Читать книгу Encyclopedia of Chart Patterns - Thomas N. Bulkowski - Страница 47
Trading Tactics
ОглавлениеThe AB=CD pattern can be a wonderful tool to help predict when price will turn and then make a substantial decline. Once you know the first three turns, you can determine when and at what price the fourth turn will appear. And when turn D appears, the stock will drop. Does it really work like that? Let's find out.
Table 2.3 shows how price behaves after point D.
How often does price reach or exceed D? I checked how often price climbed far enough to reach the calculated price of D. The table shows that nearly all of the patterns reached the target turn.
Table 2.3 Price Move after Pattern End
| Description | Bull Market | Bear Market |
|---|---|---|
| How often does price reach or exceed D? | 95% | 98% |
| How often does price turn at D? | 32% | 38% |
| How often does D appear within a week of calculated time? | 43% | 45% |
| How many drop to point A? | 24% | 36% |
| How many drop to point B? | 76% | 87% |
| How many drop to point C? | 35% | 46% |
That's terrific! Swing traders can use this to predict how far price will rise. They can even trade it by buying at the low at C and riding price higher, to D.
How often does price turn at D? I checked each pattern to see if a minor high formed at the calculated point D. I found that only about a third of the time will you see price turn lower at the calculated price of D. Because we know price rises to D nearly all of the time, we can assume that price continues beyond D instead of turning lower when it should.
This finding is not a deal breaker. Now that we know price will likely continue rising, we can just stay in our trade (when we bought after turn C) and ride the stock upward until it does turn.
How often does D appear within a week of calculated time? The pattern can work as a predictor of when point D will occur (as well as the price of the turn). I found the dates of the ABC turns and found the ratio of CB to BA. Then point D followed the equation: D = (C – B)/Ratio + C using dates instead of price.
I found that between 43% and 45% of the time point D appeared within (plus or minus a 2‐week window) a week of when it was supposed to. Because the numbers fall well short of the expected time, I don't think this measure is helpful.
How many drop to…? If price reaches D and turns down, we know that the average decline measures between 12% and 22% from Table 2.2. Let's measure how far price drops in terms of turns A, B, and C.
Point B is closest to turn D, so we would expect the stock to drop that far most often. Indeed, the table shows that price reaches turn B between 76% and 87% of the time. Price will drop to C less often (35% to 46% of the time) and reach the bottom of the pattern (point A) even less often.
Using these values, we can get a sense of how far price might decline. It could be less or more, depending on the situation, of course. But at least we have a roadmap.
