Portfolio Visualizer vs testfolio: which free tier keeps the history

Two free tiers, capped on different axes

Both tools take tickers, weights and a rebalancing rule and return CAGR, drawdowns and rolling returns in a browser. Neither has a public API. Both free tiers withhold the same two things: nothing saves and nothing downloads, on either side. Beyond that, they cap different things.

• Portfolio Visualizer needs no login, opens every module and sets no quota on how many analyses you run. It caps a portfolio at 15 assets and cuts a ticker backtest to the last ten years, with the current month left out. Our card's run of VFINX requested from 1985, on 13 September 2026, came back as January 2017 to August 2026. The asset-class allocation backtest is exempt: the same session returned January 1972 to August 2026.
• testfolio gives the full available history on every ticker, across 5 portfolios of 20 tickers each. Its caps sit on the models instead: Monte Carlo at 15 years and 500 runs, factor regression at 6 factors, PCA at 3 tickers. It needs an account for part of this, because signed out, factor regression and PCA are unavailable.

So Portfolio Visualizer's free tier withholds history and testfolio's withholds forward-looking horizon. For a question about specific funds over more than ten years, testfolio's free tier answers and Portfolio Visualizer's does not. For a question about an asset-class mix, Portfolio Visualizer's free tier answers back to 1972 with no account at all; testfolio reaches further back only through the simulated series described below.

Where the long history comes from

Portfolio Visualizer keeps two separate data sets. Its 38 asset-class series are monthly from January 1972 and stitched: academic data first, from Kenneth French's library, AQR, Alpha Architect, the q-factors and Novy-Marx, then a representative index fund from the year that fund launched. The FAQ sources the series row by row. Ticker data for US and Canadian stocks, ETFs, mutual funds and closed-end funds comes from Morningstar and CSI, and reaches back to each security's own inception: September 1976 for VFINX, January 1990 for Apple. A ticker's history ends where the fund began, and a free session sees the last ten years of it. Basic adds what the card lists as asset backfills; the card does not say how they are built. Inflation is CPI-U, and a London ticker came back as "Unknown symbol" in an anonymous session.

testfolio runs one data set, daily, and extends it with models. Its US series go back to 1885, built from Schwert's historical stock data, S&P 500 index data and Shiller's dividends and CPI, with Ken French's library, AQR, MSCI, LBMA gold and FirstRate Data behind other series. A run of SPYSIM on 14 September 2026 returned 35,435 daily values from 20 March 1885 to 11 September 2026. Most series start later, and the preset catalogue lists a coverage range for each. Thirteen exchange suffixes reach Canada, the UK, six other European markets, Australia and Japan, and results convert into seven reporting currencies, each deflated by its own CPI.

The resolution differs as well. Portfolio Visualizer computes everything from month-end balances, so a fall that recovers inside a month never shows in its maximum drawdown. testfolio computes daily, which is why it can model a leveraged fund's daily reset at all. The same portfolio over the same window can show a deeper drawdown on testfolio, with no input wrong on either side. Daily bars also land a few days late: the 14 September run ended on 11 September.

What a simulated series is, and what it is not

105 of testfolio's 121 dated preset series are SIM tickers. Each one is a modelled segment covering the years before a fund existed, joined onto that fund's actual returns. Each carries a stated source and a stated assumption. REITSIM is the Fama-French real estate portfolio less 0.13% a year, then STMDX and DFREX, then VNQ. BNDSIM is a VBMFX backfill less 0.12%. TLTSIM is a synthetic bond series built from 20- and 30-year Treasury yields. What a SIM ticker offers is a way to ask how an exposure behaved through years the fund itself never traded.

It is not a fund record. Nothing in a result marks where the model ends and the fund begins, so one line on the chart carries two different kinds of claim. The definitions are one-line summaries, not a reproducible specification, and testfolio LLC publishes no methodology paper. The leveraged proxies show how far the assumptions can carry the answer. UPROSIM is 3x daily SPYSIM with an assumed financing rate, 1.115x swap exposure, a 0.42% spread and a 0.95% expense ratio. Requested on 14 September 2026, TQQQ over its real life from February 2010 returned a 42.3% CAGR, while TQQQSIM back to 1986 returned 13.3% with a maximum drawdown of −99.98%, peaking in March 2000 and never recovering. That loss is a property of the model, not something any holder experienced.

Portfolio Visualizer's 1972 asset-class history is a splice too: academic series first, an index fund after. The difference is presentation. Portfolio Visualizer labels it an asset class and documents each row's source, where testfolio prints a ticker-like name over the join. Neither tool's pre-fund years are returns that a fund actually delivered.

Which tool models the question you came with

A rebalanced backtest of specific funds. Both run it. testfolio's free tier runs it to each fund's inception, or further on a SIM ticker; Portfolio Visualizer's free tier runs ten years of it, or its full history on asset classes.

Withdrawals. testfolio's backtester takes a cashflow schedule and returns withdrawal statistics, beside a separate safe-withdrawal calculator, all on full history on the free tier. Portfolio Visualizer puts cashflows against its Monte Carlo in Financial Goals and Asset Liability Modeling. A ten-year ticker window contains no complete retirement, so a historical withdrawal test on named funds is a testfolio job unless you pay Portfolio Visualizer.

