Rearview Join the waitlist

How Rearview compares

These are facts each vendor publishes about its own product, quoted from its own pages and checked on 23 October 2026, with the prices checked again on 6 November 2026. Prices are in US dollars. The sources are listed at the end, and if something here has changed, tell us and we’ll correct it.

Desktop backtesters on a Mac

Runs on a MacPriceSeveral strategies in one portfolioRobustness tools it lists
RearviewNative, for Apple silicon, macOS 26 or later$349 once, with 12 months of updates; $149 for each further yearYes: fixed sleeves, or one shared pool of cashDeflated Sharpe Ratio and PBO with every input, every try counted, random entries, stress tests, a Reality Check against the benchmark, a lockbox, walk-forward tests, genetic and CMA-ES searches, Monte Carlo
RealTestWindows; it “works fine on” a Mac “running BootCamp or Parallels, provided that the VM emulates an x64 CPU”$389, with “all releases and updates for the next twelve months”; $159 for each added year“combine many diverse strategies”Monte Carlo, walk-forward parameters, sequential and genetic optimization
AmiBroker“a Windows application”; Mac users “use Parallels”$299 Standard (32-bit, two threads per Analysis window), $379 Professional, $499 Ultimate Pack Pro; 24 months of free upgrades, then upgrades at 50% offNot described on its features pageMonte Carlo, automated walk-forward, exhaustive and evolutionary optimization
Wealth-LabWindows 10 or 11; “On Mac, a Windows virtual machine such as Parallels or VMware Fusion is required”$49.95 a month, $399.95 or $499.95 a year, or $1,595 for a lifetime licenceMetaStrategiesEvolutionary optimization, rolling windows

On data: a Wealth-Lab subscription includes Wealth-Data, which “generally covers the symbols in the Nasdaq 100, S&P 500, and German DAX indices”, listed and delisted, back to about 1995. Rearview includes no market data: it reads the data you choose, through a documented format, so its price carries no data bill.

Other places people backtest

  • TradingView. Its Pine Script documentation says “Each Pine Script strategy runs on one dataset at a time”, and that a script can make “up to 40 unique calls” to its request.*() functions, or 64 on the Ultimate plan. Its page on repainting says repainting “invalidates backtesting results”, and that one common way of requesting data “will return data from the future on historical bars”.
  • QuantConnect. Its terms say “Data provided through QuantConnect may only be used inside the QuantConnect platform.”
  • backtesting.py. Its issue on testing several instruments, open since September 2019, says “Multiple positions/instruments aren’t yet supported.” A brief look-ahead bias in the first bars of its ATR, which its maintainer confirmed in May 2023, is in an issue that is still open.

What Rearview adds

  • A second engine. Before every release, the engine is checked against a separate engine written only from its rules and hand-worked cases, and against three open-source backtesters on the cases they can model.
  • A look-ahead test on every release. The whole pipeline runs again on data cut off at thousands of dates, and every decision before each cut must come out the same.
  • The same answer every run, byte for byte, on any number of cores, and every saved run reruns from its own folder.
  • “Is this result luck?”, answered with every input shown, and a lockbox that holds back the latest data until you choose to look.
  • Luck checks for a single run: every try of its script counted, 500 runs with random entries at its own rate, and stress tests with costs doubled, signals a day late, entries missed and other start days.

How Rearview is checked describes each check, what it catches and what it can’t.

Sources

Each was read on 23 October 2026.

Product names belong to their owners. Rearview isn’t affiliated with or endorsed by them.