MAPLearning map for this lesson
We build it up in three steps. At the end of each step, a 'Quick check' lets you test your understanding before moving on.
Before you startThis is the last lesson. The 6-month cycle was covered in Lesson 2, the turn-of-month effect in Lesson 4, the half-year sector rotation in Lesson 5, and the three-line rule of 'condition, action, invalidation' in Lesson 7. In particular, we use the three-line rule from Lesson 7 as is in this lesson's practice.
- 1 · One number · 2 · InsightSix methods, one thing in common
You can tell the six masters' calendar signals from their confirmation filters, and explain the shared principle: 'the calendar sets the window only; trade after confirmation.'
- 3 · Data · 4 · TwistEven the masters' methods have a shelf life
You can tell whether a performance figure is a claim or a verified result, and check the numbers showing that Fosback-style calendar timing weakened after 2006.
- 5 · One actionTurn it into your own three lines
You can pick one of the six methods and rewrite it as the 'condition · action · invalidation' three-line rule from Lesson 7.
Building knowledgeConcepts built so far: 39 · 5 added in this lesson
Seasonality = tilt in probabilityMean, median, win rate, sample sizeComparing against a baselineSince 1950 vs the last 20 yearsNot 'sell' but 'lower your expectations'Nov–Apr vs May–Oct half-year comparisonHalloween effect (36 of 37 countries)The gap vanishing over the last 10 yearsFixed dates vs MACD-confirmed switchingSanta rally's 7 trading days vs the baselineJanuary barometer 2×2Small-cap January effect (excess return)The more famous the pattern, the sooner it weakensYear-end/new-year = a filterTurn-of-month effect 90% → 23%When the structure changes, the pattern disappearsThe shorter the cycle, the sooner it weakens4-year presidential cycle and a sample of 18–19One-line cause testHalf-year sector rotation (Stovall)Seasonality created by real demandExcess return vs the market and number of winsTech exception, defensive reversalMonthly gold futures record and the limits of continuous futuresKOSDAQ weakness in Jun–Oct, strength in JanuaryMajor-shareholder capital gains tax and the year-end record dateFading of the January excess return (since 2016)Patterns made by rules are changed by rulesChecking December individual net selling yourselfPatterns weaken once known (−58% after publication)Five questions before using oneOnly patterns with a built-in cause remainFilters adjust risk, not returnsCondition, action, invalidation: three linesThe six masters' methodsNEWCalendar for the window, trend and flows for confirmationNEWWeak windows: cash, bonds, defensive sectorsNEWCheck whether performance figures are claimsNEWEven the masters' methods have a shelf lifeNEW
1 · One numberThey stayed in the market only a third of the time
11.7% vs 10.6% a yearNorman Fosback's calendar timing vs buy and holdHulbert's 25+ year tracking (cited) · stock holding days were about one third of the total
Norman Fosback's Seasonality Timing System, built on research from the early 1970s, holds stocks only on the trading day before market holidays and around the turn of the month, and keeps money in a money market fund (MMF) on all other days. In Hulbert's 25+ year record tracking investment newsletter performance, this method averaged 11.7% a year versus 10.6% for buy and hold, and the days it held stocks were cited as about one third of the total (Crossing Wall Street, 2008-11-18).
Fosback is not the only one who made a name with the calendar. Yale Hirsch launched the Stock Trader's Almanac in 1967 and popularized terms such as 'Best Six Months,' the January barometer and the Santa rally, and his son Jeffrey Hirsch still carries on as editor. This lesson looks at the methods of six such masters side by side.
Quick checkWhen you see the number "11.7% a year," what is the first label you should attach to it?
Who measured it, over which period, and how. This figure is cited as the result over the period Hulbert tracked. Whether it held in other periods has to be checked separately, and we will check it ourselves in 3 · Data.
2 · InsightSix methods, one thing in common
The six masters look at the calendar in very different units. Some look at half-years, some only at month-ends and the days before holidays, some set dates for each sector, and some slice it down to single days. But when you line the six methods up, one thing in common stands out: the calendar only decided "when to look," and actual trades were executed after confirming trend, flows and statistics.
The masters' shared structure
1. CalendarSets the window only
e.g. Nov–Apr, turn of the month, sector-specific dates
→
2. ConfirmationScreens by trend, flows and statistics
e.g. , COT, win rate and distribution
→
3. ActionAdjusts the weight
In weak windows: cash, bonds, defensive sectors
A diagram of what the six methodologies summarized in the course plan have in common.
The answer for weak windows was similar too. None of the six methods has a rule saying "short in the weak season"; they moved into cash or bonds, switched to defensive sectors, or stopped new buying. This means seasonality was used not as a tool for placing directional bets but as a tool for reducing risk. The sentence this course has repeated since Lesson 1, "the calendar is a filter, not a signal," is built right into the masters' methods.
Easy analogy · The fishing calendar
Anglers check a tide calendar to decide which days to go out to sea. Even so, before launching the boat they check that day's wind and waves again, and if the weather is bad they stay in, even when the calendar says it is a good day. The masters' calendars only told them which days were 'worth heading out.'
Quick checkWhat action did all six masters choose in weak windows?
Not short selling, but cash, bonds and defensive sectors. In weak seasons they moved to reduce the risk of loss, and did not use methods that bet on declines.
