DCA vs Lump Sum in Precious Metals: What Ten Real Market Cycles Actually Show
Dollar-cost averaging is usually sold as the safer way to buy precious metals. Run it against ten real historical periods of LBMA gold and silver data and a more honest picture emerges: DCA won fewer than half the scenarios — and the reason it still deserves a place in your process has nothing to do with returns.
Published · 12 min read
Key Takeaways
- Across ten historical periods in our LBMA dataset, dollar-cost averaging beat lump-sum investing in 5 of 10 gold scenarios and 4 of 10 silver scenarios — 9 of 20 in total.
- The pattern is not random. DCA won in falling and choppy markets (2011, 2018, 2021, 2022) and lost in rising ones (2019, 2020, 2024). It holds cash while prices rise, which is a cost whenever prices rise.
- The margins are symmetric and large. DCA beat lump-sum by 29.9% in 2011 silver; lump-sum beat DCA by 32.3% in 2020 silver. This is not a low-stakes choice.
- Because precious metals have risen more often than they have fallen over most measurement windows, a strategy that systematically delays exposure gives up return on average. That is the honest finding, and it is not the usual sales pitch.
- The real case for DCA is behavioural, not mathematical: it converts one high-stakes decision into twelve low-stakes ones, which is what keeps investors invested. A plan you actually follow beats a superior plan you abandon.
Dollar-cost averaging occupies a strange position in investment writing. It is almost universally recommended, almost never tested, and almost always justified with a diagram showing prices falling — which is, unsurprisingly, the scenario in which it performs best. We have ten periods of verified LBMA gold and silver price data in our own dataset, covering the 2008 crisis through the 2024 rally, and it seemed worth simply running the comparison and reporting whatever came out. What came out is more interesting than the standard recommendation. All prices in this article are quoted in US dollars (USD) per troy ounce.
The Test
The setup is deliberately simple, because complexity here mostly serves to hide assumptions. An investor has a fixed budget of $12,000 and one of two choices. Under lump sum, the entire amount is deployed at the opening price of the period. Under dollar-cost averaging, $1,000 is deployed in each of the following twelve months. Both are then valued at the same closing price, thirteen months after the start. The only difference between them is timing.
The price series are the LBMA London PM fix for gold and the LBMA silver price, the same data that drives our DCA calculator, so every figure below is reproducible in the tool with the same inputs. The periods were not selected to make a point — they are the ten cycles already in our education dataset, spanning crashes, peaks, bear markets, chop and rallies.
Gold: DCA Won Five of Ten
| Period | Lump-sum entry price | DCA average cost | Lump-sum final value | DCA final value | Winner |
|---|---|---|---|---|---|
| 2008 — Financial Crisis | $830 | $891 | $14,414 | $13,430 | Lump sum by 6.8% |
| 2011 — Gold Peak | $1,759 | $1,661 | $11,120 | $11,778 | DCA by 5.9% |
| 2015 — Metals Bear Market | $1,128 | $1,180 | $14,223 | $13,602 | Lump sum by 4.4% |
| 2016 — Post-Brexit Spike | $1,276 | $1,256 | $11,850 | $12,041 | DCA by 1.6% |
| 2018 — Hawkish Fed | $1,335 | $1,258 | $11,560 | $12,271 | DCA by 6.2% |
| 2019 — Pre-COVID Run | $1,359 | $1,556 | $15,294 | $13,354 | Lump sum by 12.7% |
| 2020 — COVID Crash | $1,592 | $1,820 | $12,950 | $11,330 | Lump sum by 12.5% |
| 2021 — Recovery Chop | $1,867 | $1,795 | $11,672 | $12,143 | DCA by 4.0% |
| 2022 — Rate Hike Shock | $1,948 | $1,800 | $11,784 | $12,751 | DCA by 8.2% |
| 2024 — Metals Rally | $2,034 | $2,425 | $15,988 | $13,410 | Lump sum by 16.1% |
Silver: DCA Won Four of Ten
| Period | Lump-sum entry price | DCA average cost | Lump-sum final value | DCA final value | Winner |
|---|---|---|---|---|---|
| 2008 — Financial Crisis | $12.37 | $12.51 | $15,900 | $15,722 | Lump sum by 1.1% |
| 2011 — Gold Peak | $40.33 | $31.05 | $8,569 | $11,130 | DCA by 29.9% |
| 2015 — Metals Bear Market | $15.05 | $15.64 | $15,939 | $15,340 | Lump sum by 3.8% |
| 2016 — Post-Brexit Spike | $17.29 | $17.82 | $11,750 | $11,403 | Lump sum by 3.0% |
| 2018 — Hawkish Fed | $16.65 | $15.26 | $10,854 | $11,843 | DCA by 9.1% |
| 2019 — Pre-COVID Run | $15.04 | $16.83 | $14,130 | $12,624 | Lump sum by 10.7% |
| 2020 — COVID Crash | $14.88 | $21.99 | $20,685 | $13,999 | Lump sum by 32.3% |
| 2021 — Recovery Chop | $25.88 | $24.83 | $10,739 | $11,194 | DCA by 4.2% |
| 2022 — Rate Hike Shock | $25.31 | $21.28 | $10,421 | $12,396 | DCA by 19.0% |
| 2024 — Metals Rally | $22.92 | $28.61 | $15,921 | $12,754 | Lump sum by 19.9% |
9 of 20 Scenarios Where Dollar-Cost Averaging Finished Ahead. Five of ten gold periods and four of ten silver periods, across ten market cycles from the 2008 financial crisis to the 2024 rally. Dollar-cost averaging is not, on this evidence, a way to make more money.
