With the S&P 500 hitting its thirty sixth all-time excessive of 2025 final week, many traders are questioning whether or not now is an efficient time to purchase. In any case, wouldn’t you get a greater worth for those who waited? Technically, you very doubtless would. As I said in Just Keep Buying:
In the event you randomly picked a buying and selling day for the Dow Jones Industrial Common between 1930-2020, there may be over a 95% probability that the Dow would shut decrease on some day sooner or later.
So whereas we’re doubtless to see a lower cost by ready, sadly, these decrease costs are uncommon (i.e. they solely happen on 5% of days). Extra importantly, these 5% of days can’t be recognized forward of time.
However, what about after an all-time excessive? The evaluation above was finished throughout all buying and selling days. If we subset to solely all-time highs, would the likelihood of seeing a lower cost sooner or later change?
To check this, I re-ran this evaluation from January 1915 to October 2025 for the Dow Jones Industrial Common. Throughout all buying and selling days, the prospect that the Dow would shut decrease on some day sooner or later was 96.7% (even greater than throughout 1930-2020 interval).
However, for those who purchased at an all-time excessive, the likelihood that you just’d see a lower cost sooner or later was 97.0%. It was barely greater than when the market is just not at an all-time excessive.
Whereas this may appear to be proof that you just shouldn’t purchase at all-time highs, sadly, this distinction was not statistically vital. The distinction between 96.7% and 97.0% appears to be simply random probability. In different phrases, if you purchase at an all-time excessive you might be no extra doubtless to see a future lower cost than shopping for on every other buying and selling day.
So whereas shopping for U.S. shares at an all-time excessive doesn’t appear to hurt us, does it assist us?
Traditionally, probably not.
From 1950-2025, the longer term 1-year returns of the S&P 500 are literally decrease when close to an all-time excessive (“Close to ATH”) than when off an all-time excessive (“Off ATH”) [8.3% vs. 10.1%]. On this case “Close to ATH” means anytime the S&P 500 is inside 5% of an all-time excessive, “Off ATH” is all different instances:
As you’ll be able to see, the 1-year return distributions are comparable although the “Off ATH” distribution is shifted barely to the suitable (i.e. has higher returns).
But when we glance over a 3-year return window, this adjustments. For 3-year future returns, it’s higher to be close to an all-time excessive than off of 1 [8.4% annualized vs. 7.8% annualized]:
As a result of the 1-year and 3-year returns level in reverse instructions, this implies that market timing primarily based on all-time highs doesn’t work. In different phrases, being close to an all-time excessive doesn’t appear to affect future U.S. inventory returns in any constant or significant means.
Whereas there are lengthy stretches the place the S&P 500 doesn’t hit all-time highs (see beneath), these intervals have been fairly uncommon:
However what about different markets? Although the S&P 500 doesn’t appear to have any distinction in returns when close to all-time highs, what about non-U.S. equities? Let’s flip to these now.
What About Worldwide Shares at All-Time Highs?
With regards to worldwide shares, I checked out how the MSCI EAFE (“Developed shares”) and MSCI Emerging Markets (“EM shares”) indices carried out going again to 1970 and 1989, respectively. General the EAFE (Developed shares) are inclined to do barely higher when close to an all-time excessive over the subsequent yr [9.0% vs 7.2% on average]:
Nonetheless, these shares are inclined to do barely worse when close to an all-time excessive over the subsequent three years [5% annualized vs. 6.7% annualized]:
This means that any momentum for EAFE shares tends to be short-lived.
Nonetheless, that is not the case for Rising Market shares, which do about 2x higher over each 1-year and 3-year intervals when close to all-time highs. Over the subsequent yr, EM shares return 15.6% (on common) when close to an all-time excessive versus solely 7.9% when off of 1:
And over the subsequent three years, EM shares return 10.5% (annualized) when close to an all-time excessive versus solely 5.1% (annualized) when off of 1:
This means that momentum performs a much bigger function with EM shares than with U.S. or Developed shares.
