Friday, February 27, 2009

Goldman And Morgan


Looking at this Goldman chart, we'd never know that the general market has been teetering near its lows. Goldman still is up 94% off its November low.

A couple of observations about the chart:

1. Goldman had what appeared to be a beautiful TRIPLE breakout of an Ascending Triangle on December 30. Unfortunately, it morphed into a Bearish Rising Wedge, and Goldman got smacked very hard when that became apparent, when it took out the lower trendline of the pattern.

2. When a Bearish Rising Wedge gets broken, probably most technical analysts/books will tell us that the expectation is a return to the low of the pattern. I don't find that to be true. Often, the stock or index stops well above the low of the pattern, as Goldman did in this chart. I like to use lateral support levels as targets, which in this case would be Blue #4 (65.50) and Blue #2 (60.22), which were the lows of the nested Symmetrical Triangle (pattern in blue) for targets to take profits, or at least some profits, on a short position. We don't want to over-expect on trades. Goldman bottomed at 59.13 in January, so those targets both got MADE.

3. In early February, Goldman filled the 78.65 gap and staged a strong rally. In the recent selloff, that gap area again was support. Goldman is showing very good relative strength. For that reason, I played it long yesterday when I saw that it came back and filled the morning gap, and was forming an Ascending Triangle.

I got long Goldman at 91.89 after White Data Point #4 was established. On a breakout, the pattern target was 93.17. It broke out, and ANOTHER pattern emerged, the Symmetrical Triangle in yellow. The pattern target on that breakout was identical: 93.17. When you get multiple pattern breakouts, and when targets are close to each other, the chances of getting the targets are very good.

Once the Symmetrical Triangle broke out, I raised my stop to the "last low" of the pattern, 92.01, at Yellow #4. The low of the pattern was 91.87, and the breakout still would be valid unless/until that was taken down, but my entry was 91.89 and I wanted at least a small gain on the trade, so I used the 92.01 stop.

Lord, did I get punished for that! The market doesn't know what WE want, nor does it care. LOL. After the two breakouts, Goldman rallied nicely toward the 93.17 target, but then reversed and came back in my face and took out my 92.01 stop. I hit the sell buzzer and fortunately good a good fill at 92.05, but was disappointed that the trade didn't work out as planned. I made a paltry $150, or so.

Guess what? Goldman sold off to 91.87, the EXACT low of the Symmetrical Triangle, which "should have been" my stop (91.86), held right there, then rallied directly to 93.20, three cents above the 93.17 DOUBLE target. I "woulda" made $1,200+ if I had used the low of the pattern as a stop. Sheesh...

If the "last low" had been significantly above the low of the pattern, which it often is, it can be a significant difference in the amount of risk, but the difference in this case was very small. LESSON: "Don't be cheap when the choice of stops is insignificant. Use the LOWER one." Especially on a DOUBLE pattern breakout. Sheesh...

While I was sitting here slapping myself as Goldman got to the 93.17 DOUBLE target (Get over it, Melf), I noticed that Morgan was trading at 22.65, right near the bottom of the broken Bear Flag (resistance), which came in yesterday at 22.765. On Wednesday, it filled the 22.93 gap EXACTLY (high was 22.92), and failed, now here it was back for another test of the broken Bear Flag!

I shorted Morgan right there, at 22.65. At about 12:30, Morgan had sold off to the low $20's, but Goldman still was looking strong, putting in new highs on the day, so I covered my Morgan short at 20.03 for a nice gain of $1,500+, then went back to slapping myself over the Goldman trade. LOL.

I was long 1,000 Goldman. They don't all show up at the top of the page because they fill you in those annoying little 100 share batches, I have to scroll to see them all.

No comments: