With the S&P 500 hitting its 36th all-time high of 2025 last week, many investors are wondering whether now is a good time to buy. After all, wouldn’t you get a better price if you waited? Technically, you very likely would. As I stated in Just Keep Buying:
If you randomly picked a trading day for the Dow Jones Industrial Average between 1930-2020, there is over a 95% chance that the Dow would close lower on some day in the future.
So while we are likely to see a lower price by waiting, unfortunately, these lower prices are rare (i.e. they only occur on 5% of days). More importantly, these 5% of days can’t be identified ahead of time.
But, what about after an all-time high? The analysis above was done across all trading days. If we subset to only all-time highs, would the probability of seeing a lower price in the future change?
To test this, I re-ran this analysis from January 1915 to October 2025 for the Dow Jones Industrial Average. Across all trading days, the chance that the Dow would close lower on some day in the future was 96.7% (even higher than during 1930-2020 period).
But, if you bought at an all-time high, the probability that you’d see a lower price in the future was 97.0%. It was slightly higher than when the market is not at an all-time high.
While this might seem like evidence that you shouldn’t buy at all-time highs, unfortunately, this difference was not statistically significant. The difference between 96.7% and 97.0% seems to be just random chance. In other words, when you buy at an all-time high you are no more likely to see a future lower price than buying on any other trading day.
So while buying U.S. stocks at an all-time high doesn’t seem to harm us, does it help us?
Historically, not really.
From 1950-2025, the future 1-year returns of the S&P 500 are actually lower when near an all-time high (“Near ATH”) than when off an all-time high (“Off ATH”) [8.3% vs. 10.1%]. In this case “Near ATH” means anytime the S&P 500 is within 5% of an all-time high, “Off ATH” is all other times:
As you can see, the 1-year return distributions are similar though the “Off ATH” distribution is shifted slightly to the right (i.e. has better returns).
But if we look over a 3-year return window, this changes. For 3-year future returns, it’s better to be near an all-time high than off of one [8.4% annualized vs. 7.8% annualized]:
Because the 1-year and 3-year returns point in opposite directions, this suggests that market timing based on all-time highs doesn’t work. In other words, being near an all-time high doesn’t seem to impact future U.S. stock returns in any consistent or meaningful way.
While there are long stretches where the S&P 500 doesn’t hit all-time highs (see below), these periods have been quite rare:
But what about other markets? Though the S&P 500 doesn’t seem to have any difference in returns when near all-time highs, what about non-U.S. equities? Let’s turn to those now.
What About International Stocks at All-Time Highs?
When it comes to international stocks, I looked at how the MSCI EAFE (“Developed stocks”) and MSCI Emerging Markets (“EM stocks”) indices performed going back to 1970 and 1989, respectively. Overall the EAFE (Developed stocks) tend to do slightly better when near an all-time high over the next year [9.0% vs 7.2% on average]:
However, these stocks tend to do slightly worse when near an all-time high over the next three years [5% annualized vs. 6.7% annualized]:
This suggests that any momentum for EAFE stocks tends to be short-lived.
However, this is not the case for Emerging Market stocks, which do about 2x better over both 1-year and 3-year periods when near all-time highs. Over the next year, EM stocks return 15.6% (on average) when near an all-time high versus only 7.9% when off of one:
And over the next three years, EM stocks return 10.5% (annualized) when near an all-time high versus only 5.1% (annualized) when off of one:
This suggests that momentum plays a bigger role with EM stocks than with U.S. or Developed stocks.
We can see this in the chart below which shows the MSCI Emerging Markets index over time with all-time highs plotted in red. Note how EM stocks seem to have bursts of high returns followed by extended periods of stagnation:
While I would love to definitively conclude that EM stocks are more likely to trend, we must keep in mind that this only covers a few market cycles. When you have only a few observations to go off of, it’s difficult to make generalized statements.
Nevertheless, there is no historical evidence that buying international stocks (Developed or EM) near an all-time high is followed by worse short-term performance. If anything, there is some evidence that short-term performance may be slightly improved (at least for EM stocks).
Now that we’ve looked at international stocks, let’s turn our attention to two other asset classes that have done quite well this year.
What About Bitcoin and Gold?
In 2025 both Bitcoin and gold hit all-time highs and have been the bedrock of the debasement trade. How do these asset classes tend to perform when near all-time highs? Incredibly well, especially over the next year.
Historically, gold returned 37.9% (on average) over the next year when near an all-time high and only 4.1% when off of one (from 1979-2025):
When gold goes on a run, it really runs over the next year.
You could say the same thing about Bitcoin from 2014-2025:
Bitcoin returned 145.9% (on average) over the next year when near an all-time high compared to only 78.4% when off an all-time high during this period.
Unfortunately, these fantastic returns don’t seem to persist over longer periods. When looking at the 3-year timeframes, Bitcoin performs relatively worse when near an all-time high than when off one [23.1% annualized vs. 61.6% annualized]. And gold does roughly the same in either condition [5.0% annualized vs. 4.2% annualized].
Gold in particular has been very susceptible to momentum given how long it tends to stay between all-time highs:
What does this mean for you as an individual investor?
If you do own gold or Bitcoin, expect these assets to trend for a while and then have extended periods where they aren’t making new highs. However, when gold and Bitcoin do make new highs, this tends to be bullish in the short term.
Of course, all-time highs are bullish until they aren’t. Have gold and Bitcoin’s recent runs come to an end? I don’t know. Unfortunately, history is the study of everything that’s happened so far. It’s very possible that this time is different.
Is This Time Different?
Though the future is always uncertain, the data presented here should give you some relief when buying assets near all-time highs. Whether you are investing in U.S. stocks, international stocks, gold, or Bitcoin, the evidence suggests that all-time highs should not alter your regular investment plan. If anything, they’re neutral to slightly bullish in the short term.
But, is this time different?
Trust me, I get it. A few months ago I explained why I was bearish on U.S. stocks due to some extreme valuations I was seeing in the tech space. I still believe this thesis to be true. I believe AI is overhyped and some AI maximalists are starting to believe so as well.
Nevertheless, I haven’t stopped buying U.S. stocks. What did I do instead? I rebalanced my 401(k) from 100% risk assets to 80% risk assets + 20% intermediate-term U.S. bonds. It decreased my overall allocation to U.S. stocks by about 1.5%.
That’s it. I truly do sin a little. Just enough to satisfy my behavioral urge without risking my financial future.
Of course, I don’t know how the AI race is going to end. I don’t know if this time is different. But avoiding U.S. stocks entirely because they are at all-time highs would’ve been a bad strategy to follow.
If you’re truly worried, do what I do and sin a little. Then hold your nose and Just Keep Buying.
Happy investing and thank you for reading!
If you liked this post, consider signing up for my newsletter.
This is post 475. Any code I have related to this post can be found here with the same numbering: https://github.com/nmaggiulli/of-dollars-and-data