The Pattern Is Completely Consistent
Look at which periods DCA won and the mechanism is unmistakable. It won in 2011 — entering at the silver peak. It won in 2018 under a hawkish Fed, in the 2021 sideways chop, and in the 2022 rate shock. Every one of those is a falling or directionless market. It lost in 2019, in the recovery from the 2020 crash, and in the 2024 rally. Every one of those is a rising market.
Dollar-cost averaging holds cash while it waits. Cash is a drag whenever prices rise and a cushion whenever they fall. That is the entire result — everything else is a footnote to it.
This has an uncomfortable implication that most DCA advocacy avoids. If a strategy wins when markets fall and loses when they rise, then its expected performance depends on how often markets rise. Precious metals, over most measurement windows and certainly over the past two decades, have risen more often than they have fallen. A strategy that systematically delays exposure to an asset that trends upward gives up return on average. The 9-of-20 result is not a fluke of period selection — it is what the mechanism predicts.
The Margins Are Not Small
It would be convenient if this were a low-stakes decision where either answer worked out roughly the same. The silver data says otherwise. In 2011, entering at the peak, DCA finished 29.9% ahead — the difference between $8,569 and $11,130 on the same $12,000 budget. In 2020, deploying into the COVID recovery, lump sum finished 32.3% ahead, $20,685 against $13,999.
Those two numbers are almost mirror images, and they capture the trade precisely. Dollar-cost averaging is not a strategy that produces a modest, reliable improvement. It is a strategy that meaningfully protects you from one specific failure — deploying everything immediately before a substantial decline — at the cost of meaningfully reduced participation in the rallies. It reduces the variance of the outcome. It does not raise its average.
So Why Does DCA Still Deserve a Place?
Because the comparison above measures the wrong thing. It assumes an investor who will deploy $12,000 either way, calmly, on schedule, regardless of what the price does. That investor is largely hypothetical, and it is precisely the gap between that investor and the real one where dollar-cost averaging earns its keep.
We wrote recently about the difference between investing and chasing the price — the familiar cycle where an investor waits through the decline, hesitates through the base, and buys in a rush once the rally is obvious. Dollar-cost averaging is a direct structural defence against that cycle. It converts one large, emotionally loaded decision into twelve small, boring ones. Each individual purchase matters little enough that no single one requires conviction, which is exactly why the schedule tends to survive contact with a falling market.
The best strategy on a spreadsheet is worthless if you abandon it in month three.
This reframes the question usefully. The relevant comparison for most people is not "DCA versus lump sum" — it is "DCA versus the capital sitting in cash for eighteen months while I wait for a better entry." Measured against that alternative, which is the genuine counterfactual far more often than anyone admits, dollar-cost averaging wins overwhelmingly and the 9-of-20 record is beside the point.
A Framework That Uses Both
The two approaches are not mutually exclusive, and the sharpest way to use them follows the core-and-tactical structure we have written about before.
- Lean toward lump sum: The allocation decision is thesis-driven and your horizon is measured in decades rather than months; The sum is modest relative to your total portfolio, so a poor entry is survivable; You are buying physical metal and the premium saving on a single larger order is material; You have a documented history of following through on investment decisions without second-guessing them
- Lean toward dollar-cost averaging: The amount is large relative to your net worth and a sharp early drawdown would genuinely shake your commitment; You are new to precious metals and have not yet experienced a 20% drawdown in the asset class; You recognise the waiting-for-a-better-entry pattern in your own past behaviour; Your capital arrives over time from income rather than as a single existing sum, in which case DCA is not a strategy choice at all — it is simply your circumstances
- Use both deliberately: Deploy the core monetary allocation as a lump sum, since it exists for structural reasons and is intended to be left alone; Dollar-cost average the tactical or opportunistic portion where entry price genuinely matters more; Direct ongoing monthly contributions toward whichever metal the gold-silver ratio currently marks as relatively cheap; Keep the schedule mechanical so that neither portion becomes a monthly forecasting exercise
The Honest Conclusion
Dollar-cost averaging is not the safer way to earn more. On ten real cycles of gold and silver data it finished ahead in fewer than half the scenarios, and the mechanism that produces that result — holding cash while prices rise — is not going to change. Anyone selling DCA as a return-enhancing strategy is selling something the data does not support.
What it is, is a device for making a plan survivable. It narrows the range of outcomes, removes the requirement to be right about a single moment, and gives an investor a rule to follow when the price is doing something uncomfortable. Those are behavioural properties, not mathematical ones, and they are worth more than a few percentage points to the large number of investors whose actual failure mode is never deploying the capital at all.
You can run any of these ten periods yourself, for either metal and any budget, in our DCA calculator — the figures in the tables above come from the same dataset and the same calculation.