Long Term Secrets To Short-Term Trading

Table 6.5 shows the results of buying on the opening and exiting 3 days later. Any remaining random 
walk enthusiasts will tell you we should not be able to find any differences between days of the week over 
a 3-day period. An efficient market should wipe that out. Yet when we look at just the best performing day 
of each week, based on the open-to-close change, we see a large bias and taste the sweetness of knowing 
markets are not totally random. The only random market was Gold; the rest of the markets I studied beat 
the random walk. Bonds and the S&P 500 led the way showing some decent profits. 

---

86 
Table 6.5 
Best Trading Day of Week with a Three-Day Hold 
TDW does make a difference and can give us a workable bias to trade with. There are numerous ways 
to begin milking this cow, and you probably have already thought of some on your own. Certainly, it is a 
bias you want to understand, and consider, for any market you are going to trade on a shortterm basis. 
Earlier, I said the open is critical; if we start to expand or move away from the opening, price will 
probably continue in that direction. Now I will demonstrate one such approach. We will combine our 
day-of-the-week bias with one simple rule; buy on the opening of the bias day + X percent of the previous 
day's range. We target our bias day, and buy that day at an expansion off the opening price. Our exit is 
simple; we hold the trade to the close and take our profits/losses at that time. (There are better exit 
techniques, which I will get to later.) 
The S&P 500 results of buying the opening Monday +.05 percent of Friday's range are pretty 
spectacular for trading just one day a week (see Figure 6.2)! This approach shows a net profit of 
$95,150 with 435
Figure 6.2 Buying on the opening Monday. 

---

87 
winning trades out of a total 758. Thus the average profit per trade is $125 with 57 percent accuracy. Bonds 
make money buying on Tuesday opening plus 70 percent of Monday's range with $28,812 profits and 53 
percent accuracy netting $86 a trade which is a little small, but a better exit technique will radically change 
this number (see Figure 6.3). The long and short of all this data is that a simple filter, TDW, enables us to 
do what the professors say is impossible ... beat the market. 
To recap, stocks have a proven proclivity to rally on Mondays, Bonds on Tuesdays and virtually all 
the grains on Wednesday. To arrive at this opinion, we examined grain prices as far back as 1968 (30 years 
of data), Bonds to 1977 (21 years of data), and the S&P since they began trading in 1982 (17 years). In 
short, we rolled the dice enough to draw some reliable conclusions and observed enough data to determine 
there are biases; price is not solely motivated by a drunken sailor's random walk through the pages of the 
Wall Street Journal. 
From this research, we have a leg up on other traders, an advantage in the game, and a window of 
opportunity to focus on when trading. It is not how often you trade that makes you a winner, after all any 
fool can trade every day of the week. Old punters like myself know it is how often you don't trade, how 
selective you are, that will lead to a successful career. 
Astute traders are probably already asking the next question I will now answer, "If there is a bias to 
the TDW could there be a bias to the Trading Day of the Month?" 
The answer is yes, and here comes the proof. The following results were arrived at by buying/selling 
on the opening of the trading day of the month