A major synchronized cyclical low and buying opportunity across the S&P 500, NASDAQ, Dow, and Russell is projected around Tuesday, October 27, ahead of the US presidential election. At the trough, focus capital on the index showing the strongest relative strength and greatest resistance to selling pressure during the decline.
Detrended seasonal cycles isolate short-term periodic signals by removing longer-term trends, preventing distortion and enabling precise measurement of amplitude and timing.
Jeff Hirsch uses MACD crossovers as a technical filter to execute the "Best Six Months" strategy. By combining technical momentum with the calendar, he avoids buying into falling markets or exiting strong spring rallies prematurely.
► Timing Windows: The "Best Six Months" run Nov 1–Apr 30 for the Dow/S&P 500, extending through June for Nasdaq. Daily monitoring begins October 1 for buys and April 1 (June 1 for Nasdaq) for exits.
► Indicator Setup: Daily charts with 12-day fast EMA, 26-day slow EMA, and a 9-day signal line.Buy Signal: On/after Oct 1, a positive MACD crossover (12-day EMA crosses above 26-day EMA) triggers the Seasonal Buy Signal, confirming downside momentum has ended before entering equities.
► Sell Signal: On/after Apr 1 (or Jun 1 for Nasdaq), a negative MACD crossover triggers the Seasonal Sell Signal, shifting exposure to cash or defensive sectors.
► Why It Works: It eliminates buying into October market crashes, captures extended spring upside, and reduces drawdowns compared to strict calendar dates or buy-and-hold strategies.
► October ranks 8th of 12 by average across all years (avg. +0.6%). July is best (avg. +1.7%) and September is worst (avg. −1.1%). In midterm years it ranks 1st (avg. +2.4%) and September is worst (avg. −1.5%). In the nine years ending in 6 it ranks 3rd (avg. +2.1%). March is best (avg. +3.1%) and September is worst (avg. −1.5%).
► October was up in 58 of 98 years (59.18%, avg. +0.6%, med. +1.1%). In midterm years, 16 of 24 were up (66.67%, avg. +2.4%, med. +2.5%). From the September close through Oct 15 (Thu), all years were up in 63 of 98 (64.29%, avg. +0.7%, med. +1.3%). Midterm years were up in 17 of 24 (70.83%, avg. +2.1%, med. +2.1%).
► Oct 12 (Mon) is Columbus Day. The stock market is open. The bond market is closed. It is the first session of expiration week, up in 54 of 98 years (55.10%, avg. +0.1%, med. +0.1%). In midterm years, 14 of 24 were up (58.33%, avg. +0.2%, med. +0.3%).
► Oct 12 (Mon) through Oct 16 (Fri), expiration week, was up in 59 of 98 years (60.20%, avg. flat, med. +0.5%). The average is flat because a few weeks were large losses. In midterm years, 16 of 24 were up (66.67%, avg. +0.5%, med. +0.6%). Oct 16 (Fri), expiration day, was up in 47 of 98 years (47.96%, avg. −0.1%, med. flat). Midterm expiration days were up in 12 of 24 (50.00%, avg. +0.1%, med. +0.1%).
► Oct 27 (Tue) opens the midterm stretch that runs through Nov 10 (Tue), 2026, up in 22 of 24 midterm years (91.67%, avg. +2.6%, med. +3.2%). The misses were 1930, −12.6%, and 1994, −0.3%.
► Election day is Nov 3 (Tue), 2026. Oct 27 (Tue) through Oct 30 (Fri) alone was up in 15 of 24 years (62.50%, avg. +0.6%, med. +0.6%).
► May through October, which this month closes, was up in 65 of 98 years (66.33%, avg. +2.5%, med. +3.6%). In midterm years, 12 of 24 were up (50.00%, avg. −0.6%, med. +2.1%). The average is down because a few of those years were large losses. The median year was up.
► Last quarter Oct 3 (Sat), 9:25 a.m. EDT. New moon Oct 10 (Sat), 11:50 a.m. EDT. First quarter Oct 18 (Sun), 12:13 p.m. EDT. Full moon Oct 26 (Mon), 12:12 a.m. EDT. No solstice and no eclipse this month.