Monte Carlo. Portfolio Visualizer projects from historical, statistical, parameterised or forecast returns, with fat-tailed distributions and a bootstrap option. Our card records no free-tier limit on that module. testfolio's free Monte Carlo stops at 15 years and 500 runs, which is shorter than a 30-year horizon. Pro allows 70 years, and Pro+ 100 years and 10,000 runs. The testfolio card does not say how its Monte Carlo generates returns, so do not assume the two products' outputs are the same statistic. The Monte Carlo glossary entry explains why the generator changes the number.

Factor regression. Portfolio Visualizer runs Fama-French, Carhart, q-factor and AQR models against a fund or a portfolio, with performance attribution beside it, and opens the module without a login. Our card verified the ten-year cap on the ticker backtest only, and does not record whether it also trims a free regression's window. testfolio regresses on Fama-French and AQR factors on its free tier, signed in, up to six factors at a time, over full history. Either way, a regression describes the window it ran on and nothing outside it.

Optimisation. Both have an optimiser and an efficient frontier. Only Portfolio Visualizer has Black-Litterman and rolling optimisation, and they are on its free tier, inside its 15-asset and ten-year limits. This is the one axis on which testfolio has no answer at any price.

Tactical allocation. Portfolio Visualizer ships six canned models — moving average, momentum rotation, dual momentum, adaptive allocation, market valuation and target volatility — that you parameterise. testfolio's free tier includes tactical allocation with signals, and its Max plan adds a Tactical Strategy Optimizer.

What each subscription buys

The two cards carry the same normalised field, and it means different things on each. Portfolio Visualizer's price_amount of 30 dollars a month for Basic and 55 for Pro are monthly equivalents of annual bills, 360 and 660 dollars a year, with no month-to-month option offered. testfolio's 15, 30 and 100 dollars for Pro, Pro+ and Max are month-to-month prices. Billed annually they are 12.50, 25 and 80 dollars a month, or 150, 300 and 960 dollars a year. The smallest payment either will accept is 15 dollars for one month of testfolio Pro, against 360 dollars for a year of Portfolio Visualizer Basic. The trials point the other way. Portfolio Visualizer gives 14 days of everything, and testfolio gives 24 hours.

Portfolio Visualizer Basic buys full ticker history with year-to-date results, 150 assets, 50 saved models, CSV, Excel and PDF export, and asset backfills. testfolio's free tier already gives the full history. It does not give saving, export or more than 20 tickers a portfolio, and those start at Pro (downloads, uploads, saved runs, ticker aliases) and Pro+ (200 tickers a portfolio).

Portfolio Visualizer Pro buys custom return series, management fees, saved capital market assumptions, customised assumptions on a US federal and state tax model, team account sync, and a licence for commercial use; Basic is licensed for personal and educational use only. The testfolio card records none of these at any tier, including any tax model or licence terms for output, so on those points this page has nothing to set against Pro. Branded PDF reports are on Pro; testfolio's PDF reports are on Max.

testfolio Max sells the one machine interface either product supports, an MCP server that connects to Codex and Claude Code. Portfolio Visualizer has no API, no Python client and no MCP server at any tier.

The two cards were verified a day apart, Portfolio Visualizer on 13 September 2026 and testfolio on 14 September, so their prices are comparable with each other. Both are nearly four weeks old today. testfolio labels its current Pro and Pro+ rates "early member" pricing and says they will rise for new members, so check its pricing page before relying on these figures.

Which one, and when

Your question is an asset-class mix and you want no account: Portfolio Visualizer's free asset-class backtest, from 1972. Neither subscription adds anything to that question.

You are testing named funds over more than ten years, on daily data, on a non-US listing or in a non-US reporting currency, or you want a factor regression without paying: testfolio's free tier. Before trusting any result older than a fund, look up where its SIM segment ends in the preset catalogue, because the chart will not show you.

You need a forward Monte Carlo longer than 15 years: testfolio Pro for the month you need it, at 15 dollars. It is the only paid option on this page that can be rented for a single month.

You use Black-Litterman or rolling optimisation: Portfolio Visualizer, free while 15 assets and ten years of ticker data are enough, and Basic when they are not. You need its output in paid client work, or customised US tax assumptions: Portfolio Visualizer Pro, which is the only licence on this page that covers commercial use.

Otherwise: testfolio, free. Paying Portfolio Visualizer 360 dollars a year mostly buys back the ticker history that testfolio's free tier gives away. Pay it for the optimiser modules or the licence, not for the history.

Portfolio Charts is not on this page because it models asset classes on annual data and runs no simulation, so it answers neither the ticker question nor the Monte Carlo one. Its own reason to exist is a non-US home currency for an asset-class question. For each of Portfolio Visualizer's modules taken one at a time, and what replaces it, see what to use instead of Portfolio Visualizer. Other planners and backtesters are on the retirement planning page.