Replies to post #5952 on FOREX TRADERS

[Capitalist]

I've got results and although they didn't look like I predicted they would in my notes, they kind of conform to what I've been saying:

There are indicators that have a lot of "meaning", but that "meaning" changes over time.

For example, I'm looking at a combination of indicators that can have only 10 possible configurations, so I call those "chart situations" number 1 through 10. They're kind of on a scale so that situation #1 is a very bearish looking combination and situation #10 is very bullish.

Then I look at how price behaves in the future for each one. When I looked at one set of data, all the bearish looking combinations led to bullish price moves in the future, and vice-versa, so there was a very strong correlation between the indicator and what happened in the future. So I predicted that if I took a fresh set of data I would see something similar.

Well I didn't.

What I DID see though was a very strong correlation in the OTHER DIRECTION. The really bullish looking combinations led to really bullish future price moves and the bearish ones led to downturns.

So in both weeks, the indicators had a lot of "meaning" but that "meaning" was different.

It's like in one week a positive crossover of a moving average always leads to an uptrend and vice-versa. But then a few weeks later a positive crossover is always followed by a DOWNTREND and vice-versa. It's like the indicator works great, but it works in a different way at different times.

Weird huh?