How to Time the Stock Market and When You Shouldn’t Bother
Learn why the proof that time in the market beats timing the market is wrong, why most people underperform the market, and when to buy & hold.
There is an oft-repeated investment adage: “Time in the market beats timing in the market”. The idea is that investors are terrible at timing the market and, therefore, shouldn’t try. This underperformance is demonstrated by studies (source, source, source) that show that most investors not only underperform the overall market but underperform their chosen investments (often by 3%–4%)—in effect, they tend to buy high and sell low. Thus, the conclusion seems to be that buy & hold is the only sensible investment strategy.
The Greed/Fear Cycle
The likely cause of this chronic underperformance is investors use intuition (i.e. emotions) to dictate their investment strategy (even if they’re unaware of this propensity). The following pattern will look familiar to anyone who’s gotten emotionally invested in their investments or business venture:
When you’re making lots of money, it’s easy to get excited and feel invincible—you start imagining everything you’ll buy with your newfound wealth. Then, when your fortunes turn, all your confidence disappears, the negative emotions set in, and you stop making rational decisions. Magnifying this rollercoaster is that financial markets tend to fall much faster than they rise, and our fear of loss is more potent than our fear of missing out (i.e. greed), which makes us more likely to overreact to downturns. In general, our emotions lead us to do the opposite of what we should be doing.
“Be fearful when others are greedy and be greedy when others are fearful.” —Warren Buffet
Justifying Buy & Hold
Managing the emotional side of investment decision-making can be challenging. So, advising people to buy & hold makes sense because it’s clearly better than buying high & selling low.
My problem is with those who claim that the Best Days Theory shows (“proves”) that timing the market is not possible, or at least highly improbable (example, example, example, example). This popular theory can be summarized as follows:
- If you miss the top 20 (or whatever) best days in the market, you’ll generate much lower returns.
- The best days are often during market downturns.
- Therefore, you must endure market downturns to capture these best days.
- Thus, buy & hold is the best strategy because otherwise, you’ll be tempted to sell at the wrong time and miss the best days.
The apparent takeaway of this theory is that capturing the best market days is the key to investment success and that you cannot have good returns if you miss the best days. As I’ll demonstrate below, while the first two claims are verifiably true, the conclusions are nonsense. The obvious flaw is that they focus exclusively on the impact of the best market days. How cosmically unlucky do you need to be to miss only the best days? What about the worst days? Can you miss some of the best days if you also miss some of the worst days?
Effect of Skipping Specific Days
To quantify the specifics of this theory, I’ll use the S&P 500 Index ETF (symbol SPY) since its inception (30 years) as our target investment. In case you’re curious, the exact time range and index are unimportant—the effect is pervasive across time and any broad index. I will measure how much our annualized return changes, assuming we can selectively skip the next day when we know it will be one of the best (or worst) days.
Annualized Returns of S&P 500 Index ETF, Feb 1993–Dec 2022
| Skip Best Days Only |
Skip Worst Days Only |
Skip Best & Worst Days |
|
|---|---|---|---|
| Skip 0 days (baseline) | 9.5% | 9.5% | 9.5% |
| Skip 10 days | 6.9% | 12.5% | 9.8% |
| Skip 20 days | 4.9% | 14.8% | 9.9% |
| Skip 30 days | 3.3% | 16.6% | 10.0% |
| Skip 40 days | 1.9% | 18.4% | 10.1% |
| Skip 50 days | 0.5% | 20.1% | 10.2% |
As predicted, missing the best days substantially shrinks our returns. As a side note, remember that earning 2% more per year over 30–35 years will double the value of your investment. So, as extraordinary as it sounds, missing just ten specific days over 30 years does the opposite—it cuts your investment in half, and missing an additional ten best days will cut it in half again.
What they neglected to tell us is the best and worst days largely cancel out—and if anything, they skew slightly positive, showing that the magnitude of the worst days is larger, on average, than the best days.
So, will missing the best days guarantee poor returns? Clearly not.
Bull and Bear Markets
Since we know it’s OK to miss the best days if we also miss the worst days, let’s analyze the second claim—that we must endure market downturns to not miss the best days.