However, the leak precedes the rally. Both the all-year seasonal map and the midterm composite sag from the autumn equinox through month-end. September 27 falls directly within that washout zone, with the midterm path averaging a trough near September 30 (−1.3% from mid-September levels). Because September 27, 2026, falls on a Sunday, the active trading window shifts to Friday the 25th and Monday the 28th. That cluster represents a dip-buying opportunity, not an immediate breakout zone.Seasonality Midterm October Map for the S&P 500: Midterm years leave September weaker than the all-year path—then pull away from October 1 through mid-November. S&P 500 calendar-day path from the September 14 close: crimson is midterms 1950–2022 (+3.0% in October, 74% up); navy dashed is all years 1928–2025. Late September is still the washout (window trough ~Sep 30). 2026 has already rallied +21% off the March low, so this is a Q4-bid midterm, not a crash-then-rally analog, unless 7,550 breaks.
Presidential Election Cycle in US StocksYale Hirsch is most widely credited with detecting and popularizing the Presidential Election Cycle in US stocks. A market historian, he introduced the pattern in the first edition of his Stock Trader's Almanac in 1967 and refined it in later volumes. Drawing on decades of data, Hirsch showed that equity returns tend to be weaker in the first one or two years of a presidential term and stronger thereafter—especially in the third, pre-election year—as administrations shift toward stimulus to support re-election prospects. Although related political-business-cycle ideas appear in economics (notably William Nordhaus's 1975 model of pre-election stimulus), the specific documentation and popularization of the stock-market version by year of the term belong to Hirsch's Almanac work; academic testing of the equity pattern followed mainly from the 1980s onward.Overlap with the Kitchin and Broader Business CyclesThe presidential cycle is a fixed four-year political calendar. It overlaps in length with several economic cycles of roughly three to five years, yet the mechanisms differ and the patterns should not be treated as identical. The Kitchin Cycle, identified by Joseph Kitchin in 1923, is a short inventory-driven business cycle averaging about forty months. Firms over-order in expansions, then destock, producing production swings that feed into the broader economy. Because its typical span nearly matches a presidential term, many observers refer to the dominant four-year equity rhythm as both the Kitchin and the presidential cycle; some studies even treat the observed market pattern as encompassing both. The crucial distinction remains that Kitchin is an economic-inventory process while the presidential cycle is a political-calendar effect tied to election incentives. They can reinforce each other—late-term stimulus coinciding with inventory rebuilding—or drift out of phase because one is rigid and the other variable.The presidential cycle also interacts with the general Business Cycle through policy timing: adjustment-oriented measures early in a term give way to growth-oriented stimulus later. Stocks, as a leading indicator, historically show weaker average returns and more frequent recessions in the first half of the term and stronger performance, especially in year three, in the second half. Empirical work finds that standard business-cycle variables do not fully account for the presidential return pattern; the equity effect persists as something of a residual puzzle.Links to Hurst CyclesTechnical cycle analysis supplies a further parallel. J. M. Hurst's model organizes markets into a hierarchy of harmonically related nominal periods governed by commonality, synchronicity, and proportionality. The 54-week cycle is an intermediate member; longer relatives include the 40-week, 18-month, and especially the 54-month (approximately 4.5-year) cycle. Four 54-week periods nest into roughly 4.15 years, close to a presidential term, and Hurst practitioners often map the four-year political pattern onto their 4-to-4.5-year or 46-to-54-month rhythms. Historical studies, including those referencing Edward Dewey, note a roughly 46-month cycle of high regularity that aligns in period with the presidential timeframe. The political calendar can help phase or contextualize Hurst troughs and peaks, yet pure Hurst analysis remains grounded in price action and nested harmonics rather than external politics; the two tools are complementary, and presidential turning points coincide with major Hurst lows only intermittently.Interaction with the Decennial CycleA still longer calendar regularity, the Decennial Cycle, interacts with the presidential pattern through systematic overlaps. First detailed by Edgar Lawrence Smith in the 1930s and later popularized in Hirsch's Almanac, the decennial pattern tracks average performance by the year's ending digit. Early-decade years (ending in 0, 1, or 2) tend to be softer—a "decade hangover"—while mid-decade years, especially those ending in 5, have been almost invariably positive and often strongly so; later years are more mixed. Because ten is not a multiple of four, the two cycles nest in shifting combinations: every decade contains two full presidential terms plus part of a third, so a year's place in the presidential sequence systematically aligns with particular ending digits. Years ending in 5 frequently fall in the first or third presidential year—both historically stronger—helping to amplify the mid-decade strength. Early-decade softness often coincides with post-election or transitional periods that overlap the weaker half of the presidential cycle. Practitioners like Ned Davis therefore treat the two as additive filters: a year that is both a strong presidential year (particularly year three) and a favorable decennial year is viewed more constructively, while alignment of weak slots raises caution.
