TALLIVO
toolsabout
Your numbers will appear here as you use the tools.
home / data / lump sum vs dca
Data study · 1871–2026

Lump Sum or Spread It Out?

You have a sum today. Invest it all at once, or feed it in over months? We ran both against every start month in the record — same money, same dates, only the timing different.

All at once wins
72.0%
of start months, 12-month spread
Median advantage
+4.7%
to investing at once
Worst case, all at once
41.9%
of your money left
Worst case, spread out
60.9%
a much higher floor
Spread overAll-at-once winsMedian edgeBest ever for spreading
3 months64.6%+0.9%17.3% (1929-10)
6 months68.1%+2.3%38.2% (1932-03)
12 months72.0%+4.7%49.6% (1931-08)
24 months75.4%+9.5%76.4% (1931-07)
36 months79.6%+15.0%83.7% (1930-07)

Real (inflation-adjusted) total returns, every start month with a complete window — 1,855 of them for the 12-month spread.

Both sides of this argument are right

Investing at once beat spreading over a year in 72.0% of all historical start months, by a median of 4.7%. The reason is not clever: markets rise more often than they fall, so money in the market sooner earns more, most of the time.

But look at the floors. The worst 12-month stretch left someone who invested at once with 41.9% of their money. The worst for someone spreading it out left 60.9%. Averaging in gives up expected return to buy a substantially better worst case — which is a real thing to want, not a misunderstanding.

So the question is not "which is better". It is which mistake would you rather make: leaving money on the table most of the time, or being fully invested on the worst possible day.

What it looked like at specific moments

StartingAll at onceSpread over 12 moWinner
January 1995a bull market$13,174$11,273all at once by 19.0%
October 2007the month before the crash$6,204$6,951spreading by 7.5%
January 2000the dot-com peak$9,138$9,089all at once by 0.5%
January 2019a strong year$12,504$11,127all at once by 13.8%

$10,000 deployed, valued 12 months later in today's money.

If you do spread it, spread it fast

The advantage to investing at once grows with the delay. Over three months it won 64.6% of the time; over three years, 79.6%, by a median of 15.0%. Every extra month you hold back is another month you are probably waiting through a rise.

Averaging in over three to six months captures most of the psychological benefit at a fraction of the cost. Three years of averaging is mostly just being out of the market.

The comparison most articles get wrong

A common version of this test pits a large lump sum against small monthly contributions over a decade. That is not the same decision and not even the same amount of money at risk — it mostly measures that investing more, earlier, produces more.

Here both paths start with the same sum on the same date and are valued on the same date. The only difference is how fast the money goes in. One assumption to be aware of: uninvested cash earns nothing in nominal terms, because the historical record we use has no short-term rate series and we will not invent one. That mildly favours investing at once, so read the win rate as a slight overstatement.

And a distinction worth being clear about: investing each paycheck as it arrives is not this question at all. That is just investing money when you get it. This only applies when you already hold a sum and are deciding how fast to deploy it. Run your own dates in the investment backtest.

Cite or republish this study

These figures are free to reuse — in an article, a newsletter, a class, or a report — with attribution and a link. No permission needed and no paywall.

Suggested citation

Tallivo, “Lump Sum vs Dollar-Cost Averaging: every start month since 1871,” updated July 1, 2026. https://tallivo.com/data/lump-sum-vs-dollar-cost-averaging

Key findings

  • Investing a sum all at once beat spreading it over 12 months in 72.0% of all historical start months, by a median of 4.7%.
  • But the worst 12-month case left a lump-sum investor with 41.9% of their money against 60.9% for averaging in — DCA buys a better floor, not a better average.
  • The advantage to investing at once grows with delay: 64.6% over a 3-month spread, 79.6% over 36 months.
  • Averaging in won biggest entering the Great Depression (best case began 1931-08); October 2007 is the modern example.

The data

Download the full dataset (CSV)

All 50 states with the federal, FICA and state components broken out. Generated by the same engine that renders the table on this page, so the file and the page cannot disagree.

Chart image

Open the shareable chart (PNG, 1200×630)

Free to republish alongside a credit to Tallivo and a link to this page. Please do not alter the figures in the image.

Dates & method

Published July 25, 2026 · Figures last verified July 1, 2026.
Full methodology · how we check these numbers · report an error

Primary sources

Writing something and need a figure we have not published — a different salary, a specific state, a filing status? Ask and we will run it.

Common questions

Is it better to invest a lump sum or spread it out?
On the numbers, investing at once usually wins: it beat spreading over 12 months in 72.0% of all historical start months, by a median of 4.7%. The reason is unexciting — markets rise more often than they fall, so money in the market longer tends to earn more.
Then why does anyone dollar-cost average?
Because the average is not the only thing that matters. The worst 12-month deployment left a lump-sum investor with 41.9% of their money; the worst for dollar-cost averaging left 60.9%. DCA gives up expected return to buy a much better floor. If a bad outcome would make you sell at the bottom, the floor is worth more than the average.
Does the length of the spread matter?
Yes, and in one direction. Spreading over 3 months, investing at once won 64.6% of the time; over 36 months, 79.6%. The longer you hold back, the more likely you are to be waiting through a rise — so if you do average in, shorter is better on the numbers.
When did dollar-cost averaging win biggest?
Entering right before a crash. The best 12-month case for DCA in the whole record started in 1931-08, during the Great Depression, when it finished 49.6% ahead of going all in. October 2007 — the month before the financial crisis — is the modern example.
What do you assume the uninvested cash earns?
Nothing, in nominal terms. The historical record we use has no short-term interest rate series, and assuming one would mean inventing data. That choice slightly favours the lump sum, because real cash would earn something — so treat the lump-sum win rate as a mild overstatement.
Does this apply to money I earn each month?
No, and this is the most common confusion. If you invest each paycheck as it arrives, that is not dollar-cost averaging as a strategy — it is simply investing money when you get it, and there is no lump sum to compare against. This question only applies when you already hold a sum and are deciding how fast to deploy it.

How we calculate this

For every start month, one path invests the whole sum immediately; the other invests an equal slice at the start of each month of the spread. Both are valued at the end of the same window using a total-return index (dividends reinvested) deflated by CPI. Uninvested cash earns nothing nominally and loses purchasing power in real terms.

Index returns before fees and taxes, US large-cap only. See the backtest for the method and rolling returns for the underlying distribution.

Robert J. Shiller, Online Data (shillerdata.com) — S&P Composite, dividends and CPI, monthly from 1871 →

Historical index returns 1871-01 to 2026-07, before fees and taxes. Past performance does not predict future results. Not financial advice — and notably not a recommendation either way, because the right answer depends on your own tolerance for a bad outcome.