Timing Test Results!
A week ago I decided to perform a real-time public experiment with one of the timing strategies that had been discussed here (see post 8072).
The point was not so much to predict the course of the general market or even a particular stock, but rather to shop a bit for a better short-term entry point once a decision to buy had already been made. I split some capital into two pockets, and used the first to buy shares in six stocks popular around here at the market price between 10:30-11:00 am Monday morning. I reserved the second pocket for a later purchase of the same six stocks, placing limit orders for them at 5% less than the market price at the time of the first buy, and committing myself mentally to buy in three days later if the limit orders had not hit by then. It turned out to be quite a wild week, but perhaps less wild than one might have expected for the stocks in question. Here are the results of the test, along with some observations:
Stock 1: JDSU
The first buy was made on Monday at 133, and the limit order was set at 126 3/8. It hit on Wednesday, with the actual purchase coming in at 126 5/16. By Friday's close the price was 128 5/8.
Stock 2: CTXS
The first buy was made on Monday at 65 15/16, and the limit order was set at 62 5/8. It hit and was executed at that price on Tuesday. By Friday's close the price was 60 9/16.
Stock 3: CREE
The first buy was made on Monday at 39 7/8, and the limit order was set at 37 7/8. It hit Wednesday morning, with the actual purchase coming at 37 /14. By Friday's close the price was 37 3/4. Earnings two cents above estimates were posted on Thursday. (NB: I believe CREE to be a Shiny Pebble, or more likely a DITR; these were my first purchases of the stock, and I intend to hold it lightly until--and perhaps after--I complete my research into its prospects, which I will post when complete.)
Stock 4: SEBL
The first buy was made on Monday at 86 1/16, and the limit was set at 81 3/4. With high earnings expected soon the price shot up afterwards. When I saw it dip briefly to about 82 1/2 during the mid-week carnage and then start back up, I broke the rules <g> and jumped in, calculating that this was the best chance I was likely to get during the week. I was correct. I bought in at 83 3/8 on Wednesday, it never hit the limit, and closed on Friday at 85 1/8.
Stock 5: GMST
The first buy was made on Monday at 74 5/8, and the limit was set at 71 1/8. It never came close to that--never dropped below the initial purchase price, as far as I saw--and closed the week at 75 7/16. I did not buy in on Thursday morning, however, because I broke the rules of the game a second time <g> and used the second half of the GMST cash for....
Stock 6: The Mighty Q
The first buy was made on Monday at 222 5/8, and the limit was set at 211 1/2. It hit and was executed at that price on Tuesday. It continued to drop, however, and by Thursday morning was well below 200. I said to myself, "Screw the test," and used the remaining GMST money to buy into QCOM a third time at 197 1/8--11% below the first buy. It closed on Friday at 198 1/2.
Lessons
1. Well, for what such a small-N test is worth, the strategy did generally work as expected. All the stocks except one dropped substantially, and four out of six hit their limits. This was not, moreover, because they were destroyed in the end-of-the-week catastrophe: the limits were hit Tuesday and Wednesday, when the market was simply giving back Monday's gains. So I do not think the strategy requires a general plunge in order to work. In fact, since good earnings were expected to be posted soon for most of my companies, this week could actually have been considered a tough test for the strategy to pass.
2. I set the limit orders at 5% below the initial price because I wanted to give them a good chance of hitting, and this was borne out. The author's original strategy of using a 10% decline instead would also have worked, I suppose, but more because of the general decline late in the week rather than because of normal fluctuation. Only QCOM dropped 10%, and only CTXS ended the week between 5-10% down, but three out of the other four closed the week below the initial purchase price. The optimal amount to set the limit at remains a good subject for further exploration.
3. Using a mechanical strategy was definitely a good thing, because even during such a short period the stocks were all over the place--again, even before the disastrous second half of the week. On Monday, when everything was up, I wanted desperately to buy in and cut my expected losses. Later, when everything was down slightly, I wanted desperately to buy to lock in my (minor) gains. Having a fixed plan help instill discipline, and made sure that when I did in fact deviate--buying SEBL above its limit, buying QCOM instead of GMST at the end--I did so not on a whim, or in panic, but for very well-grounded reasons.
4. Somewhat unrelated to this test, for me this week confirmed the value of holding only a few stocks that I knew very very well. Because I was secure in my knowledge of the companies I was invested in, I was able to weather the turbulence much more serenely than I would ever have thought possible--and even, as someone had predicted, to root for my favorite stocks to go down so I could buy in more cheaply!
In the end, I am pleased enough with the results to consider trying a similar test again. If I do so, I will try to do it in public so that others can benefit from the experience.
Tekboy@enoughofthisshit,nowthemarketshouldgoup.com |
