How to use the DCA Scenario Explorer

The short version

1. 1Load the prices

   One click on a Quick Start loads an example. Or list tickers on the Data tab, or make up an asset from a return and a volatility, and set the Date Range.

2. 2Set each scenario

   On Scenarios, each plan gets an asset, an Amount per top-up and an Investment Style: a fixed day, a price move, a technical signal, or a hindsight benchmark.

3. 3Simulate and read

   Press ▶ Simulate. One sentence names the best plan; the Final Summary gives each one's value and ROI. Portfolio mode does the same for whole portfolios.

Nothing is uploaded. Prices come from Yahoo Finance through the site's own server and are kept in your browser; every run is worked out there. This guide is general information, not financial advice.

The screen

Inputs on the left, the answer and the charts beside them, the detail under the charts. Point at a part in the list to see where it is.

Portfolio mode, linked above the inputs as the Portfolio DCA Simulator, runs whole portfolios: several assets at target weights, rebalanced, with fees and a cash account.

Top right on both pages, Tour replays a one minute walk through the page, and Guide opens this page. The ticker reference lists the codes Yahoo Finance uses for shares, funds, indices, currencies and crypto.

Quick Start

Three worked comparisons on real prices: Monthly vs Weekly ($520 a month against $120 a week into SPY), Equity vs Money Market ($500 a month into DHHF.AX against AAA.AX) and Stock vs ETF ($500 a month into AAPL against SPY). One click replaces the scenarios and runs them.

On a phone, press Full screen under a demo and turn the phone sideways to read it.

• Every demo here runs 1 Oct 2021 to 30 Sep 2026, on prices saved with this guide. The tool opens on the last five years to today, so your own run of an example will differ.
• A Quick Start is a fresh start. It clears the scenarios and settings first; prices already loaded stay, so no download is wasted.

The Data tab

What every scenario can buy, and the days it runs over. Load prices once here; scenarios then pick from them.

• Five loads a day, 25 tickers each. The data server is shared, so list everything at once. A ticker already loaded costs nothing, and both pages share the same prices.
• Prices from 1990 on. A load fetches each ticker's whole history from 1 Jan 1990, so changing the dates within it needs no new load.

The Scenarios tab

A scenario is one plan: an asset, how much each top-up puts in, and when it buys. Add as many as you like; each runs on its own and they are compared.

• The amount is per purchase. A weekly style buys 52 times a year, a monthly one 12. Switching between them restates the amount so a year's money stays the same: $500 a month becomes $115.38 a week.

Investment Style

When a scenario buys. Four families, each opened by its pill: only the options of the open family, and the settings of the style picked, show.

• A signal that never fires buys nothing. With the end-of-period box off, a month without a signal is skipped, so less money goes in. Compare such a plan on ROI, not on its final value.
• Indicators start from the first date. They are worked out on the prices in the Date Range alone, so a 200-day average gives no signal in the first 200 trading days.

The Assumptions tab

Three settings every scenario shares.

The answer and the charts

The sentence at the top is the whole answer. Under it, the figures behind it and two charts.

• Blue and gold are options, not good and bad. Scenarios take the site's neutral colours; red is kept for a price falling.
• Zoom and pan a chart with the wheel or a drag; ⟳ resets it. Hover to read every scenario on that day.

Detailed Breakdown and What this assumes

Under the charts, the working behind the answer.

Optimise

The gold gauge beside a setting sweeps it: the scenario on screen runs once for every value in a range, up to two settings at once, so you see how the answer moves and where it peaks.

• Show picks what the sweep ranks by: ROI, Final value, Net, Sharpe, Sortino, CAGR (TWR) or CAGR (MWR).
• Up to 625 runs. 50 values for one setting, or 25 by 25 for two.
• The best is the best for these past prices, not a forecast. Here every day from the 1st to the 28th lands within a few hundred dollars.

Save, export and share

The tool keeps your scenarios in this browser between visits. The rest makes something to keep or send.

• An embed carries its own prices, just the dates it draws, so it shows the same chart on any site without loading anything. A link opens the tool on your scenarios and loads prices as a visit does.

Same inputs, two screens

Portfolio mode runs the same kind of plan on the same prices: a tab per plan, an amount per top-up, a yearly increase, a day it buys on. A plan there holds several assets instead of one. Point at a field in the list to see it on both.

• What it adds. Several assets in one plan, kept at their weights or not, with fees on every trade and interest on idle cash.
• What it holds fixed. The engine: a portfolio of one asset at 100%, with no fees, topped up on the same day, gives the main page's figures, buy for buy, metric for metric.

Portfolio mode: Quick Start

Three Quick Starts: 80/20 vs 60/40 (DHHF.AX shares and AAA.AX cash), Rebal Monthly vs Quarterly and Rebal vs Constant Alloc (both 50/50 SPY and GOVT bonds). Each tops up $1,000 a month.

Portfolio mode: assets and weights

Each portfolio holds its own list of assets, and a target share for each.

Portfolio mode: rebalancing and fees

The Method decides how a top-up becomes units, and whether anything is ever sold to stay on target.

