The Little Book of Stock Market Cycles
Jeffrey Hirsch · 2012 · 155 pages
Hirsch (of the Stock Trader's Almanac family) packages two real calendar effects, the Best Six Months and the four-year presidential cycle, into one switching strategy. The seasonal data is solid through 2012. The long-range market forecast bolted onto it already missed.
The one idea
The book's engine is the Best Six Months Switching Strategy: own the Dow from November 1 through April 30, move to cash or bonds the other six months. Hirsch shows that from 1950 to 2012 the November-to-April months gained 14,654.27 Dow points across 62 years (37 up, 25 down), while May through October lost 1,654.97 points (48 up, 14 down) over the same span. A $10,000 investment compounding only through the winning half of the year turned into $674,073 more than the same money left in stocks year round, because the losing half added almost nothing.
Layered on top is the four-year presidential election cycle. Hirsch's claim, traced back to Andrew Jackson's first term, is that the last two years of a presidential term (preelection and election year) produced 724.0 percent in combined Dow gains across 44 administrations since 1833, against 273.1 percent for the first two years. Bear markets cluster in the post-election and midterm years while administrations 'prime the pump' heading into the next vote. The book's real contribution is combining the two: instead of switching in and out every six months for four years straight, you trade the Best Six Months only in the post-election and midterm years, then just hold from October 1 of the midterm year straight through to sometime after the following April, skipping two full switches and nearly tripling the plain Best Six Months return on paper.
A third layer, borrowed from Gerald Appel, uses MACD as a timing trigger around the October/November and March/April switch windows rather than trading the calendar date mechanically: an 8-17-9 MACD for entries, a 12-25-9 MACD to confirm an exit after a strong run. The rest of the book (roughly half its page count) is a month-by-month tour of smaller calendar quirks: options expiration ('witching') Fridays, the worst-day-of-week pattern, and holiday-specific moves like the Santa Claus rally and pre-Good Friday drift.
What it actually teaches
Hirsch's own trading process, laid out mainly in the final chapter, runs in this order.
IDENTIFY THE BEST SIX MONTHS WINDOW
Be long the Dow (or an S&P/NASDAQ proxy) from November 1 through April 30, and out of stocks from May 1 through October 31. Hirsch's own backtest: since 1950 the winter half gained 14,654.27 Dow points versus a 1,654.97-point loss in the summer half. On NASDAQ since 1971 the effect is bigger: an eight-month November-to-June window turned $10,000 into $394,337 versus a $3,196 loss in the four-month gap from July to October.
LOCATE YOURSELF IN THE FOUR-YEAR CYCLE
Classify the current year as post-election, midterm, preelection, or election. Since 1913 the Dow has fallen 20.9 percent on average from its post-election-year high to the following midterm-year low. Since 1914 it has gained 48.6 percent on average from that midterm low to the preelection-year high. No down Dow year has occurred in a preelection year since 1939.
STACK THE TWO CYCLES INTO FOUR TRADES
Instead of switching every six months for four straight years, buy at the start of the Best Six Months in the post-election year and sell at the end; buy again at the start of the Best Six Months in the midterm year and sell; then simply hold the position from October 1 of the midterm year through some point after April 1 of the preelection year, skipping the switches in the preelection and election years entirely. Hirsch presents this as four trades every four years that beat the plain Best Six Months return, with lower transaction costs and fewer taxable events.
CONFIRM ENTRY AND EXIT WITH MACD
Around the October/November and March/April windows, use MACD rather than a fixed calendar date. Appel's 8-17-9 MACD (an 8-period fast average, 17-period slow average, 9-period signal line) generates the buy signal; a 12-25-9 MACD confirms a sell after a strong bullish run. A buy signal fires when the MACD line crosses above its signal line while the indicator sits below the zero line (oversold); a sell fires on the opposite cross above zero (overbought). The book does not specify whether to run these settings on daily or weekly bars for the seasonal switch, so that choice is left to the reader.
LET DAY-OF-WEEK TENDENCIES PICK THE EXECUTION DAY
Since 1990, Monday and Tuesday have been the most consistently bullish days for the Dow and Thursday and Friday the most bearish; Monday and Tuesday combined gained 11,992.54 Dow points against a 2,677.45-point loss for Thursday and Friday. Use that pattern to choose which day within the switch week to actually place the trade, rather than executing on whatever day the calendar or the MACD signal happens to land.
