Lump sum or dollar-cost averaging?

If you already hold a lump sum, the statistics favour investing it sooner rather than spreading it out. In our test on roughly a century of broad-index data, front-loading beat steady dollar-cost averaging (DCA) in about 67% of years.

That is the whole quantitative answer. The rest of this article is about why that number is what it is, what the losing 33% feels like, and why the right choice for you depends on one thing the data cannot measure: which regret you can actually live with.

The mechanism is not cleverness — it is calendar time. Money invested in January spends eleven more months in the market than the same money dripped in monthly over a year, and broad equity markets have drifted upward in far more years than they have fallen.

DCA, by construction, holds part of your money in cash while it waits its turn. In every year the market rises — the majority — that waiting cash is a drag. The ~67% win rate for front-loading is simply the market's upward drift made visible.

This is the same reason 'waiting for a dip' quietly underperforms: cash waiting for a better price is making an implicit bet that prices will fall, and most years they don't.

A two-in-three win rate is a tendency, not a promise. It says nothing about the next twelve months in particular, and the margin of victory is often small — many winning years are won by a little, while the losing years can lose by a lot.

Treat the number the way this project treats every result: as an honest description of the past distribution, measured across calm decades and catastrophic ones alike — not as a forecast.

The years where DCA wins are the years that start falling — 1929, 2008, 2020. In those years a January lump sum buys the top, then watches the drawdown with everything already committed. DCA, by contrast, keeps buying on the way down and ends the year with a better average price.

This is why the choice is really about regret, not arithmetic. A lump-sum investor who panics and sells in the middle of a 2008 has converted a temporary statistical loss into a permanent real one — and at that point every percentage point the strategy 'should' have earned is irrelevant.

The strongest argument for DCA has never been the math. It is that the people using it tend to keep using it, because no single decision is ever big enough to be terrifying.

The lump-sum question only exists if you have a lump sum — an inheritance, a bonus, a windfall. If you invest a fixed amount from each paycheck, you are not 'choosing DCA'; you are simply investing money as it arrives, which is the earliest possible moment for each yen.

In that sense, salary investors are already front-loading. The practical advice for them is unchanged by anything in this article: automate the contribution, and don't let cash pile up waiting for a better price.

Japan's NISA puts a ceiling on the question. The annual quota (¥3.6M under the current system) means even a large windfall cannot be fully invested tax-free at once — front-loading inside NISA mostly means filling each year's quota early rather than late.

That version of the question has the same answer for the same reason: quota filled in January compounds tax-free for eleven more months than quota filled in December. Our front-loading study was motivated by exactly this mechanic.

Anything beyond the annual quota is a separate, taxable decision — and the tax-free compounding inside NISA is precisely why our research treats the NISA index holding as the benchmark every strategy must beat.

If you hold a lump sum and can genuinely hold through a 2008-sized drawdown without selling: the data favours investing it early, accepting that roughly one year in three you will briefly look wrong.

If you know — honestly — that buying the top with everything would break you: spread it over a fixed schedule and automate it, because a slightly worse expected outcome you can actually hold beats a better one you abandon at the bottom.

The only clearly losing move is the popular one: keeping the money in cash indefinitely, waiting for a dip that is never officially announced.