<<ok--so now i know for sure microsoft can either go up or down>> predicting stock prices is not an exact science, and a lot of what you do, is based on statistical correlations.
For example, if you observe, that a given stock tends to gap up in the morning, and sell off toward the end of the day, 85% of the time, then that is statistically significant. What you do, is try to identify all the conditions surrounding such behavior, and from that you can further refine your data. You take the 15% of the time when the stock is down and you look at other factors. In the example above, perhaps you notice that when the market was down the previous day, the odds of the stock *not* gapping up in the morning, increase to 50%. That tells you, that the 15% of the time when the stock does not gap up, can be further refined - now, you are in the possession of the following information: if the market is up the previous day, the odds that the stock will *not* gap up, is only 7% (remember the 50% reduction of 15%). With additional factors, you can parse the odds finer and finer. However, there is ALWAYS a small chance that you will hit that 7% or 2% or 1% or whatever "freak" thing. But as a trader, what are you going to bank on? the 93% chance, or the 7% chance?
The same with a lot of TA. There are patterns which are statistically associated with certain behavior. It makes sense to bet with the odds. Sometimes, a certain chart formation is hard to read, as it can mean the beginnig of a trend up or a trend down, and with time, it shows which way it is - the chart in this case indicated to you that a turning point is at hand. What it did NOT indicate, was whether the turn was up or down. That was the meaning of the statement I made, of which the exact version is <<a platform from which a move can be launched (usually up, unless there is a breakdown, in which case TA practitioners would revise what the nature of the preceding action was)>>. In other words, the chart looks at this point like a certain kind of formation (basing), however, instead of an expected move UP, the move actually is down - in which case a TA guy would go back and say: the formation "X" which could have been either a "base" or a "crest" turned out to be a "crest" not a "base"."
Obviously, as a trader, you try to get as clear a picture as you can. Sometimes the picture is good enough, so that the odds of your bet being right are 85% - in which case you go with it. What if it turns out that the 15% case transpires? Basically, you're f****d. All you can do, is go back and say: is there something I can learn from this, was there some other factor which had I spotted it, would have told me to bet the other way. But often, you are no wiser, and in retrospect, all you can say is: if I had the same situation presented to me again, I would have done exactly the SAME thing. I must go with the odds. If you go to a surgeon, and he tells you the success rate of this operation is 75% - you take your odds, depending on how much you have to gain/lose. Here other considerations enter - risk/reward ratios etc.
Naturally, as I always say - TA is a limited tool, FA is a limited tool. Unexpected events, can turn a bet that was 95% in your favor to go against you. For example, core inflation numbers shoot up, and whatever TA indicated for the following day, is rendered meaningless.
The way to think of TA is as follows: odds, odds, odds. Bet the odds, but use FA, try to anticipate other factors, and adjust your bet to the risk/reward ratio.
If we know that structurally, after examining an airplane's construction, the odds are, 1 in a million that if you take a flight, the plane will have a structural failure, you like the odds and take the flight. Think of TA as that statistic. Now, a terrorist shoots a missile and takes the plane out. You lost. Well, TA cannot anticipate that. And so, yes, there is always a chance that external events will overwhelm TA. But you still go with the odds.
That is why, I'll often say things like: "on purely TA grounds, the stock looks weak/strong - but of course, if the DOJ case goes well/badly, all bets are off".
So, how useful is TA? I find, that properly practiced it significantly increases my odds of making a successful trade. It is not a guarantee. But in my book, if I win 75% of the time, and lose 25% of the time, I'm ahead of the game - providing I look at every trade from a risk/reward ratio point of view - obviously, I try to pick only high-probability trades, and I bet more money the higher the probability.
Morgan |
