Pure logic: Random entries don't produce the same result. Market is no different than ever, same stuff works for years already. What else can it be then?
ORB's. here's an interesting article by ET's own Murray Ruggerio. http://www.futuresmag.com/2012/08/01/opening-range-breakout-past-present-and-future?page=3
Just keep on preaching this mantra--- I guess it ads security to an uncertain endeavor. Markets have changed radically and continue to change....you need to be flexible or the flexions will crush you. surf
@ Fireplace and cornix When Surf used the word "volatility" that pretty much outed him. His absence of a response is also quite telling.
True, but you can't use that as a "proof" that TA isn't what makes guys/gals like Cornix and NoD profitable; it doesn't even qualify as an argument for any side. Which is what you tried to do, I think. Saying that "entries don't matter" is just another way of saying you're discussing a group of systems (profitable or not) that use random entries. Such a system could still be based on TA for scaling/profit taking/mm/stop loss logic etc. No, why?
Of course there are ways to objectively define trends, and you can use the definitions to perform tests. The reason I wondered how you would define trend is you said tests had shown the "trend isn't your friend". So what you meant was that the definition of trend used for the tests in question was useless; I was curious to what definition that was. A few "objective trend definitions": http://www.elitetrader.com/vb/showthread.php?s=&postid=3247539#post3247539 Are they usable? You have to test them to know .
I have, and continue to do so----- sorry, trends and trend following simply do not hold water when properly tested in the stock market. Trend is quite discernible using commonly used statistical testing procedures. Confidence intervals, serial correlation coefficients, regression coefficients of current change vs past change, magnitude of the impact of past moving averages on the future, distribution of the length of runs and the correllogram are used to differentiate randomness from trending behavior-- --When these tests are applied to the stock market, there is no trending behavior in price--- In fact, these tests indicate negative serial correlation. \ none of anyone's claims to the opposite invalidate the facts. sorry. surf
Dude, the other day you got caught with your pants down watching 5 minute bars on a price chart. You enjoy pulling your own chain.