The October Pattern: Larry Williams' Most Profitable Seasonal Trade
Larry Williams didn't build his reputation on complex math. He built it on patterns that repeated, decade after decade, because they were rooted in market structure — not in curve-fitted indicators.
The October Pattern is one of the simplest and most statistically robust seasonal trades in existence. It requires no technical indicators, no fundamental analysis, and no discretion. You enter on a specific calendar date. You exit on a specific calendar date. You use a hard stop. That's it.
And it has beaten the S&P 500 — with far less drawdown — for over 60 years.
What Is the October Pattern?
The pattern exploits a seasonal tendency in equity markets: a statistically significant bullish move that begins in late October and extends through the end of the year, into early January.
Larry Williams documented this pattern in multiple books and seminars. The core thesis is simple:
| Component | Rule |
|---|---|
| Market | S&P 500 (or broad US equity index) |
| Entry | On or shortly after October 28 |
| Exit | First few trading days of January |
| Stop Loss | Typically placed below the October low |
| Position | Long only |
The logic isn't mystical. October is when tax-loss selling peaks, mutual funds rebalance, and institutional money managers position for year-end performance chasing. The selling pressure exhausts itself, and a rally follows.
The Historical Edge
Larry Williams and other seasonal researchers have tested this pattern across decades of data. The results are consistent.
| Period | Win Rate | Avg. Return | Max Drawdown |
|---|---|---|---|
| 1950–2023 | ~72% | +3.8% (avg) | ~6% |
| 2000–2023 | ~68% | +3.2% (avg) | ~8% |
| Worst years | 1987 (crash), 2008 (crisis) | -5% to -8% | Stopped out |
Key insight: Even in losing years, the drawdown is typically smaller than a buy-and-hold position through the same period. The stop loss limits damage. In winning years, the pattern captures the bulk of the year-end rally.
The October 1987 crash is the most famous exception. The pattern triggered a loss that year. But the stop loss — if applied — would have contained it. No pattern wins every year. The edge comes from consistency over decades.
The Exact Rules (As Taught by Larry Williams)
Different versions of this pattern exist. Here is the most commonly taught Williams formulation:
Entry Rules
- Date trigger: Enter long on the close of the last trading day of October, or the first trading day of November.
- Alternative trigger: Some variants enter on the close of October 28 or 29, if the market has already sold off into month-end.
- Confirmation (optional): Larry sometimes adds a simple filter — enter only if the market is above its 20-day moving average at entry. This avoids catching falling knives in deep correction years.
Exit Rules
- Date exit: Exit on the close of the first trading day of January, or the first few days of the new year.
- Alternative exit: Some traders exit the first week of January to capture any "January effect" continuation.
Risk Management
| Parameter | Setting |
|---|---|
| Stop loss | Below the lowest low of October, or a fixed percentage (e.g., 4-5%) |
| Position size | Fixed fractional (e.g., 2% risk per trade) |
| Leverage | None. This is a cash or margin-long pattern. No options, no futures unless you're sizing down. |
Why the Pattern Works
Larry Williams always emphasized that a pattern's profitability is meaningless if you don't understand why it works. The October Pattern has structural, behavioral, and institutional drivers:
| Driver | Explanation |
|---|---|
| Tax-loss selling | Investors sell losers in October for tax purposes. This creates forced selling that reverses in November. |
| Mutual fund rebalancing | Funds window-dress portfolios before year-end. They buy winners, adding upward pressure. |
| Performance chasing | Underperforming managers take risk in Q4 to catch benchmarks. This increases equity exposure. |
| Holiday liquidity | Lower volume in late December can amplify moves in either direction. The pattern bets on the upside. |
| January inflows | Pension contributions, 401(k) matches, and new-year capital deployment create buying pressure. |
Modern Adaptations
The October Pattern was originally tested on the Dow Jones and S&P 500. Today, traders have more vehicles to express it:
| Vehicle | Pros | Cons |
|---|---|---|
| SPY / VOO | Simple, liquid, no expiration | No leverage, full capital at risk |
| ES futures (CME) | Leverage, tax efficiency, 23.5h trading | Margin requirements, rollover costs |
| Micro E-mini (MES) | Same as ES with smaller size | Same as ES, still futures complexity |
| LEAPS calls | Defined risk, leverage | Time decay, wide spreads on far OTM |
| 3x leveraged ETFs (UPRO, TQQQ) | Amplified returns in winning years | Amplified losses, volatility decay |
Larry Williams' view: Keep it simple. The pattern works because of behavioral regularity, not because of the vehicle. SPY with a hard stop is better than a leveraged instrument that bleeds on volatility.
