Most traders use technical analysis to find an edge. Some use fundamentals. Very few use seasonality. Jay Kaeppel has spent decades building trading systems from seasonal patterns that have produced results most traders would not believe possible, all from a handful of calendar-based rules applied consistently over time.
Jay Kaeppel has over 30 years of trading experience. He worked as a software developer for futures and options trading, then spent nine years as a head trader for a CTA, where the firm made money in eight of nine years. He spent nine more years as an instructor at OptionsEdX and published four books including Seasonal Stock Market Trends. In episode 34 of the Better System Trader podcast, Jay shares the seasonal systems he still trades, why so few traders use seasonality despite the evidence, and how to build a diversified portfolio of strategies that reduces the damage any single drawdown can do.
Watch the full episode below, then read on for the complete breakdown.
Why seasonality is an underused edge
Jay opens his talks with a simple experiment. He asks the audience how many have used technical analysis to find a trade. Everyone raises their hand. Fundamental analysis? Most raise their hand. Seasonal trends? Almost no one.
His point is direct: if you are looking for an edge, it makes more sense to look where other traders are not looking. Seasonality sits in that space. The patterns have been documented for decades, yet very few systematic traders build them into their models. That underuse is part of what keeps the edge intact.
“The primary job of a trader is to find something that works and that other people aren’t using. Seasonal trends fit that description.”
The Santa Claus rally: 84% win rate over 66 years
Jay’s definition of the Santa Claus rally is more precise than the popular version. He defines it as starting at the close on the Friday before Thanksgiving and running through the close of the third trading day of January. That is roughly six weeks, and over 66 years it has produced a gain 55 times, giving it an 84% win rate.
He is careful to frame what this means in practice. A seasonal trend is a bias, not a guarantee. His suggestion is to use it as a filter: if the seasonal trend says the market should be rising and your trend-following signals agree, that is a high-quality setup. Seasonality improves the odds, but the trade still needs confirmation from price.
Holiday trading: $1,000 to $10,000 in only 50 days per year
One of the most compelling statistics in the episode comes from Jay’s holiday trading research. He found that the three trading days before and the three trading days after every market holiday have shown a consistent positive bias going back to 1955.
Starting with $1,000 in 1955 and investing in the Dow Industrial Average only during those six days around each holiday, about 50 days per year, that position grew to over $10,000. The return is not spectacular in raw dollar terms, but the rate of return per day in the market is notable. Most of the year you are sitting out, and the returns come from those concentrated windows around holiday periods.
Jay admits he does not have a complete explanation for why it works. His working theory is some general positive sentiment around holiday periods, but his practical conclusion is simple: it has worked consistently for 70 years, and that is enough.
The monthly seasonal system: beating buy-and-hold over 60 years
The core of Jay’s quantitative approach is a monthly seasonal system built around specific trading days. The rule is straightforward. Be long the stock market during the last four trading days of the month, the first three trading days of the following month, and trading days nine through twelve. Add in the three days before and after each holiday. Everything else, stay out.
The long-term result: starting with $1,000 at the end of 1955, buy-and-hold the Dow grew to a 3,400% return over 60 years. The seasonal system over the same period returned 38,000%. That is not a typo. The same time frame, the same market, just a filter for when to be in it.
There is a structural reason the mid-month window works that did not exist before the 1980s. Trading days nine through twelve showed no edge from 1955 through 1980. After that, the edge appeared consistently. Jay’s explanation: IRAs were introduced in the early 1980s. From that point on, payroll money automatically flowing into retirement accounts created built-in demand for stocks every month at the end, beginning, and middle of the month. The pattern has a mechanical cause, which makes it more durable than a statistical coincidence.
RYTNX: turning seasonal signals into 1,600% returns
Knowing when to be in the market is step one. The second step is figuring out how to exploit the opportunity. Jay calls this the full trading process condensed to four words: spot opportunity, exploit opportunity.
For exploiting the monthly seasonal signals, Jay points to RYTNX, a Rydex fund that tracks the S&P 500 at 2x leverage on a daily basis. He is not a general advocate of leveraged funds. Over long periods they tend to decay relative to their benchmark. But for short-term seasonal trades where you are never in the market for more than seven to ten days at a time, the leverage amplifies the edge without the compounding problems that hurt long-term holders.
The numbers from inception in June 2000 through 2015: the Dow gained 72% over that period. Trading the seasonal system with RYTNX returned over 1,600%. The comparison covers the dot-com crash, the 2008 financial crisis, and the recovery that followed. The system held up through all of it.
The seasonal sector rotation model
Beyond the monthly system, Jay outlined a sector rotation approach based on calendar patterns. Different sectors show different seasonal characteristics in different months. His model assigns specific Fidelity sector funds to specific months: S&P 500 exposure in December and January, retail stocks in February, March, October, and November, energy in April, intermediate-term Treasury bonds from May through July, and gold in August and September.
The model is rebalanced monthly with no discretionary input. Starting with $1,000 at the end of 1989 through September 2015, the result was over $200,000. Twenty-six years of consistency. He followed it personally from around 2005, which included the 2008 drawdown where the model lost roughly 50%. Since then, it recovered and went on to gain over 600%.
His point about the 2008 drawdown illustrates a broader principle: this is why diversification across multiple systems matters. A 50% loss on one system is a portfolio crisis when it is your only system. When it is 10% of your overall allocation, you can trade through it.
The Known Trends Index
In his book, Jay describes about 13 different seasonal trends. Some are his own research, others come from Yale Hirsch, Norman Fosbeck, and other seasonal trading pioneers. The Known Trends Index is his attempt to combine them into a single composite reading.
The idea is to look at how many of the 13 seasonal factors are currently pointing bullish versus bearish. A high reading means multiple seasonal factors are aligned favorably. A low reading means the opposite. Rather than acting on any single seasonal signal, the index gives a broader sense of the seasonal backdrop, which he uses as one input in his overall market bias assessment.
September is the worst month, not October
A common belief among traders is that October is the most dangerous month of the year, based on events like the 1987 crash. Jay pushes back on this. September is the only month that has shown a net loss in the Dow over the last 70 years. If you had only ever been long the stock market in September and out every other month, you would have lost money. October gets the reputation because that is when some dramatic crashes have occurred, but the consistent underperformer on the calendar is September.
Related episodes
- Cornering the Market: seasonality, risk, and counterparty protection – Kurt Sakaeda
- Algorithmic Forecasting – Larry Williams (Episode 40)
- How to Get Unlimited Trading Ideas – Rob Hanna (Episode 7)
- How to Leverage Market Biases – Lawrence Chan (Episode 204)
Get the show notes & transcript
Want to find and test your own seasonal edges? Subscribe to the Better System Trader podcast for weekly interviews with the world’s top systematic traders and quantitative researchers.