To classify the market’s direction (i.e. bull versus bear markets), I’ll calculate the price’s 200-day simple moving average (SMA200). This value is just the average closing (last) daily price over the previous 200 days. The SMA200 is among the oldest and most popular technical financial indicators and is often quoted in the financial media. It’s useful because it lets us compare today’s price versus the average price over the past ~10 months, giving us a clue as to the market’s recent performance—above the SMA200 means the price has been trending up recently (on average), and below means it’s been trending down. Why 200 days? There’s no good reason other than it’s a nice round number and is what everyone else is looking at (creating a self-fulfilling prophecy if other investors use it to make decisions). However, 190 days and 210 days also yield similar results.
To illustrate the effect of using this indicator, let’s analyze investing in the S&P 500 Index ETF (symbol SPY), buying or selling at the end of the month based on whether the closing price is above or below our reference price (SMA200).
S&P 500 Investment Results, Dec 1993–Dec 2022
| Always Invested | Bull Market — Above SMA(200) |
Bear Market — Below SMA(200) |
|
|---|---|---|---|
| % of months | 100% | 76% | 24% |
| Total return | 1,307% | 1,478% | –11% |
| Average month return | 0.86% | 1.10% | 0.07% |
| Best month return | 12.7% | 10.9% | 12.7% |
| Worst month return | –16.5% | –14.1% | –16.5% |
| % Positive months | 65% | 69% | 51% |
| Best day return | 14.5% | 5.8% | 14.5% |
| Worst day return | –10.9% | –7.2% | –10.9% |
| % of 50 best days | 100% | 18% | 82% |
| % of 50 worst days | 100% | 30% | 70% |
| Average daily volatility | 1.2% | 0.9% | 1.8% |
| Average monthly volatility | 4.4% | 3.5% | 6.4% |
The most interesting things I noticed here are:
- Bear markets are much more “exciting”—they have twice the volatility and contain most of the best and worst days.
- Bear markets represent a quarter of the total time but contribute almost nothing to the total returns, with disproportionately more anxiety.
- Bear markets have more best days than worst days, yet the returns are terrible.
So why must I stay invested through market downturns to have good returns?
In Practice
Using this simple market filter (SMA200), we can build a basic trading system that I’ll test across several market indexes.
The rules are, at the end of each month:
- Sell if the monthly closing price <= SMA200
- Hold (or reenter) if the monthly closing price > SMA200
I’ll compare this timing strategy to the recommended buy & hold strategy across the available history for several common assets.
S&P 500 Index (symbol SPY), Jan 1994–Dec 2022
| Timing Strategy | Buy & Hold | |
|---|---|---|
| Annualized return | 9.9% | 9.5% |
| Maximum drawdown | –23% | –55% |
| Maximum drawdown length in months (peak to new peak) |
41 (2000–2003) |
79 (2000–2006) |
| Average trades per year | 1.2 | 0 |
| % Time in the market | 76% | 100% |
NASDAQ 100 Index (symbol QQQ), Jan 2000–Dec 2022
| Timing Strategy | Buy & Hold | |
|---|---|---|
| Annualized return | 7.8% | 5.8% |
| Maximum drawdown | –44% | –83% |
| Maximum drawdown length in months |
120 (2000–2010) |
179 (2000–2015) |
| Average trades per year | 1.7 | 0 |
| % Time in the market | 71% | 100% |
TSX Composite Index (symbol XIC), Oct 2001–Dec 2022
| Timing Strategy | Buy & Hold | |
|---|---|---|
| Annualized return | 6.7% | 7.1% |
| Maximum drawdown | –28% | –48% |
| Maximum drawdown length in months |
38 (2007–2010) |
32 (2008–2011) |
| Average trades per year | 1.5 | 0 |
| % Time in the market | 77% | 100% |
US Bond Index (symbol AGG), Aug 2004–Dec 2022
| Timing Strategy | Buy & Hold | |
|---|---|---|
| Annualized return | 3.0% | 2.9% |
| Maximum drawdown | –10% | –18% |
| Maximum drawdown length in months |
32 (2016–2019) |
29 (2020–2022) |
| Average trades per year | 1.7 | 0 |
| % Time in the market | 77% | 100% |
The results look very impressive. Not only have the drawdowns been cut in half, but the returns are the same or slightly better, even while only being invested three-quarters of the time. This system appears to be an easy way to significantly reduce the downside risk of investing without impacting your potential upside. But before you get too excited, there are a couple of important caveats.