Portfolio mode: top-ups and cash

How much goes in, when, and what money earns while it waits to be invested.

Portfolio mode: the results

The same answer and tiles as the main page, then one portfolio at a time in detail.

• Optimise works here too, from the gauges beside the amount, the schedule, the fees, the risk-free rate, trigger settings and the weights.
• Every portfolio runs on the trading days all of them share, the risk-free ticker included.

The model in one picture

Dollar-cost averaging (DCA) means putting a set amount in on a schedule, whatever the price. Each top-up buys as many units as its money buys that day, in fractions where needed. The holding is worth its units at each day's close. At the end the tool asks two things: what is it worth, and what did it return on the money put in?

• a_ithe money in top-up i
• p_ithe price it fills at (Trade at)
• Uunits held at the end
• P_endthe close on the last day
• DTopped up: the sum of every a_i

The last form is the whole of DCA in one line: your return is the last price over the average price you paid. And a fixed amount pays less than the average price, because it buys more units when the price is low and fewer when it is high. With equal top-ups the average cost is the harmonic mean of the fill prices, which is never above their ordinary mean:

• In the example below, SPY's 60 fills averaged $512.90, but $520 a month paid $486.74 a unit on average: 5% less, from buying more in the 2022 fall.
• The answer compares like with like. When every scenario put in about the same money (within 0.5%), the highest final value wins. When they did not, more money in is not a better plan, so the sentence ranks them by ROI.
• DCA against a lump sum. The tool has no lump-sum scenario. A lump sum puts all the money to work on day one; a monthly plan has, on average, about half of it at work. In a market that rose, as SPY did (from $406 to $763 over the window, up 88%), the lump sum ends ahead. What DCA buys is protection from putting everything in at a peak. CAGR (TWR), below, is the nearest stand-in for the lump sum's rate.

The prices include dividends. Yahoo Finance's adjusted close folds every dividend and split into the price, so a holding grows as if dividends were reinvested. A ticker that falls back to Stooq is unadjusted: it shows ⚠️ in the loaded list, and What this assumes says its dividends are left out.

Data: the prices and the dates

A real ticker is its daily prices. A simulated asset is one random path built from a yearly return and a volatility, by geometric Brownian motion: each weekday's price is the last one times a random factor.

• μReturn, a year
• σStandard deviation, a year
• z_ta random draw, close to a standard normal (the sum of three uniform draws)

Watch out: two markets share only their common days. Put an ASX fund beside a US one and each skips the other's holidays; a seven-day crypto series loses its weekends. The run says how many days it used under the dates.

Watch out: on the main page a simulated path follows the scenario's name. Its seed is the Simulation Seed with the scenario's name, return and volatility, so two scenarios on one simulated asset, named differently, buy into two different paths, and renaming a scenario redraws its path at the next Simulate. To compare two styles on one simulated path, compare them in Portfolio mode, or give each scenario the asset's own name.

Scenarios: the top-ups and the asset

A scenario's money is its amount on each buy day, stepped up once a year if you ask. Nothing waits in cash: a top-up is deposited on the day it buys.

• AAmount per top-up
• gYearly increase
• yyears since the run's first day; ⌊ ⌋ keeps the whole years

Watch out: no fees and no tax here. Each top-up buys its whole amount. That flatters a weekly plan most, which trades four times as often. Portfolio mode charges a fee on every trade.

Investment Style: the day it buys

Every style names a decision day, and the order fills by Trade at. Date styles buy every period. Momentum and technical styles buy at most once a period: the first day their rule holds, or the period's last trading day if Invest at End of Month (or Week) is ticked.

• C_tthe day's close
• C_firstthe close on the period's first trading day
• θThreshold
• nRSI period

Watch out: the Forward styles look ahead. The tool says so in amber, and What this assumes calls them hindsight. Even so, in the example below perfect hindsight within each month moved the result by only about $1,300 either way on $31,200.

Assumptions: the fill, the seed and the risk-free rate

Every rule decides on a day's close, using only prices up to that close (Forward styles aside). Trade at sets where the order fills.

• dthe decision day
• O, Ca day's open and close

The figures: ROI, CAGR (TWR), CAGR (MWR), Sharpe and Sortino

ROI is the plain gain on what went in. The rest come from the daily value, starting on the first day money is at work, and turn the result into a rate a year and a reward for the risk taken.

• F_tthe money topped up that day
• Yyears from the first buy to the last day
• t_i, Tyears from the first buy to top-up i, and to the end
• r_fthe daily risk-free rate, (1 + R_f)^1÷252 − 1

• TWR against MWR is the timing test. TWR grades the asset; MWR grades your money. A monthly plan holds little in year 1 and the most at the end, so MWR leans on the later years. For SPY $520 a month, TWR was 14.21% and MWR 18.06%: the 2022 fall hit only the first few thousand dollars, and most of the money was at work in the strong years after.
• The display switches (Advanced metrics, Candlesticks, Show Buy Date, Show Indicators, Show Total Deposited) change what you see, never a figure.

