ETFS - TEST THE MIX, NOT THE HUNCH
ETF backtesting: backtest an ETF portfolio with an ETF backtesting tool
Name the ETFs and the rule the way you would say it out loud, such as "60% VTI, 40% BND, rebalanced yearly", and the bench tests it on 20+ years of split- and dividend-adjusted daily data, then stamps an honest verdict.
- 1 THESIS
- 2 EVIDENCE
- 3 BACKTEST
- 4 RISK
- 5 VERDICT
02 EVIDENCE · FUNDAMENTALS
04 RISK · IN PLAIN ENGLISH
Past performance does not guarantee future results. Educational analysis only, not financial advice.
In short
ETF backtesting applies a portfolio mix or a switching rule to historical ETF prices to show how it would have behaved, and doing it honestly means three things most tools skip: total returns with dividends reinvested, a realistic cost for every rebalance, and a comparison against simply holding the benchmark. Agenttrading takes the whole thing as a sentence. Type "60% VTI, 40% BND, rebalanced annually" or "hold QQQ above its 200-day moving average, otherwise cash" and it restates the rule for you to confirm, runs it on 20+ years of split- and dividend-adjusted daily data with a 0.1% cost per trade assumed by default, shades the worst drawdown window, and stamps HELD UP, MIXED, or UNDERPERFORMED against buy-and-hold. Most ETF backtesting still happens either in a long configuration form like Portfolio Visualizer, which returns statistics with no explanation, or in a spreadsheet that quietly uses price-only returns and understates what dividends did. Agenttrading starts at $19 per month, needs no formulas, and executes no trades: it is educational analysis, and past performance does not guarantee future results.
Past performance does not guarantee future results. For educational and informational purposes only. Not financial advice. Consult a licensed advisor.
WHAT YOU GET - ETF BACKTESTING
ETF backtesting, run on the bench
Dividends counted, not quietly dropped
ETF results live or die on distributions. Every test runs on split- and dividend-adjusted data, so a bond or dividend fund is judged on total return rather than the price line that makes it look flat.
Rebalancing charged like it costs something
A yearly rebalance is a set of real trades. Costs are charged at 0.1% per trade by default and printed on the result, so a mix that only wins before costs cannot hide it.
Rotation rules in one sentence
Momentum switches, moving-average filters, and cash sleeves are typed the way you would say them. No condition builder, no formulas, and the restated rule is shown before anything runs.
The verdict is allowed to say no
Most tactical ETF rules lose to holding the index once costs are in. HELD UP, MIXED, and UNDERPERFORMED are stamped honestly, because a backtest that cannot disappoint you is not evidence.
HOW IT WORKS - 4 STEPS
From a sentence to a stamped verdict
Name the ETFs and the weights
Type the mix in plain English, such as "60% VTI, 40% BND" or "equal weight VOO, VXUS, and GLD". Tickers and percentages are enough to start.
Add the rule or the schedule
Say how it behaves over time: rebalance annually, switch to cash below the 200-day, or hold and never touch it. The bench restates the rule for you to confirm.
Run it across 20+ years
The mix is tested on split- and dividend-adjusted daily data with costs charged per rebalance, plotted against buy-and-hold with the worst drawdown window shaded.
Read the risk panel and the verdict
Drawdown, concentration, and sample size in plain English, then the honest verdict. Historical, never predictive, and never personalized advice. You decide.
Past performance does not guarantee future results. For educational and informational purposes only. Not financial advice. Consult a licensed advisor.
On the same bench
Multi-asset mixes and rebalancing schedules are covered in more depth on portfolio backtesting, and the adjusted 20-year record every test runs on is described on historical stock data. If you hold funds for the long haul rather than trade them, research tools for ETF investors is the persona view, and the drawdown you would have had to sit through is explained on investment risk analysis. For the step-by-step method, read how to backtest a portfolio; plans start at $19 per month.
QUESTIONS - ASKED AND ANSWERED
ETF backtesting: the common questions
How do you backtest an ETF portfolio?
Set the holdings and weights, choose a rebalancing schedule, then run the mix against historical total-return data over at least 20 years and compare it to holding a single benchmark fund. The three details that decide whether the result is trustworthy are dividends reinvested, a realistic cost charged on every rebalance, and a long enough window to include 2008 and 2020.
Is backtesting ETFs worth it?
Yes, mostly as a way to retire ideas cheaply. Backtesting rarely uncovers a market-beating ETF rotation, but it reliably shows that a clever-sounding switching rule underperformed a plain index fund after costs, or that a mix you were comfortable with required sitting through a 50% loss. Both answers save real money.
How far back should an ETF backtest go?
At least 20 years where the fund history allows it. Many popular ETFs launched after 2003, so a shorter test can miss the dot-com unwind entirely and quietly measure a strategy only through bull markets. When a fund is too young, the honest move is to say so rather than to splice in a proxy without flagging it.
Can you backtest leveraged ETFs?
You can, and the backtest usually explains why they are not buy-and-hold instruments. Daily-reset leveraged funds decay in choppy markets, so a 3x fund can lose money over a year in which its index finished flat. Testing one on 20+ years of adjusted data with costs shows that volatility drag directly instead of arguing about it.
Does an ETF backtest include dividends?
It should, and here it does. Agenttrading runs on split- and dividend-adjusted daily data, so distributions are treated as reinvested. Price-only backtests, which is what most spreadsheet versions produce, understate bond funds, dividend funds, and broad index funds badly over a 20-year window.
What is the difference between backtesting an ETF and backtesting a stock?
The mechanics are the same, but the sensible questions differ. With a single stock you are testing an entry and exit rule against company risk; with ETFs you are usually testing an allocation, a rebalancing schedule, or a rotation between funds, and the result is dominated by weights and costs rather than timing.
Past performance does not guarantee future results. For educational and informational purposes only. Not financial advice. Consult a licensed advisor.
Your next idea deserves a verdict, not a hunch.
Bring a thesis or a ticker. Agenttrading restates the rule, shows the evidence, runs 20+ years of history, and stamps an honest verdict. You decide.
Past performance does not guarantee future results. For educational and informational purposes only. Not financial advice. Consult a licensed advisor.