The first is that trading at the end of the month outperforms most of the other days of the month (sometimes as much as 1%–2%). Therefore, it’s possible that we just got lucky with our timing by avoiding a significant drop at the beginning or end of a month by a couple of days, making these returns not robustly predictive of the future. But digging into the numbers, it appears this end-of-month advantage persists across assets and time, and no small set of “lucky” timing events exist to explain this end-of-month performance advantage. Therefore, rather than luck, it’s more likely this benefit is due to repeatable seasonal or calendar behavioural effects or institutions’ quarter-end (or year-end) window dressing. I’ve looked for evidence to disprove this hypothesis but have not found any.
The second caveat is that this style of timing system underperforms during bull markets while dramatically outperforming during bear markets. The problem is there are times during a bull market when we’ll exit early and miss some of the upside because it falsely predicts a bear market. This early exit risk is an inherent limitation of using moving averages for timing—no amount of fiddling with their length will overcome it. As such, my results are sensitive to the date range used and the relative size and lengths of the bull and bear markets during the analysis. For example, our stellar returns for the NASDAQ and TSX were undoubtedly helped by their trading history starting at a market top (year 2000).
Given the relatively short history of these assets (only 2 to 3 complete market cycles with significant bull/bear periods), it’s hard to draw any definitive conclusions from this analysis about whether the bear market overperformance is enough to compensate for the bull market underperformance. Fortunately, Meb Faber, in his paper A Quantitative Approach to Tactical Asset Allocation, demonstrated that this type of timing system has worked very well across multiple assets (US stocks, foreign stocks, bonds, commodities, real estate) over the past 100 years.
The takeaway is that, in general, a simple moving-average timing system (like SMA200) will regularly mildly underperform in the medium term but should keep pace with buy & hold over the long term while dramatically reducing drawdowns.
Conclusion
Would you call using the results of the SMA200 trading system successfully timing the market? I would. It’s challenging to outperform buy & hold by looking only at total returns (rather than risk-adjusted returns), but this comes very close. The interesting point is that we’re doing the opposite of what they said we must do to achieve good returns—we’re skipping most of the best days and bear markets. So, why would people champion the can’t-miss-the-X-best-days theory? I don’t rightly know.
Is this style of timing system worth using? In general, no. For long-term investing, reducing volatility (as we’re doing here) reduces returns when making regular ongoing contributions. Instead, it’s better to lean into the natural market cycles and load up when good assets are on sale. Thus, using dollar-cost averaging and buy & hold is usually the best strategy for savers. But market timing can be beneficial when you’re more sensitive to drawdowns (e.g. during retirement) or have contribution timing risk (e.g. a significant lump-sum contribution). Smaller drawdowns give us a higher safe withdrawal rate during retirement because we don’t need the same amount of buffer to guard against market downturns.
There are also tax implications for this type of strategy when investing outside a tax-sheltered account (e.g. capital gains), so it’s not a magic bullet. Still, this demonstration should be enough to debunk the theory that you can’t time the market successfully.
Key Points
- The most significant risk to your investments is you and your emotional decisions.
- Just because most investors don’t time the market successfully doesn’t mean you can’t time the market—but it must be a mechanical (preplanned, mathematical, backtested) strategy rather than a knee-jerk emotional reaction to current events.
- Simple market timing (like using SMA200) has a cost, but it’s often small (besides taxes), so it can be a helpful tool in your toolbox.
Points to Ponder
- Do you find yourself panicking or discarding your plan during market crashes? Could you be more systematic?
- Have you held large positions during recent significant market crashes (e.g. 2000, 2008, 2020)? What was your emotional reaction? In retrospect, were you tempted to sell at the wrong time?
- If you’re approaching retirement or in retirement, could this type of timing strategy help reduce the downside volatility of your portfolio?
Related Concepts
- Why Dollar-Cost Averaging Outperforms Lump-Sum Investing
- Why Aggressive Portfolios Are Usually Safer Than Conservative
- How to Estimate How Much Money You’ll Need for Retirement
If you have comments, questions, or constructive feedback, you can contact us by email at questions@essentialsofinvesting.com.