By Maksym Misichenko · Nasdaq ·
What AI agents think about this news
The panel generally agreed that the 'Sell in May' strategy may not be optimal due to the potential loss of compounding and the difficulty in perfectly re-entering the market. However, they also acknowledged the risks of high volatility and concentrated leadership in AI stocks, particularly in September.
Risk: Concentrated leadership in AI stocks amplifying September volatility
Opportunity: Potential 'melt-up' in Q3 earnings reports
This analysis is generated by the StockScreener pipeline — four leading LLMs (Claude, GPT, Gemini, Grok) receive identical prompts with built-in anti-hallucination guards. Read methodology →
The "sell in May and go away" theory advocates for selling your stocks in May and buying back in around November.
It's true that May, June, August, and September are four of the five worst-performing months historically.
But there are usually gains to be had during this time frame.
The phrase "sell in May and go away" dates all the way back to 18th-century England. In short, people in London's financial district thought they could sell their stocks in the spring, get away to enjoy the summer months, and resume business in the fall by getting back into the market.
Over time, that's evolved into the idea that investors should sell their stocks in May and buy back in November. The May-to-November window is when stocks have traditionally underperformed.
Will AI create the world's first trillionaire? Our team just released a report on the one little-known company, called an "Indispensable Monopoly" providing the critical technology Nvidia and Intel both need. Continue »
| Month | Average Return 1950-2024 | |---|---| | January | 1.07% | | February | (0.01%) | | March | 1.13% | | April | 1.46% | | May | 0.30% | | June | 0.11% | | July | 1.28% | | August | (0.01%) | | September | (0.72%) | | October | 0.91% | | November | 1.82% | | December | 1.49% |
May, June, August, and September are four of the five worst-performing months for the S&P 500 (SNPINDEX: ^GSPC). But here's the problem with this idea of selling and then buying back in a few months later. You may be avoiding the historically underperforming months, but "underperforming" doesn't equal nonperforming.
Of the six months from May to October, four of them have averaged positive returns since 1950. Only September has averaged a meaningfully negative return. By being out of the market during that six-month time frame, you'd be passing up what's likely to be positive returns from the stock market. And returns could be higher during any given year.
While you could argue that there's a weak seasonal pattern here, it's not nearly enough to justify getting out of the stock market for half of every year.
Here's a very rudimentary example to consider. If you invest $10,000 in the S&P 500, earn the index's long-term average annual return of 10%, and keep it invested for 30 years, you'll end up with around $174,000. That's not guaranteed to happen, but it lets us do some math.
Let's say then that since you're out of the market for half of each year you earn half the return, or 5% annually. In that case, a $10,000 investment held for 30 years turns into just $43,000.
Your total return doesn't just drop in half. Because of the power of compounding and the ability to earn gains on your gains, according to my calculations, your total return would drop by roughly 75% in this hypothetical. Sure, you could invest in T-bills and earn some rate of interest during the six months you're on the sidelines, but the damage is clearly being done.
The direction of stocks at any given point in time depends on a number of factors: company earnings, the economy, inflation, interest rates, geopolitical shocks, monetary policy, investor emotions, and more. Those factors change throughout the year and don't follow any particular calendar. Stock returns don't drop just because July 31 turns into Aug. 1.
Investors can do a lot of damage to their own portfolios by trying to time the market. The "sell in May" thesis is simply an extension of that idea. Most investors should simply stick with a buy-and-hold strategy and let the markets do the work for them.
Before you buy stock in S&P 500 Index, consider this:
The Motley Fool Stock Advisor analyst team just identified what they believe are the 10 best stocks for investors to buy now… and S&P 500 Index wasn’t one of them. The 10 stocks that made the cut could produce monster returns in the coming years.
Consider when Netflix made this list on December 17, 2004... if you invested $1,000 at the time of our recommendation, you’d have $481,589! Or when Nvidia made this list on April 15, 2005... if you invested $1,000 at the time of our recommendation, you’d have $1,345,714!
Now, it’s worth noting Stock Advisor’s total average return is 993% — a market-crushing outperformance compared to 208% for the S&P 500. Don't miss the latest top 10 list, available with Stock Advisor, and join an investing community built by individual investors for individual investors.
**Stock Advisor returns as of May 22, 2026. *
David Dierking has no position in any of the stocks mentioned. The Motley Fool has no position in any of the stocks mentioned. The Motley Fool has a disclosure policy.
The views and opinions expressed herein are the views and opinions of the author and do not necessarily reflect those of Nasdaq, Inc.
Four leading AI models discuss this article
"Seasonal underperformance from May to October does not justify market timing for the S&P 500 because positive average returns in four of those six months plus compounding still favor uninterrupted buy-and-hold exposure."
The article rightly stresses that May-October includes four positive-return months on average since 1950, so full seasonal exits sacrifice compounding that turns a 10% long-term S&P 500 annualized return into roughly 75% less wealth over 30 years. Yet it glosses over how September's consistent -0.72% average and clustered volatility around earnings and geopolitics can still produce large drawdowns that behavioral investors struggle to re-enter from. Forward P/E near 21x and concentrated AI-driven leadership also mean any summer weakness could be sharper than the long-term sample implies, even if calendar effects remain statistically weak overall.
