Something very simplistic

Status
Not open for further replies.
Originally posted by echo
I would say the edge, if you want to call it that, is just that it forces a set discipline. I would be surprised if the fib based entry times proved to be more than that. Although it could possibly be interesting to see if over the long haul, a certain time period of entry generated the most income given a certain long vs. short methodology.
More...

Exactly. This type of system forces a exact discipline. It also had other benefits - if you know exactly when you are going to make a trade you can be off doing something else and not worry about missing signals that could happen at any time. You know up front exactly what the possible outcomes are.
 
Quah,

I must still be missing something, please bear with me for I can be pretty slow sometimes :eek:

If you are taking gains less then 1to1 then how could this method be profitable if the entry is not drasticlaly putting the probability of a winner in your favor???? What is it about entering at those specific Fib time intervals that is providing this???

PEACE and good trading,
Commisso
 
Originally posted by J_Commisso
I must still be missing something, please bear with me for I can be pretty slow sometimes :eek:
Commisso
More...
Sounds more like Colombo than Commisso :D
 
Originally posted by J_Commisso
Quah,

I must still be missing something, please bear with me for I can be pretty slow sometimes :eek:

If you are taking gains less then 1to1 then how could this method be profitable if the entry is not drasticlaly putting the probability of a winner in your favor???? What is it about entering at those specific Fib time intervals that is providing this???

PEACE and good trading,
Commisso
More...

I'm not saying entering at those times IS NOT putting the probability of a winner in my favor. It seems that it does, but I can't tell you why it does because I don't know why. And if it does, it does, no matter if I can explain it or not. :)

Lets say I have two indicators - one being those times and the other being "Indicator A".

Let's say if "Indicator A" is above 80 that is a long signal, if it is below 20 it is a short signal. And when I look through a day those signals (for example) are only correct, say, 40% of the time.

But then I come along and look at "Indicator A" only at those specific times and notice that it is correct 95% of the time if you use it only at those times (hypothetical situation here).

Given that situation, I couldn't really give any reason why "Indicator A" seems to work well at those specific times. There may be other times that "Indicator A" works just as well - but why look for those? At that point I've satisfied the requirements of my system.
 
Originally posted by AAAintheBeltway
Why not just get your dog to make the trades? That way you don't have to do anything. And dogs are very intuitive.
More...

If I had a dog, and if the dog could trade in a way that gives me what I'm looking for, I'd be all for it. Why not? :D
 
Originally posted by Quah

Here is the system:

1. Use a minute bar chart

2. Trades will be entered only at specific times - they will be entered at the opening price of the following bars after the open - 3, 5, 8, 13, 21, 34, 55, 89, 144, 233, 377. So, on a normal 9:30 open, trades would take place at 9:33, 9:35, 9:38, 9:43, 9:51, 10:04, 10:25, 10:59, 11:54, 13:23, and 15:47.

3. Use any indicator you like to decide if you will go long or short - doesn't really matter - I use stochastics. Almost anything will work - even something as simple as going in the direction of the previous bar.

4. Once you are filled at the opening price of the bar, enter a limit order to close your position +1.

5. Enter a stop at -1.25. You can reverse at the stop and enter a limit order to close that position at +1.25, leaving you at B/E for that bar's trade (less commish of course).

That's about it. I'll post today's results in a separate post.
More...


a fib. time series counting method, I guess the next step is to decide when to start the counting

edit: re-draw as suggest (3 posts down)

edit: I get it right the first time but people can count bar and draw their own conclusion :eek:
 
QUAH.... I think you are correct in ignoring backtesting and I also agree if the numbers "work" in real time it isnt necessary to know why.
 
Status
Not open for further replies.
Back
Top