Three investors, and the one who does worst#
Picture three people.
The first refuses to buy at all-time highs and waits patiently for a 10% pullback. Disciplined. Sensible-sounding.
The second has catastrophic luck — invests every single year, always at the exact peak.
The third is too nervous to start, so the money sits in a savings account while inflation quietly removes purchasing power.
Intuition says the patient one wins and the unlucky one gets punished. Intuition is wrong on both counts.
What the record actually shows#
Buying at highs is not a penalty. Citi Wealth’s Investment Lab found that investors who bought the S&P 500 on record-high closes earned returns essentially identical to those buying on any other day, across 1-, 3- and 5-year horizons back to January 1960.
J.P. Morgan, using 1988–2024 data, found it’s actually slightly better:
| Horizon | Bought at a record high | Bought on any day |
|---|---|---|
| 1 year | 14% | 12% |
| 3 years | 46% | 41% |
| 5 years | 82% | 76% |
The reason is structural rather than mystical. All-time highs cluster inside bull markets — periods of earnings growth, economic expansion and heavy flows. Healthy markets don’t set one record and collapse; they set a run of them. A new high is evidence you’re in one of those periods, not evidence the period is ending.
Waiting has a price that never appears on a statement#
The dip-waiter’s cost is invisible, which is why it goes unnoticed.
Sometimes the 10% pullback simply doesn’t arrive. 31% of all-time highs since 1950 acted as a floor — the index never traded more than 5% below that level again. Nearly a third of the time, the entry point you’re waiting for is already behind you permanently.
And corrections are rarer than the fear implies: drops greater than 10% have occurred only about 9% of the time within a year of an all-time high, and rarer still over longer horizons.
Even when the dip does come, it often arrives at a level well above where you could have bought. You get your 10% discount — off a price 25% higher than the one you refused.
Meanwhile the meter runs: missed gains, missed compounding, missed reinvested dividends, and above all missed time, which is the input you can never buy back.
The best days hide inside the worst weeks#
Hartford Funds’ analysis of the flip side is stark. Missing only the 10 best days over a 20-year period cuts total returns by more than half. Missing the 20 best days cuts them by over 70%.
The essential detail is when those days happen. They cluster right after crashes, in the most volatile and frightening stretches — nine of the ten largest single-day gains landed during recessions, six of them during bear markets.
So the person waiting for clarity systematically misses the recovery. Clarity arrives after the rebound. That’s what clarity is.
Your brain is running the wrong software#
None of this is a knowledge problem. It’s a wiring problem, and it has a name.
Loss aversion, from Kahneman and Tversky’s prospect theory, describes the finding that losses feel roughly twice as powerful as equivalent gains — experimental estimates cluster between 1.5x and 2.5x, with 2.0 as the textbook value.
Add regret aversion and you get the specific paralysis around record highs. Buying at a peak and watching it fall produces a vivid, blameable error. Failing to buy and watching it rise produces a vague, forgettable non-event. The two outcomes can cost identical amounts and feel nothing alike.
Markets spend a surprisingly large share of their time near record levels. The alarm your brain sounds at a new high isn’t reading the market. It’s reading your imagined regret.
The one legitimate caveat#
There is a real objection to all of this, and it isn’t about timing.
Every study above uses US market data, and the S&P 500 today is unusually concentrated. Technology has reached roughly 38–41% of the index, a level not seen since the dot-com era. The MSCI All Country World Index holds around 2,500 stocks — about five times the S&P 500’s count.
The S&P 500 has outrun the MSCI World for decades, and that’s not an accident of measurement; US firms genuinely dominated. But “all-time highs are safe to buy” is a claim resting on one index’s history during a period of exceptional US performance. If a handful of mega-caps stumble, the index falls because of them specifically.
That’s an argument for broader geographic diversification. It is not an argument for sitting in cash — the dip-waiter’s problem is unaffected by which index they’re refusing to buy.
Summary — and what to do about it#
The record high isn’t the risk. Waiting for permission is.
- Invest on a schedule, not on a signal. Dollar-cost averaging removes the entry-point decision entirely, which is the decision you’re most likely to get wrong.
- Treat “waiting for a dip” as an active market call. It is one, and it’s the same bet as picking a top — just dressed as caution.
- Recognise the feeling for what it is. Losses register about twice as hard as gains. The alarm at a record high is loss aversion, not analysis.
- Remember that recovery days hide inside crashes. Nine of the ten best days came during recessions; sitting out for clarity means missing them.
- Diversify globally rather than defensively. The concentration caveat is real; the answer is a broader index, not cash.
- Zoom out on the peak. A market top today routinely looks like a valley from ten or twenty years away.
The unlucky investor who bought every peak for decades still beats the careful one waiting for the right moment. Time in beats timing, even when the timing is deliberately terrible.
Sources & further reading#
- TheStreet on Citi Wealth’s all-time-high analysis — returns from record-high entries essentially match any-day entries since 1960.
- IndexBox on J.P. Morgan’s 1988–2024 study — the 1/3/5-year return comparison and the 31% floor finding.
- Hartford Funds, “Timing the Market Is Impossible” — the cost of missing the 10 and 20 best days.
- A Wealth of Common Sense, “Missing the Best & Worst Days” — why the best days cluster in recessions.
- BehavioralEconomics.com on loss aversion — Kahneman and Tversky’s prospect theory and the ~2x coefficient.
- AlphaEx Capital, S&P 500 sector breakdown 2026 — technology’s share of the index.
