Seasonality & Calendar Patterns

The calendar has patterns

Markets are supposed to be efficient. If everyone knew that stocks rally in January, they'd buy in December, moving the rally earlier until the anomaly disappeared. And yet — calendar effects persist. Not as reliably as they once did, and not always strongly enough to trade on their own, but they keep showing up in the data.

Kaufman and Pring both document these patterns with decades of evidence. The patterns are real. The question is whether they're still tradeable.

Sell in May and go away

The most famous seasonal pattern. Kaufman documents the data:

Historically, the stock market's strongest performance period runs from November through April. The weakest period is May through October. The strategy is simple: buy an equity index on November 1, sell on May 1, park the money in bonds or cash for six months, repeat.

But Kaufman is careful to note that the edge has compressed over time as more participants have become aware of it. The pattern still shows up in long-run statistics, but in any given year it can fail spectacularly — some of the strongest market rallies (and crashes) have occurred during the "weak" May–October period.

Practical use: Don't sell everything in May. Instead, use seasonality as a tilt — slightly reducing risk during the historically weak period, slightly increasing it during the strong period. It's a background factor, not a primary signal.

The January effect

Small-cap stocks have historically outperformed large-caps in January. The commonly cited explanation: tax-loss selling in December drives small, beaten-down stocks to artificially low prices, which then bounce back in January as selling pressure evaporates and bargain hunters arrive.

Kaufman notes that this effect has been documented since at least the 1980s and was one of the first calendar anomalies to receive academic attention. However, the effect has weakened significantly in recent decades — partly because awareness led traders to front-run it (buying in late December instead of waiting for January), and partly because changes in tax law and market structure have altered the dynamics.

The January Barometer is a related concept (from Yale Hirsch's Stock Trader's Almanac): as January goes, so goes the year. If the S&P 500 is up in January, it tends to be up for the full year. Pring notes that the barometer has a decent but imperfect track record — it works better as a sentiment gauge than as a mechanical trading rule.

The presidential election cycle

Stock markets have shown a tendency to perform differently depending on which year of the four-year presidential cycle you're in. The historical pattern:

• Year 1 (post-inauguration): Weak. New presidents push unpopular policies early.
• Year 2 (midterm): Often the weakest. Markets price in political uncertainty.
• Year 3 (pre-election): Strongest. The incumbent stimulates the economy to win re-election.
• Year 4 (election year): Moderate to strong, especially in the second half once uncertainty resolves.

This pattern has held loosely over many decades. Pring relates it to the broader cycle of government policy tightening and loosening. But as with all calendar effects, individual data points deviate wildly — 2008 (election year) was catastrophic; 2017 (Year 1) was a powerful rally.

Month-end and turn-of-month effects

Kaufman documents a subtler but more consistent pattern: stocks tend to perform better in the last few days of the month and the first few days of the next month. The explanation centers on institutional fund flows — pension funds, payroll investments, and monthly rebalancing create recurring buying pressure at month-end/start.

This effect is smaller in magnitude than the others but more frequent and more consistent. Some systematic traders incorporate a slight long bias around month-turn dates.

Seasonal patterns in commodities

Pring emphasizes that seasonality is most reliable in commodity markets, where it has a fundamental basis:

• Energy: Heating oil tends to rise into winter and decline in spring. Gasoline tends to rise into summer driving season.
• Grains: Planting and harvest seasons create predictable supply-driven cycles.
• Gold: Often shows strength in September–February (jewelry demand, Indian wedding season).

These commodity seasonal patterns have physical underpinnings — weather, crop cycles, consumption patterns — which makes them more durable than pure market-sentiment anomalies.

The Santa Claus rally

Pring discusses the tendency for stocks to rally in the last five trading days of December and the first two of January. This is a short window — only about seven trading days — but historically shows a positive bias.

When the Santa Claus rally fails (the market declines during this window), it's sometimes treated as a warning sign for the coming year. The absence of the rally matters more than its presence.

How to use seasonality correctly

1. As a filter, not a signal. Seasonality should confirm a trade, not initiate it. A bullish breakout in November (start of the strong period) is more convincing than one in June.

2. As a risk-management tilt. Slightly tighter stops during the weak season, slightly wider during the strong season. Slightly smaller positions May–October.

3. In combination with other indicators. A seasonal tailwind plus a bullish crossover plus expanding breadth is a powerful three-factor setup. Seasonality alone is not enough.

4. With awareness that any single year can deviate. The 2020 crash bottomed in March (weak season) and the recovery powered through the entire "Sell in May" period. Slavishly following the calendar would have left you on the sidelines for one of the strongest recoveries in history.

Quick check

What you now know

• Sell in May: November–April historically outperforms May–October, but the edge has compressed.
• January effect: small-caps outperform in January, driven by tax-loss selling recovery — weakened by front-running.
• Presidential cycle: Year 3 (pre-election) has historically been strongest; Year 2 (midterm) weakest.
• Month-end effect: institutional flows create a recurring positive bias around the turn of each month.
• Commodity seasonality is the most reliable type — driven by physical supply/demand cycles.
• Use seasonality as a filter and tilt, never as a standalone signal.

Next: Market Breadth — the advance-decline line, McClellan oscillator, and new highs vs. new lows — the tools that tell you whether the market's "troops" are keeping up with its "generals."