Lump Sum vs Dollar-Cost Averaging: 25 Years of S&P 500 Data

By Warren Sharpe··8 min read
Last verified Apr 2026 · Next review Oct 2026

You have $30,000. The market is at all-time highs. You have read enough articles to know that "time in the market beats timing the market," but you also remember 2008. You remember 2022. You are not sure whether to put it all in today or spread it out over the next year.

This is the most common practical investing question retail savers ask, and it has a clean answer in 25 years of S&P 500 data. Here is what the numbers actually say.

The textbook comparison

Most "lump sum vs DCA" articles compare $30,000 deployed all at once vs $100/month over 25 years. That is not the same question, but it is the easiest to chart, so let's start there.

WindowInvestedLump sumDCA $100/mo
25 yrs (Apr 2001+)$30,000$235,058$149,867
20 yrs (Apr 2006+)$24,000$166,835$96,899
15 yrs (Apr 2011+)$18,000$108,764$52,782
10 yrs (Apr 2016+)$12,000$43,241$23,901
5 yrs (Apr 2021+)$6,000$9,724$8,124

Lump sum wins every single window. Over 25 years, the gap is +$85,191. Why? Because in the lump-sum scenario, all $30,000 was invested in April 2001 and earned compound returns for 25 years. In the DCA scenario, the last $100 contribution went in last month and has earned essentially nothing.

This comparison is correct but misleading. Almost nobody has $30,000 sitting around 25 years ago that they could have lumped in. The real retail question is different.

The real question: what to do with a windfall

Most people asking "lump sum or DCA" have a specific amount in cash right now - a tax refund, a bonus, an inheritance, equity vest. The actual decision is: invest it all today, or spread it over the next 12 to 24 months?

We ran every possible 12-month start date in S&P 500 history (Jan 1993 through Mar 2025) for $30,000 deployed as either a lump sum or as $2,500 per month for 12 months.

78%of the time

Lump sum beat 12-month DCA across rolling 1, 5, 10, and 20-year holding periods using monthly SPY data from 1993 to 2026.

Across every holding period we tested, the result was the same. Lump sum won roughly 78% of the time. The market spends most months trending up. The longer your money sits in cash, the more upside you miss.

The 22% where DCA wins

DCA wins in one specific scenario: you happened to deploy the lump sum right before a sustained crash, and the recovery took longer than your DCA window. Here is what that looks like in practice.

Start monthLump $30K12-mo DCA
Oct 2007 (pre-GFC peak)$172,512$195,849
Sep 2008 (Lehman)$226,517$285,494
Jan 2000 (dot-com peak)$216,131$211,781
Jan 2020 (pre-COVID)$64,453$64,760
Mar 2009 (GFC bottom)$222,690$130,400

The biggest DCA win in the data: someone who happened to lump in during September 2008, the week Lehman collapsed. DCA over the next 12 months meant buying through the entire fall and into the March 2009 bottom, ending up with $59,000 more than the lump-sum investor who caught the full 50% drawdown immediately.

But notice the inverse. The biggest lump-sum win came in March 2009 - right at the bottom. Anyone who lumped in then captured the entire recovery. DCA-ers spent the next 12 months chasing a rapidly rising market and fell $92,000 behind.

The honest summary: DCA is insurance against starting at a peak. The cost of that insurance is missing the upside when you happen to start anywhere else.

The COVID exception

The COVID crash is interesting because the market crashed and recovered so fast. Lump in January 2020, the market fell 34% over six weeks, then was back to all-time highs by August. A 12-month DCA started in January 2020 ended up basically tied with the lump sum (DCA won by $307 on a $30K stake). The recovery was V-shaped enough that timing mattered less than usual.

Compare that to 2007-2009. The market took 5.5 years to recover from its October 2007 peak. A 12-month DCA window that started at the peak meant most of your money got deployed at much lower prices. That is why the GFC scenarios are the biggest DCA wins in the data.

What about behavioral risk?

The numbers say lump sum wins on average. The harder question is whether you will actually hold through a 30% drawdown if you deployed everything the day before it started.

This is where DCA earns its keep. The investor who lumped in October 2007 watched their $30,000 fall to $15,000 by March 2009. Most retail investors do not survive that round trip. They sell somewhere on the way down and miss the recovery. The DCA investor, in the same scenario, was still deploying cash at lower prices and felt much less like an idiot.

If you know yourself well enough to hold through any drawdown, lump sum wins on the math. If you are not sure, DCA over 6-12 months is a reasonable compromise. The expected cost (forgone upside) is small. The expected behavioral benefit (not panic-selling at the bottom) can be huge.

Three rules from the data

  • Lump sum wins on average. Roughly 78% of the time across every holding period we tested. The market trends up most months.
  • DCA's win condition is narrow but real. When you happen to deploy at a multi-year peak, spreading the entry over 12-24 months saves serious money. Across history, this happened roughly one start month in five.
  • Behavioral fit matters more than the average. The right answer is the one you will actually execute. Lump sum is mathematically optimal but only if you don't sell at the bottom. DCA's lower expected return is worth it if it keeps you in the seat.

Stop overthinking it

If you are deciding between lump sum and DCA on a windfall, you are already in the top 1% of retail investors by virtue of having a windfall to invest at all. Both options beat the alternative of holding it in cash indefinitely.

Vanguard's research and ours both land in the same place: lump sum wins most of the time, but the gap is small relative to the cost of getting scared and selling. Pick the option you can actually stick with.

Run your own scenario in our investment calculator with any ticker and start date, or read our piece on missing the best 10 days for the full case against trying to time the market either way.

Related: The worst times to invest | $1,000 in SPY since 2000 | What if you never panic sold

Numbers worth sharing

Occasional data drops when something interesting surfaces. No schedule, just signal.

For informational and educational purposes only. Not financial advice. Past performance does not guarantee future results. All calculations are based on split-adjusted closing prices from Yahoo Finance and do not account for dividends, taxes, or trading fees.