# Prismatic Algos Reports Trading-Filter Results Across 150 Live Accounts By Prismatic Algos, Prismatic Algos · Published 2026-11-12 · Software & Technology on MarketScale Canonical: https://www.marketscale.com/industries/software-and-technology/prismatic-algos-reports-trading-filter-results-across-150-live-accounts > Prismatic Algos released research on 105 strategy backtests and 150 live accounts testing trading-halt and risk-scaling filters in its free ACI benchmark. ## Key points - No tested trading-halt or risk-scaling rule outperformed its randomized comparison across 56 trend, 26 mean-reversion, and 23 session-clock backtests plus 150 live accounts. - The free ACI dashboard uses 12 reference strategies across 17 symbols and three mechanism families, with data transparency about sources, dates and sample sizes for EA developers and portfolio teams. - ACI is descriptive research not validated for predicting returns or improving trading decisions; Prismatic Algos removed the proposed sizing multiplier on Oct. 28 based on the findings. ONLINE, Nov. 12, 2026 — [Prismatic Algos](https://prismaticalgos.com/press/algo-consistency-index) announced the public research package for its Algo Consistency Index (ACI), including tests of 105 strategy backtests and 150 live trading accounts that did not establish an advantage for the tested trading-halt and risk-scaling rules over matched random comparisons. The company is making the findings available alongside a free reference dashboard for Expert Advisor developers, trading-technology researchers and algorithmic portfolio teams. “A dashboard can describe what strategies have experienced without telling you what to trade next,” wrote Cree Lawson, founder of Prismatic Algos, in a companion article published Nov. 6. The article is available at https://www.mql5.com/en/blogs/post/779529 . The underlying T5 findings are dated Oct. 28, 2026. The study examined 56 trend strategy backtests, 26 mean-reversion backtests and 23 session-clock backtests, together with 150 live accounts. It tested rules including halting below the 20th percentile, scaling exposure by a reading band and halting after a market-drift alert. No tested family or rule beat its same-window randomized comparison. Limited statistical power in the trend tests means the findings do not establish that every possible trading filter is ineffective. For software teams building automated trading controls, the findings address a practical development decision: whether a dashboard reading supplies enough evidence to justify stopping a system or changing its risk. The project's tests did not validate those uses, and Prismatic Algos removed the proposed ACI sizing multiplier on Oct. 28. The public methodology records the test questions, results and limitations so other researchers can inspect the evidence before designing their own controls. The ACI dashboard, first made public Nov. 6, uses 12 frozen textbook reference recipes across 17 symbols and three mechanism families: trend/breakout, mean-reversion and session-clock. Probe readings include stated trading costs and random-entry comparisons. The probes are hypothetical; separate panels report observed account data and external hypothetical model-account records. These sources are labeled separately. The Nov. 12 announcement brings together the dashboard, methodology and two educational guides on using ACI in an EA review and distinguishing a benchmark from a trading signal. Traders can inspect each card's data date and sample size, review the published validation record and download the public history. Recent external-panel observations may be revised as records arrive. The free index and methodology are available at https://thecryptocrat.github.io/algo-consistency-index/ and https://thecryptocrat.github.io/algo-consistency-index/methodology.html . The announcement and guides are linked from https://prismaticalgos.com/press/algo-consistency-index . ACI is descriptive research. Its current readings have not been shown to predict future strategy returns or improve sizing and trading-halt decisions. The findings report the project's own research and do not constitute independent validation or investment advice. Prismatic Algos publishes algorithmic-trading research and develops algorithmic portfolios. Its research distinguishes historical simulations, account observations and unvalidated hypotheses. Tags: algorithmic trading, trading research, Expert Advisors, risk management, backtesting, fintech --- Source: MarketScale, https://www.marketscale.com/industries/software-and-technology/prismatic-algos-reports-trading-filter-results-across-150-live-accounts. Published for AI indexing and citation; cite the canonical URL. Site guide for agents: https://www.marketscale.com/llms.txt