Maybe I'm way off base here but as was pointed out earlier, aren't we judging human behavior in a trend? For example, if enough people hear that resistance is at a specific number, will this not incite many people to sell at this juncture and fullfill the prophecy, therefore not making this random?
Why do I see "Trends" in Randomly Generated Data?
- Thread starter Rahula
- Start date
If you flipped an unbiased coin 10 times and by an accident this coin has landed all 10 times on âheadsâ does it mean that you are in a âhead trendâ? Trends are the observations made on the past behavior. At any point of time the probability of the trend to continue or reverse is still 50/50. Price patterns are for idiots who have no education. It is a religion and as any religion it has rituals â one of them is trend! Spotting patterns in no pattern environment is like telling your girlfriend, âHoney, doesnât this cloud look like a dogâs head?â If she wants you to buy her a few shiny things she would say âYesâ. Patterns, trend lines, head and shoulders and other bullshit is like dancing around the fire and praying for rain. It takes about the same level of intelligence.
that makes sense on the surface. however, remember just as many traders are trying to GAME the system, doing the opposite at any point of time.
surf
While it is statistically possible for a fair coin to perform in this manner, the odds of it doing so are about 0.000977. I could use a coin like that.
MAESTRO! the master has returned, nice to see you back.
how is the super computer, fund and monitor wall?
look forward to reading your musings.
regards,
surf
Quote from Thunderdog:
While it is statistically possible for a fair coin to perform in this manner, the odds of it doing so are about 0.000977. I could use a coin like that.
me too!
snowed in surf
Quote from Thunderdog:
While it is statistically possible for a fair coin to perform in this manner, the odds of it doing so are about 0.000977. I could use a coin like that.
There are about 300,000 of price ticks (up or down) in the ES each month so 300,000 x 0.000977 = 293. That is how many times a month you might see the trend on a tick chart. Given that there is always a slight bias on the markets (on average 4 - 5 % difference between up ticks vs. down ticks) the frequency of such event is a lot higher.
I am glad to be back. It was hard for me these two years. I have spent most of this time in the hospital (almost checked out of this life). But I am much better these days, thank you for remembering me. The super computers and the wall of monitors are still there. Our average performance for the past 4 years is 22.3% annualized. Not much, but puts us right in line with other good guys.
Cheers.
Hey, MAESTRO, I haven't seen you in a while. I hope all is well.
I'm not sure I completely agree with you. I personally don't use trend lines or the more intricate patterns such as triangles, diamonds and the like. (To me a head-and-shoulder pattern is merely a lower high or, in the inverted case, a higher low.) I don't think that higher lows should readily be dismissed in a rising market, or lower highs in a falling market. These are demonstrable biases thus far. The bias may change at any time, but it at least momentarily exists. I think it is fallacious to equate market behavior with the tossing of coins because, although there is an undeniable random component in the markets, it is not of the order of a fair coin toss. The market comprises traders who tend to base their decision, in one way or another on preceding price action. Coins tend to be somewhat less concerned (as in completely) about what happened at the last toss.