3 · DataComparison table of the six methods
Click a card below to see each method's calendar signal, confirmation filter and response in weak windows. Performance figures are marked with claim. It means the figure was stated by the person or the cited outlet and has not been independently verified by this course.
Six masters of seasonality and their methods
Click a card.
Based on the sources cited in the course plan: Stock Trader's Almanac introduction and strategy, CXO Advisory (Sy Harding verification), Crossing Wall Street (2008-11-18), Stocks & Commodities (2002-12), CFRA-Stovall (2019), Global X Canada (2015-08-14), Wealth Professional (2023-09-20), StockCharts (2023-03), Picture Perfect Portfolios (secondary source), Seasonax (2018-02-02). The original texts were not separately verified.
Looking at the table again, the six methods divide by the unit of the calendar. Hirsch, Harding and Stovall use half-years (an extension of , Lessons 2 and 5), Fosback and Kaeppel use a few days (, Lesson 4), Thackray uses sector-specific date ranges, and Williams and Speck use weekly and daily probabilities. And except for Stovall and Fosback, all of them added one more layer of confirmation on top of the calendar, such as MACD, technical analysis, the and event filters.
Quick checkWhat is the biggest difference between Hirsch-style 6-month switching and Stovall-style half-year sector rotation?
Whether you leave stocks in the weak window or just switch sectors. The Hirsch method moves into bonds or cash in May–Oct, while the Stovall method stays in stocks and switches to defensive sectors such as consumer staples and health care.
4 · TwistEven the masters' methods have a shelf life
Most performance figures are claims. Fosback's 11.7% a year, Sy Harding's cumulative 213% in 1999–2012 (versus 93% for the Dow) and Larry Williams' 1987 contest result are figures stated by the people themselves or by the cited outlets, and have not been independently verified. And even if the numbers are right, once the period that produced them has passed, the basis for the method can weaken.
The basis of Fosback's method was the pattern that gains cluster around the turn of the month and the trading day before market holidays. As we saw in Lesson 4, this pattern weakened sharply after 2006. So what happens if we apply the same rule today? We calculated a simplified Fosback-style rule ourselves from S&P 500 daily closes. It holds stocks only on the last trading day of the month, the first 3 trading days of the month, and trading days followed by a weekday market closure before the next trading day, and keeps cash (0% interest) on all other days.
Fosback-style calendar timing vs buy and hold (S&P 500)
Calendar timing (about 21% of days held)Buy and hold
Tap a bar.
Calculated directly from Yahoo Finance ^GSPC daily closes (price index, excluding dividends and trading costs, assuming 0% interest on cash periods; the actual Fosback method earns MMF interest, so this calculation is unfavorable to calendar timing). Trading days before market closures were estimated from gaps in the daily data. 1950–1989: 10,053 trading days; 1990–2005: 4,036 trading days; 2006–2025: 5,031 trading days. Retrieved 2026-10-09.
The results flipped from period to period. In 1950–1989, holding stocks on only 21% of all trading days still produced an average of 8.51% a year, higher than buy and hold (7.94% a year). But in 1990–2005 it was 4.85% vs 8.21% a year, and in 2006–2025 it lagged badly at 2.70% vs 8.88% a year. Adding interest on cash would hardly close the gap of the last 20 years. Once the pattern became widely known and the market structure changed, the basis for the method disappeared. As we saw in Lesson 7, patterns whose cause is not rooted in rules or real demand weakened first.
Common misconceptions
"It's a famous master's method, so it still works?" No. Fosback-style calendar timing returned 2.7% a year in 2006–2025, far behind buy and hold (8.9%). The period in which a method was created can differ from today.
"A big performance number means a verified method?" No. Most are the person's own claims or secondary citations. You need to apply the same rule to a different period (an ).
"The masters traded on the calendar alone?" No. Most went through a confirmation step such as MACD, COT, technical analysis or event filters.
Quick checkA calendar strategy beat buy and hold in 1950–1989 but lagged badly after 2006. What should you do first before using it?
First check whether it also worked over the last 10 years. If it fails 'is it still valid over the last 10 years?' or 'can you explain the cause?' among the five questions in Lesson 7, don't use it for trading.
5 · One actionTurn a master's method into your own three lines
This is the final assignment. Pick one of the six methods and rewrite it as the three-line rule of condition · action · invalidation from Lesson 7. When you choose a master, a default sentence is filled in, and you can edit it freely. The goal of this assignment is not to follow a master's method as is, but to convert it into your own account's position size and invalidation condition.
Practice · Turn it into a three-line rule
What you enter is stored only in this device's browser. The default sentences are educational examples and do not recommend any specific trade. Once you have made a rule, check it against the five questions in Lesson 7, and first confirm that it also worked over the last 10 years.
The calculations in this course are based on price indexes, so dividends, trading costs and taxes are left out. Once costs and taxes are included, the edge of a switching strategy shrinks further. There is no guarantee that a method that worked in the past will keep working.
Quick checkWhen you turn Hirsch-style 6-month switching into three lines, what condition could go in the invalidation line?
A condition that says you don't raise your weight when the trend has broken, even in a strong window. For example, write "If the S&P 500 is below its 200-day moving average, don't apply it even in Nov–Apr."
The one sentence left from Lesson 8The calendar sets the window only; trade after confirmation