Watch out: no drawdown, and Sharpe is daily noise. The tool does not report the deepest fall. And with daily returns a few large days dominate Sharpe and Sortino; over five years they rank plans, they do not promise the next five.

Optimise: a sweep, not a forecast

Optimise copies the scenario, writes each tested value in where the form would, and runs the very engine Simulate uses. One setting runs its values; two run every pair.

Watch out: a rate can be won by putting less in. A threshold that rarely fires buys less, and its ROI can top the sweep. When the best run topped up a different amount, the answer says so; rank by Final value or Net to compare the money. And the best setting is the best for these past prices: the next stretch can differ.

Portfolio mode: weights, rebalancing, cash and fees

A portfolio is cash plus units of each asset. A top-up lands in cash, and the Method decides how cash becomes units. Cash waiting earns the risk-free account every trading day; every trade pays its fee.

• u_k, P_kunits and price of asset k
• w_kits Target Weight
• Tcash plus everything held, that day
• V_kwhat asset k is worth
• RRisk-free rate
• f_b, f_sBuying fee, Selling fee

• How return and risk combine. On a day with no trades, the portfolio's return is its holdings' returns weighted by their value: r_p = Σ w_kr_k, cash at its rate. Its swing is less than the weighted swings whenever the assets do not move together. The tool measures that from the portfolio's own daily values, so the benefit shows up in its Sharpe.
• 80/20 against 60/40. More in shares earned more: $78,915 against $75,604 on $60,000, CAGR (TWR) 8.42% against 7.19%. But 60/40 had the higher Sharpe, 1.08 against 0.96: per unit of swing, the cash-heavy mix paid better.
• Rebalancing costs and helps. Over these five years, splitting each top-up 50/50 and never rebalancing ended $1,720 ahead of keeping SPY and GOVT at 50/50: shares beat bonds, and the unrebalanced mix let the winner grow.

Watch out: waiting cash earns nothing unless you set a rate. The Risk-free rate starts at 0%. A quarterly rebalance or a rare trigger can leave real money idle; the tile's Cash share and the table's Interest show how much.

Worked example: SPY, $520 on the 1st of every month

The first scenario of the Monthly vs Weekly Quick Start, 1 Oct 2021 to 30 Sep 2026, filled at the next day's open (the default). The figures are the tool's own, from its Detailed Breakdown and Final Summary.

• The rates a year. Advanced metrics: CAGR (TWR) 14.21%, CAGR (MWR) 18.06%, Sharpe 0.63, Sortino 0.91 at the 4% Risk-Free Rate.
• Its twin, $120 a week, bought 264 times for $31,680 and ended at $49,379: more money, a little less on it (ROI 55.87%). Same money a year, nearly the same result.
• The amount does not change the rate. The Stock vs ETF example puts $500 a month into SPY over the same dates: $47,004 on $30,000, the same 56.68%.

What moves the answer

Each input moved on its own from the worked example, on the tool itself, with the prices this guide ships.

• When you start, and how much you put in, are the answer. A year either way on the start date moves the result by $12,000 to $13,000: more top-ups, and a different stretch of market. The amount moves it in proportion.
• Timing within the month hardly matters. The 10th or the 20th instead of the 1st: under $200. A fill at the close instead of the open: $24. Even perfect hindsight, buying every month at its lowest close, adds only $1,321.
• ROI tells a different story from value. Starting a year earlier lifted ROI from 56.68% to 65.68%, starting later cut it to 48.62%; the amount left it unchanged; a 5% yearly increase put $3,280 more in but cut ROI to 54.50%, because later money had less time to grow.
• So spend your effort on the plan, not the day. Choose an asset and an amount you can keep up through a fall; a clever buy day is worth a rounding error by comparison.

What it leaves out

What the answer rests on, beyond the figures you enter. What this assumes, under the results, lists the ones your run uses.

• The past only. Every real run replays prices that already happened. A plan that won over five years can lose over the next five.
• No tax, and no fees on the main page. Portfolio mode charges a % fee on every trade; neither page has tax, brokerage minimums or spreads.
• One currency, never converted. The symbol is a label; mixing an ASX and a US ticker adds Australian and US dollars as if they were one.
• Dividends ride in the adjusted price, as if reinvested on the day, with no tax or delay. A Stooq series leaves them out.
• Fractional units at the fill price. Every top-up buys its whole amount at the open or close, whatever its size.
• Nothing is sold on the main page, and the holding is valued, not sold, on the last day.
• A simulated asset is one random path with a constant return and volatility: a what-if, not a forecast. Change the seed to see another.
• Indicators see only the Date Range. A long average needs months of the window to warm up before it can signal.
• No drawdown or worst-case figure. Read the Portfolio Value chart for the falls.

Do not lean on it when the best plan topped up a different amount from yours, a winning style is a Forward one, a sweep's best value sits alone on a spike, or your money is in a different currency from the ticker.

This guide and the tool are general information, not financial advice. Past returns do not predict future ones; talk to a licensed adviser before you invest.