Complementary FrameworkTaken together, these patterns form a nested set of calendar and economic regularities. Length similarities produce natural correlations and frequent joint discussion, especially around the four-year rhythm shared by the presidential, Kitchin, and certain Hurst cycles. The presidential cycle functions as a political overlay that can influence or coincide with inventory-driven, business-cycle, and pure price-based rhythms, particularly around policy timing and major turning points. The decennial cycle supplies an additional independent calendar layer that modulates the four-year pattern at predictable intersections. None of the cycles causes the others; each is an empirical tendency best used as a parallel lens. Real markets approximate the historical averages but never duplicate them exactly, because exogenous events, monetary policy, and larger forces continually interact with and sometimes override the calendar regularities.
S&P 500 80-day cycle (primary): Starts at the Aug 20 trough. Wavelength 71 days. High already in on Aug 28, at 11% of the wave—left, not in the middle. Sep 21 did not beat it. Printed highs and lows as of Sep 21. "Expected" repeats that cycle's last translation. Oct 31 is the 80-day low only—the 20-day low before it is Oct 17. 80-day FLD 7,790 is still lost. The 20-week from this same Aug 20 trough does not bottom until Jan 9, 2027 (timing schematic, not a price forecast).
S&P 500 80-day cycle (alternate): Same Aug 20 start. On this count the low is Oct 5, not Oct 31, and Oct 5 is not a high. Sep 21 did not beat Aug 28.No crest between Sep 21 and Oct 5. The Nov 10 bounce is drawn smaller because this alternate cycle count does not expect it a new high (timing schematic, not a price forecast).
DJIA 80-day cycle (primary): Starts at the Aug 20 trough. Wavelength 71 days. High already in on Aug 28, at 11% of the wave—left, not in the middle. Oct 30 is the 80-day low only—the 20-day low before it is Oct 17. The 20-week from this same Aug 20 trough does not bottom until Jan 9 (timing schematic, not a price forecast).
DJIA 80-day cycle (alternate): On this count the low is Oct 4, not Oct 30, and Oct 4 is not a high. The Sep 22 bounce already failed. Next week is a bounce only if Sep 26 holds above 51,172. The high of that bounce is Oct 4, and it is a sell. It is a short, left-translated 40-day high, about a week, and it does not repair 52,364. If 51,172 breaks, next week is down into Oct 4. Then Oct 4 is the low near 50,000, not the high (timing schematic, not a price forecast).
► Sep 1–8 (Tue–Tue) = Trading Days 1–5: Sideways-to-up / modestly higher. Most midterm series (especially Russell 2000 and DJIA) rise, with several peaking near +0.5% to +1.0%.
► Sep 9–17 (Wed–Thu) = TD 6–12: Sideways to mildly fading. Early gains are largely held or only slowly given back. S&P 500 midterm often remains the strongest (still near its peak), while NASDAQ and Russell lines begin drifting lower.
► Sep 18–25 (Fri–Fri) = TD 13–18: Steady decline. The mid-month advantage disappears; indices trend lower and most move into negative territory.
► Sep 28–30 (Mon–Wed) = TD 19–21: Accelerating sell-off / sharp weakness. Losses deepen, particularly in NASDAQ and Russell 1000 (historically finishing around –1.6% to –1.8%). Russell 2000 also shows a late plunge.
Average S&P 500 total-return path (indexed to 100 on midterm Election Day) for all midterm years since 1970 (1970–2022), spanning roughly ±6 months. X-axis centers on Election Day (first Tuesday in November); y-axis tracks cumulative total return. The average line rises in the final ~22 trading days before the election (= October 2, 2026) and continues higher afterward (+14.1% average in the following six months). A separate “Lost Control” series (party loses presidential trifecta) lags the broader average post-election (+10.4% vs. +16.1%).
S&P 500 Strategy: Go long during bullish zodiac phases and short during bearish phases; hold through phase end with a 2% stop-loss (1% tighter stops improved results). Starting with $1,000, a 10 year walk-forward case study produced $5,410. The win rate exceeded RSI, Stochastic, and MFI, slightly outperforming Stochastic. Low-win-rate signs (e.g., Gemini at 39%) can be omitted.
Augmented version: Enter long only when Stochastic (14,1,3) ≤20 during a bullish phase; enter short only when ≥80 during a bearish phase; hold through the full phase. Python backtest: $18,571.79 over two years with a 66.67% win rate.
Using 10 years of S&P 500, Mercury retrograde periods frequently preceded major pivots, with post-retrograde directional flips.