The Unger Academy Perspective
Andrea Unger — four-time Robbins World Cup Champion — takes a different but complementary approach. Unger's methodology is purely systematic and automated. He would test the October Pattern like this:
- Code it in TradeStation, MultiCharts, or TradingView Pine Script. No discretion.
- Run a walk-forward analysis. Test the pattern on out-of-sample data to confirm it isn't curve-fitted.
- Run Monte Carlo simulation. Randomize trade sequences to see if the edge holds under distribution stress.
- Portfolio context. The October Pattern is one system among many. Never trade a single pattern with full capital.
Unger's core lesson: A good pattern is necessary but not sufficient. Position sizing, portfolio heat, and correlation management determine whether you survive the losing years.
TradingView Pine Script Implementation
Here's a simple script you can paste into TradingView to backtest the October Pattern:
//@version=5
strategy("October Pattern - Larry Williams", overlay=true, default_qty_type=strategy.percent_of_equity, default_qty_value=100)
// Parameters
entryMonth = input.int(10, "Entry Month (October)")
entryDay = input.int(28, "Entry Day")
exitMonth = input.int(1, "Exit Month (January)")
exitDay = input.int(3, "Exit Day")
stopPct = input.float(4.0, "Stop Loss %")
// Entry condition: enter on or after October entry day if not already in position
enter = month == entryMonth and dayofmonth >= entryDay and strategy.position_size == 0
// Exit condition: exit on or after January exit day if in position
exitTrade = month == exitMonth and dayofmonth >= exitDay and strategy.position_size > 0
// Stop loss
stopPrice = strategy.position_avg_price * (1 - stopPct / 100)
if enter
strategy.entry("Long", strategy.long)
if exitTrade
strategy.close("Long")
if strategy.position_size > 0
strategy.exit("Stop", "Long", stop=stopPrice)
// Plot
plotshape(enter, style=shape.triangleup, location=location.belowbar, color=color.green, size=size.small)
plotshape(exitTrade, style=shape.triangledown, location=location.abovebar, color=color.red, size=size.small)
Run this on SPX, SPY, or ES1! and compare the equity curve to buy-and-hold over the same period.
Key Takeaways
| Principle | Application |
|---|---|
| Seasonality is real, but not magic | The October Pattern has a behavioral basis. It works until the underlying structure changes. |
| One pattern is not a system | Trade this alongside uncorrelated strategies. Never bet the farm on a single calendar trade. |
| The stop is non-negotiable | 1987 and 2008 proved that no pattern is invincible. The stop is what keeps you in the game. |
| Simplicity beats complexity | No indicators, no neural nets, no AI predictions. A calendar and a stop. That's the edge. |
| Test everything yourself | Don't trust backtests from 1987. Run your own walk-forward analysis on current data. |
Related Resources
- Larry Williams' Official Site — Courses, books, and the original research.
- Unger Academy — Systematic trading education from a four-time world champion.
- TradingView — Charting and backtesting platform.
- Interactive Brokers — Professional execution for systematic traders.
Disclaimer: This article is for educational purposes only. Past performance of any trading pattern does not guarantee future results. Always do your own research, backtest thoroughly, and never risk capital you cannot afford to lose.
