Observed arrival · 2026-10-05
Adversarial Backtesting puts trading strategies on trial
A paid audit service re-simulates algorithmic trading backtests independently, then compares trades to check whether a strategy’s apparent edge holds up.
○Open
⊠Login
$Paid
†Ads
✦Pretty
●Pro
◎Niche
◉Human
⚑Risk
ƒJS
- For
- Algorithmic FX traders checking backtest results
- Worth noticing
- The case study describes two separately written simulators compared trade-for-trade before results were reported.
Field notes
The case study says configurations were selected on a training window and then scored on a held-out test window, with multiple-comparison rates disclosed. It also describes checking results with two simulators written separately: a fast vectorized engine and a simpler bar-by-bar version. The page reports that this comparison exposed a reversal-handling bug; these are the service’s own reported findings.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
○OpenPublic substance visible
$PaidCommerce or pricing visible
✦PrettyNotable craft visible
●ProPolished or operationally mature
◎NicheUnusually specific use
One card from the complete issue