Is Technical Analysis Worthwhile?

As we've already noted in this BLog the global answer to that question is - oh hell no - but, and there is always a "but" - any trader who is making a living at trading who can't look at three candlesticks and "guess" the direction of the market for the next three candlesticks with at least 65% accuracy isn't going to continue making money. And 65% accuracy is better than "guessing" and 65% accuracy over a stream of 3 discrete events is far better than guessing.

For example here is SBUX and my guess today is that SBUX has had enough of the down side and will, market forces permitting, turn up for the next three weeks at least.


What does "market forces permitting" mean? It means if the market as a whole goes up then SBUX should go up.

And where is the market as a whole going? According to SPY - up


Why am I guessing "up" - because there is only one other way to guess and it has more letters in it.

No - not really - I'm guessing up because the SPY weekly chart shows a return to 4 over the past 4 weeks. You can see that 4 weeks ago the open was above EMA 4 and the close was above the open. The following week repeated that pattern. The following week moved sideways and last week the open hit right on the EMA 4. That is a return to 4 pattern and that pattern always means that with a 65% probability the market will be higher three weeks from now. Which means SBUX will, with 65% probability be higher three weeks from now.

CROX on the other hand will not be.


Oh my God, Amy, I can hear you blubbering over here. But, but, but, but - you sound like an old motor boat - just look at the chart and tell me what you see --- that's right a stock that has run away from its base - the EMA 4. So my guess is that CROX needs a rest and will go sideways-down for the next three weeks (return to 4) - after that - who knows - those candlesticks haven't been printed yet.

The CROX chart has a lot of interesting items on it. First is the cross over back in July '06 and if you bought it there as I would have advised at the time (had I seen it then) you would be several hundred percent to the good right now.

It did a return to 4 in Oct and posted a "cathedral of dead money" in the second week of November. (The Japanese literature calls that an evening star). Regardless of what it is called - that that would have been a good place to sell because that's when the funds did. This would have been followed by a reacquisition in the first week of January off the BOB that formed in December.

In November/December do you think that was basing, consolidating, or resting? I'm going to opt for the "Christmas-holidays-end-of-the-year-mutual-fund-tape- painting-and-reacquisition-because-we-must-have-CROX-in-our-fund-
going-into-the-New-Year" formation that is so popular among the TA crowd. (Although I'm pretty sure it isn't called that).

It did another run away from the 4 in February (a crummy month for many stocks) with the end of February general market melt-down and then received new life in the second week of March. Whenever your stock is down on the 21 EMA and pulls back and leaves a long tail - buy it - put your stop under the tail and have no fear. You can see this also on the CROX chart back in May and June '06 - so you wouldn't have had to waitfor the cross over. And remember - the cross over is a very conservative entry - the move has begun by then.

So for about a year or so you could have bought and sold CROX from just technical analysis with no regard to fundamentals, product roll outs or any of the other nonsense that the analysts like to jibber-jabber about. And the only two things that mattered: was anyone else buying/selling the stock which the candlesticks make abundantly clear; and, what is the overall market doing?

So is technical analysis worthwhile? Forget the global answer (remember rule 1) and just learn it and do it. What works works.