What it looks like Monday morning
A reader who takes this seriously stops trading a flat calendar (sell in May, buy in November on autopilot) and instead marks two things on a real calendar before doing anything: which of the four presidential-cycle years the market is currently in, and whether that year is scheduled for a Best Six Months switch under Hirsch's four-trades-per-cycle table or should just be held through. From there, the mechanical part is small: set a MACD alert (8-17-9 for buys, 12-25-9 for sell confirmation) around the late-October and late-March windows instead of committing to a fixed date, and if there is a choice of which day that week to place the order, favor Monday or Tuesday over Thursday or Friday.
The bigger change is treating the rest of the book, the witching-day and single-holiday chapters, as color rather than a plan. None of those chapters carries a position-sizing or entry rule; they are context for why a given week might be more volatile, not something to trade on their own.
| Cycle year | Worst Six Months (May-Oct) | Best Six Months (Nov-Apr) |
|---|---|---|
| Post-election | Sell | Buy |
| Midterm | Sell | Buy |
| Preelection | Hold | Hold |
| Election | Hold | Hold |
Hirsch's combined strategy: trade the Best Six Months switch only in the post-election and midterm years, then hold the position from October 1 of the midterm year through sometime after April 1 of the preelection year, skipping the next two switches entirely.
There is no such thing as a perfect trading strategy, tactic, or methodology, but our Best Six Months Switching has an undeniable track record.Jeffrey Hirsch, The Little Book of Stock Market Cycles
Where it fails
- The book's own long-range forecast already missed, badly, on timing. Hirsch's headline 'super boom' call, made in May 2010 and repeated in this 2012 edition, projects the Dow moving sideways in a 7,000 to 14,000 range through 2017 or 2018 before rallying to 38,820 by 2025. The Dow actually crossed 38,820 in December 2023, roughly two years ahead of the book's own schedule, and the sideways phase ended years earlier than forecast. This is the same cycle-reading apparatus the book asks you to trust for the four-year presidential timing, and on its one big, checkable, dated prediction it was wrong on timing by years.
- The Best Six Months backtest stops at May 2012 and is never updated in this edition. Every headline number (14,654.27 Dow points, the $674,073 compounded gain, 37 up years versus 25 down) covers 1950 through 2012. A reader applying this in 2026 is extrapolating 14 years past the book's own sample with no way to check from the text whether the edge held, widened, or decayed since.
- The four-year cycle case studies are labeled after the fact, not predicted in advance. Tables 5.1 through 5.4 walk year by year through outcomes like '1925 Coolidge: Peace and prosperity. Hallelujah!' and '1929 Hoover: Worst market crash in history,' both preelection or post-election years. That is a history lesson, not a forward rule: the book gives no test a reader could apply in real time to tell a 'textbook' cycle year from an exception like 2001 or 2008, both of which it excuses as crisis-driven rather than counting against the pattern.
- The MACD confirmation step is underspecified. Chapter 14 gives Appel's 8-17-9 and 12-25-9 settings but never states what timeframe (daily or weekly bars) to run them on for the seasonal switch dates, and it explicitly warns 'one should not rely on a single indicator' without naming which others to pair it with. As written, the confirmation step cannot be mechanically replicated from the book alone.
- Roughly half the book is calendar trivia with no trading rule attached. Chapters 7 through 12 work month by month through options-expiration witching days, Martin Luther King Day, Good Friday, Rosh Hashanah and Yom Kippur, and Labor Day, mostly reporting how many of the last N years a given single day was up or down. None of these chapters produces a threshold, a position size, or a rule beyond 'this day has historically leaned up or down,' and they pad a book whose two real ideas (Best Six Months, the four-year cycle) are fully covered in two chapters.
Who it is for
Buy it if
You already trade or invest using an index ETF and want a documented, mechanical seasonal overlay (when to be long the market versus in cash) backed by a 1950-2012 dataset, and you are willing to treat the political-cycle and MACD layers as loose confirmation rather than precise rules.
Skip it if
You want a forward-tested or post-2012 validated edge, a book that actually specifies indicator timeframes and position sizing, or credibility on long-range market forecasting; the book's own dated forecast for the Dow was off by roughly two years on the one call a reader can check against history.
As an Amazon Associate I earn from qualifying purchases. It costs you nothing extra, and it does not change what gets recommended or what the verdict says.