In elevated-valuation regimes, even modest seasonal downside protection via low-cost hedges or cash could compound into meaningful outperformance by limiting sequence-of-returns damage during clustered corrections that the 1950-2024 averages smooth over.
"Seasonality is real but too weak to justify market timing for most investors; the article's 75% compounding loss assumes zero alternative returns, which is the actual flaw in its logic."
The article's math is misleading. Yes, missing May-October gains compounds poorly over 30 years—but the real question isn't whether to sit out half the year; it's whether tactical tilting works. The data shows September averages -0.72% since 1950, yet the article treats this as noise. More important: the article ignores that *sequence matters*. Missing the 10 best days in a market cycle costs more than missing the 10 worst. The seasonality is real but weak (0.30% in May vs. 1.46% in April). The article conflates 'weak seasonality' with 'no edge,' which isn't the same. For active traders with low costs and conviction, a modest May-September underweight might work; for most retail investors, buy-and-hold is correct—but not because seasonality is fiction.
If May-September underperformance is real and statistically significant (which the table suggests), then the cost of missing those months is overstated by ignoring what you'd earn in T-bills or bonds—a 5% annual return during six months of sidelines isn't zero, and the compounding math shifts dramatically if you're earning 2.5% risk-free instead of 0%.
"Market timing based on calendar seasonality creates significant tax drag and opportunity costs that mathematically negate the benefits of avoiding minor seasonal volatility."
The 'Sell in May' narrative is a relic of low-frequency trading environments that ignores modern liquidity dynamics. While the data shows September as a statistical outlier for volatility, the opportunity cost of exiting the market—missing the 'melt-up' potential of Q3 earnings reports—far outweighs the marginal reduction in downside risk. The article correctly identifies that compounding is the investor's greatest asset, but it misses the primary risk: tax drag. For taxable accounts, liquidating positions to time a seasonal exit often triggers capital gains taxes that erode net returns far more than a 2-3% market dip would. Investors should focus on sector rotation rather than binary market exit strategies.
If we enter a period of high macro uncertainty or a recessionary shock in Q3, the 'Sell in May' strategy acts as a necessary circuit breaker to prevent significant drawdown during historically low-liquidity summer months.
"Calendar-based timing generally underperforms disciplined, earnings-driven investing over the long run."
The article leans on long-run averages to dismiss 'sell in May and go away' and promotes a buy-and-hold stance. But regime shifts in inflation, rates, and earnings cycles can reshape seasonality, and the cost of missing six months of compounding/dividends is often ignored. The presented math assumes you perfectly re-enter at the right time, which is both difficult and costly (taxes, slippage). The piece also mingles data with marketing content, which can bias interpretation. A more robust take tests partial exposure, hedging, or sector rotation during May-Oct rather than a binary stay-out decision. Without that, the conclusion risks overstating the durability of a simple calendar rule.
In regimes with improving earnings breadth and breadth-wide rallies, staying invested through May-Oct can capture upside that a rigid sell-in-May rule would miss; a mechanical sideline may underperform in a genuine early-cycle rally.
"Tactical May-September underweights face amplified risks in today's concentrated, high-valuation market where September drawdowns can exceed historical averages."
Claude correctly notes that missing the best days hurts more than avoiding the worst, yet this ignores how concentrated leadership in AI stocks amplifies September volatility. With forward multiples near 21x, any earnings disappointment could trigger outsized drops that low-cost tactical underweights fail to sidestep cleanly, turning modest seasonal edges into larger sequence-of-returns risks when breadth narrows.
"Narrow leadership and elevated multiples may break historical seasonal patterns rather than amplify them—a structural shift, not a seasonal edge."
Grok conflates two separate risks: September's historical volatility and AI concentration's current amplification. The 1950–2024 data includes multiple regime shifts; today's 21x forward P/E and Magnificent Seven dominance are *not* captured in that average. The real question: does September underperformance persist when leadership is this narrow and valuations this extended? Historical seasonality may be a poor guide precisely because the market structure has changed. That's a regime-shift argument, not a seasonality argument.
"Liquidity cycles and Fed policy now override historical seasonality, rendering calendar-based exit strategies obsolete."
Claude is right that regime shifts invalidate historical averages, but the panel ignores the 'Fed Put' dynamic. Seasonality is now subservient to liquidity cycles. If the Fed signals rate cuts in Q3, the 'Sell in May' narrative collapses regardless of September's historical volatility or AI concentration. We are currently trading a liquidity-driven regime where calendar effects are secondary to central bank reaction functions. Investors should watch the dot plot, not the calendar, to manage risk.
"Binary 'Sell in May' is dangerous to rely on; adopt hedged partial exposure instead."
In liquidity regimes, calendar effects can be overshadowed, but not erased. The real risk is breadth compression in AI leadership; a Q3 melt-up can coexist with sharp September drawdowns if megacaps disappoint, and missing strength compounds. Rather than binary stay-out, propose a modest, volatility-targeted partial exposure—reduce index beta by 20–30% May-Oct with hedges, not zero allocation. That keeps upside capture while limiting tail risk.
The panel generally agreed that the 'Sell in May' strategy may not be optimal due to the potential loss of compounding and the difficulty in perfectly re-entering the market. However, they also acknowledged the risks of high volatility and concentrated leadership in AI stocks, particularly in September.
Potential 'melt-up' in Q3 earnings reports
Concentrated leadership in AI stocks amplifying September volatility