the equator on March 21st. Fall begins on this date and stocks make important changes in trend. 42V5ber8ih is 46 days from September 23rd and equals 45 degrees Tou will find, buy checking the records that many important tops and bottoms and changes in trend occur on this date. December 22nd is 91 days from September 23rd, and six months or 180 days from June 22nd, This is where winter begins and it is important to watch for important changes in trend."

From this, it is obvious that Gann was using the year as a cycle, and the geometric divisions as cycles for change. Seasonal studiee from 1897 show the period from the 27th of July to the first week in August as having a definite bullish bias. And that Fridays have the highest probability of being up days and Monday the highest probability of being down days. August has a very high probability of being an up month. September, historically, has a bearish bias.

Another, interesting and tradable aspect of this is, the phenomenon of days preceding holidays. Using all days before holidays, including times during bear markets since 1897, the following is the percentage of times the market advance before certain holidays: Labor Day- 82% of the time. Independence day-77% of the time. Memorial Day- 74% of the time. Thanksgiving- 59% of the time Christmas- 73% of the time. New Years- 72% of the time, Good Friday- 61% of the time. Presidents Day-no significant percentage. If the percentage is woHced out for all holidays combined, it comes to almost a 70 percent likelihood that the day preceding a holiday will show a positive close. Of course this is a well know fact among professional tradere, and they will often accumulate positions a day or two earlier, and then sell these the afternoon of the day before the holiday.

THE CYCLES OF THE INNER YEAR

The year is a cycle. There are 360 degrees in that full revolution. We divide that up, based on the seasons, into 90 day periods, and we divide it into months,based on the lunar cycles. So, as divisions of a year are concerned, we have the following important time periods for moves in the market. Ninety days, 180 days, 270 days, and 360 days (one year) from the seasonal changes from significant highs or lows in the item that is being analyzed. We also have divisions by months, or 30 days, 60 days, etcetera. Making one more step inward with the seasonal aspect gives us a 45 calendar day count.

From all of this are the baeic calendar day counts which we will use in our examination of price movement through time. Counts of important highs and lows (within the year) are 30, 45, 60, 90, 135 (90+45), 150, 180, 210, 225, 270, 315, 330, and 360. These could also be extended outward, which I only do in 90 day increments. These calendar day counts should be placed on you daily charts from significant highs and lows. Obviously, you

day c60TfR~and market day counts is quite close to producing the same date. As 11 market days oan be 15 calendar, 22 market days is almost exactly 30 calendar days, and 33 market days is very close to 45 calendar. Once you study these cycles for change in trend youwill understand why the calendar day counts or cycles, will a better understandimi of timer

I no longer put trading day counts on my daily charts and I rely on the overlays for that purpose. You can also determine where the calendar day counts are, by moving the overlays to a 45 degree angle.

These counts start with the exception day, the high or low day being counted as zero. That high or low day is the last day, or the completion day of the cycle, and the next day, is the first day of the next cycle. So, a low or a high should be treated as the zero day, with the day after being day one of the next cycle.

Counting from all of the swing lows and highs, using the trading day counts, will often produce a grouping of those counts on the same day, where you find as an example, a 22 count, an 11 count, a 56 count, and a 90 count. Price might be 22 1/2 points from high or low. When you see a situation like this, specially with a 45 or 90 count involved, it is usually a definite cyclical vibration. Under the right circumstances you will be able to find a certain day on a daily chart from whioh every 11, 22, 33, 45, 56, and 90 market day count changes the trend (minor swings). And, in hindsight, this will look very intriguing, but the true

should not always expect to see exact changes in the market- to the day- from the calendar day counts, but should use them as a road map to measure the duration of moves, and also as general time periods to look for more significant changes in direction.

What you are looking for from the calendar day cycles are the moves that go up or down for a full 90 calendar days, or a longer, period of time. It becomes much easier to trade if you know the market has a probability to up for 90 days. When you see this happen, you will get trading situations from the counts which are "setups." As you will see m the exercises, you can also find 45 day and 30 day moves fairly often.

Gannin his work said, "you can look f"r top on the 90th to 99th day V This is somewhat misleading. If you are to tHat to every move in the markets, you will find that it seldom works. There are circumstances under which it works nearly all of the time though, and these are the 90 day blow off moves. In fact, if you looked at every significant high before the onset of a bear campaign, you will find a significant low 90 days back from that high, a vast majority of the time.

Another set of counts which Gann used in his work on daily charts, and whioh you osn look for changes in market direction are the trading day counts. These trading day counts come from the

understanding of cyclic vibrations lies in those of the calendar days.

Obviously to people who have studied Gann, what I have just described is the square of 90, and its time use on the daily trading day chart. The square of 90 is one quarter of the 360 degrees of the Earth traveling around the sun. The price divisions of the square of 90 are the same as the time divisions, and should be monitored closely.

This brings up the question of which should I trade? The market day or trading day cycles. First, when viewing the market day cycles, you are in many instances looking at calendar days. Only the counts are different- as 33 market is 45 calendar. There are many times, as you will see in the exercises, that the angles generated on market day charts along with the angles on the square of 90 are very important when used on market day charts. But you are really trading the calendar day cycles. This question should answer itsejf by the end of this book.

On a weekly basis, the counts from 90 are also important, although I pay much more attention to a complete 90 or 180 weeks. There will usually be a quick sharp move as these cycles expire, and if qualified with other work, could be a change in trend. Although on a weekly chart use 13 weeks (90 days), 26 weeks (180 days), 39 weeks (270 days), 45 weeks (315 days) and of course 52 weeks, and so on. This is the square of 52 which is discussed later in the text.

Gann described the 7 week period as a death zone. The reason for this is many markets and individual stocks and commodities will start or terminate fast moves with this 45 calendar day cycle.

A stock makes a low or high and then begins to accelerate in the other direction, A significant movement against this acceleration occurs, with the end of this corrective move coming very near 45 calendar day count in time from the start of the move. Then the stock begins to accelerate again, and continues to accelerate, with movements against the trend becoming shorter in time and less in prioe as the movement continues. Obviously, what I am describing here is a 90 day blowoff. The key here is the death zone week is also a significant time period as 52 weeks or 78 weeks.

There are many setups from this cyclical work on the weekly charts. A certain week may be 26 weeks from a bottom and 13 weeks from a top and move down in price into that tima. That is an excellent time to look for a change in trend, or a short term trade, especially if you can find other cycles lining up, possible time is 180 weeks from a previous top, and 208 weeks from a bottom. If I see that type of situation on a weekly chart, I will then check the monthly charts to determine support and resistance, and if that chart presents a good possibility, I will go to the daily chart to find the day to position, but more on this later.

Finally, the monthly charts are the outward extension of the cycle of the years. From significant highs and lows on the monthly charts the method which I now use is to count by 6 months, giving