We are able to see this within the chart beneath which exhibits the MSCI Rising Markets index over time with all-time highs plotted in crimson. Be aware how EM shares appear to have bursts of excessive returns adopted by prolonged intervals of stagnation:
Whereas I might like to definitively conclude that EM shares usually tend to pattern, we should remember that this solely covers just a few market cycles. When you may have only some observations to go off of, it’s troublesome to make generalized statements.
However, there isn’t any historic proof that purchasing worldwide shares (Developed or EM) close to an all-time excessive is adopted by worse short-term efficiency. If something, there may be some proof that short-term efficiency could also be barely improved (a minimum of for EM shares).
Now that we’ve checked out worldwide shares, let’s flip our consideration to 2 different asset courses which have finished fairly effectively this yr.
What About Bitcoin and Gold?
In 2025 each Bitcoin and gold hit all-time highs and have been the bedrock of the debasement trade. How do these asset courses are inclined to carry out when close to all-time highs? Extremely effectively, particularly over the subsequent yr.
Traditionally, gold returned 37.9% (on common) over the subsequent yr when close to an all-time excessive and solely 4.1% when off of 1 (from 1979-2025):
When gold goes on a run, it actually runs over the subsequent yr.
You would say the identical factor about Bitcoin from 2014-2025:
Bitcoin returned 145.9% (on common) over the subsequent yr when close to an all-time excessive in comparison with solely 78.4% when off an all-time excessive throughout this era.
Sadly, these incredible returns don’t appear to persist over longer intervals. When trying on the 3-year timeframes, Bitcoin performs comparatively worse when close to an all-time excessive than when off one [23.1% annualized vs. 61.6% annualized]. And gold does roughly the identical in both situation [5.0% annualized vs. 4.2% annualized].
Gold particularly has been very vulnerable to momentum given how lengthy it tends to remain between all-time highs:
What does this imply for you as a person investor?
In the event you do personal gold or Bitcoin, count on these property to pattern for some time after which have prolonged intervals the place they aren’t making new highs. Nonetheless, when gold and Bitcoin do make new highs, this tends to be bullish within the brief time period.
In fact, all-time highs are bullish till they aren’t. Have gold and Bitcoin’s current runs come to an finish? I don’t know. Sadly, historical past is the examine of every little thing that’s occurred thus far. It’s very doable that this time is completely different.
Is This Time Completely different?
Although the longer term is at all times unsure, the information introduced right here ought to offer you some reduction when shopping for property close to all-time highs. Whether or not you might be investing in U.S. shares, worldwide shares, gold, or Bitcoin, the proof means that all-time highs mustn’t alter your common funding plan. If something, they’re impartial to barely bullish within the brief time period.
However, is that this time completely different?
Belief me, I get it. A number of months in the past I explained why I was bearish on U.S. stocks attributable to some excessive valuations I used to be seeing within the tech area. I nonetheless imagine this thesis to be true. I imagine AI is overhyped and some AI maximalists are starting to believe so as well.
However, I haven’t stopped shopping for U.S. shares. What did I do as a substitute? I rebalanced my 401(okay) from 100% threat property to 80% threat property + 20% intermediate-term U.S. bonds. It decreased my total allocation to U.S. shares by about 1.5%.
That’s it. I actually do sin a little. Simply sufficient to fulfill my behavioral urge with out risking my monetary future.
In fact, I don’t know the way the AI race goes to finish. I don’t know if this time is completely different. However avoiding U.S. shares fully as a result of they’re at all-time highs would’ve been a foul technique to comply with.
In the event you’re actually apprehensive, do what I do and sin slightly. Then maintain your nostril and Just Keep Buying.
Blissful investing and thanks for studying!
In the event you preferred this put up, take into account signing up for my newsletter.
That is put up 475. Any code I’ve associated to this put up could be discovered right here with the identical numbering: https://github.com/nmaggiulli/of-dollars-and-data
