plain English, you are asking for a profitable system for free.

Clever!
More...

Say I could give you a working version in NT, which I have, would you even know what to do with it? Do you have about $50,000 to $250,000 in an account to make it worth your time at 1% sizing? Other things, and partially why I quit offering the system on collective2, do you have 3 hours per night to setup 100 charts on 100 different symbols with the same settings each night for completely automated hands off trading? Would you know how to correct the script to fix it? If any of these answers are no, which for me there's a couple, then you probably shouldn't trade it. Anyway...

I don't. If I was in an institution I'm sure I could, but this is still public domain. I'm not sure how much institutional interest there is.
 
Quote from Bob111:


but..i have to add-it take few years of search and try for almost 24/7 to get there..
More...

You have no idea.... I was on vacation in a million dollar condo with my future wife, and luckily, WL would allow you to work there. The first version was the most profitable to be released at the time, beating a simple dip buyer that bought 8% below yesterdays close and sold on the open. It's flaw was in buying on the open, so I corrected that, the next versions doubled in performance when adding linear regression analysis and predictive values for use intraday. This combination took a few years, but it's public, because it's not my best work.
 
Quote from Bob111:

heh.. how do you know,if it's most profitable EVER?
More...


There is not a doubt in my mind. Why? It's a problem of data and WL has only been around since 1999. Systems designed for today's markets might not have worked in the past, but have huge performance results for the moment. Over time, the results start to even out to a quite normal return nearing buy and hold. The real system that will last is one that gets significantly higher returns over a long period of time with far less risk. This is where I see too many ET yahoos start thinking they can get 100% per year with 20% risk forever. It's bullshit. You'd be good to do 15-20% annually over a 40 year period, but you'd probably blow up a few times along the way. I challenge anyone else to post better results from 1/1/1996 with greater than 30% returns and less than 17% drawdown. If you wanted to double the size you'd miss a few trades but your performance would be 1.3^2-1 and drawdown at 1.17^2-1.
 
You must log in or register to reply here.
Back
Top