=================== T-3 Tek analysis can be profitable ,with enough work, so he maybe more profitable, than an accurate speaker/writer.LOL. But 100% would be a prediction. So no, tech analysis is not prediction/100% I was talking to a ''certified'' appraiser; guess what ,he was not infallible/100%].LOL/Belly laugh Even 90% [which may or may not be true, for a larger period of time]; could easily be wrong 10 months in a row,in 10 years . But that is not a prediction either .LOL Wisdom is profitable to direct Derivatives may not be suitable for all investors.
We have discussed this very detailed in this thread. http://www.elitetrader.com/vb/showt...103&perpage=6&highlight=win rate&pagenumber=1 About the definition of a win rate. Win rate means to win. It doesnt mean how much you win. But it doesnt matter if you believe my 90% win rate, it will still work, if you dont believe in it, important is only that i believe in it, based on facts. But of course it only works if, i dont make any mistake, because how i posted before, if i do a wrong thing, i probably will lose money, but this have nothing to do with the accurany of the strategies.
They sure do... And your unsuccessful trades don't count as successful trades! Uh huh... Make a dollar 9/10 times in a row and then lose $ 20 on the last one...! It's actually about how much money you make... not about how many times you win. That being said... You'd better be right about the market sometimes because if you're not you'll never make money.
Seriously, with a couple of exceptions, the rest of you guys are missing the point as it pertains to the OP's TA acquaintance: http://en.wikipedia.org/wiki/Circular_reasoning
Then you obviously subscribe to the Will Rogers (or was that Mark Twain?) philosophy of, "buy a stock. When it goes up, sell it. If it don't go up, don't buy it."
I had a short-term edge that pulled around 85% accuracy briefly while I ran it, but generally my edges drop to around the 55% mark as well (on the whole.) The biggest dragons I fight are the fees and the bid-ask spread. Even if I design something with 80% accuracy, the fees associated with such a strategy will chop me to bits.
Even if this is true, you are still left with the fact that identifying a legitimate pattern in real-time is an exercise in probability. In other words, the only way this would matter is if someone had 100% accuracy in identifying the pattern, which would be very unlikely. Even then, a circumstance could arise where a valid (in this guy's definition) pattern played out and a TA practitioner missed it. Generally, arguing absolutes in the market is a waste of time. In my case, I give every trade setup a grade from 0-100 and only trades with a 30 or above get taken, although any score above 0 is technically valid, but scores of 1-30 have performance which is so close to breakeven that there is no real value in taking the trade. But, the biggest winning trade I've ever seen triggered by my model scored less than a 30, while I've had trades which scored 100 turn out to be losers. In general, I just think this is how the market works for any trader and that trying to find a way for it to work otherwise is similar to trying to jump over your own shadow.