How MELANY is tested

Most research tools ask you to trust their ratings. Market Eyes Live prefers that you check them. The MELANY engine is validated three ways: its rules are stress-tested against 19 years of U.S. market history (2007 to 2026), its testing program publishes the ideas it rejected, and since August 6, 2026 every rating it publishes is captured and graded daily against what the market actually does, on a record that is public from day one.

The complete statistical write-up behind this page, with full tables, confidence intervals, and limitations, is published as Working Paper No. 01: Validating a Systematic Equity Rating Engine.

1. Stress-tested against 19 years of market history

MELANY's risk and portfolio rules are validated on historical daily data reaching back to 2007, a window that spans the 2008 financial crisis, the COVID crash of 2020, and the 2022 bear market. The deepest test bed holds 541 large U.S. names with daily bars since 2007, sampled across roughly 714,000 reference days. A broad companion bed adds 4,390 names including small caps.

Survivorship bias, the classic way backtests flatter themselves, is addressed directly: portfolio construction was tested point-in-time on an eligibility universe that includes 1,789 delisted companies, so the names that went to zero are in the test, not quietly missing from it. Where a data set is survivor-only, our findings say so instead of hiding it.

What this does and does not claim: it is the engine's rules (risk bands, entry behavior, portfolio construction, drawdown control) that carry 19 years of testing. The published ratings themselves are younger, and they are held to a harsher standard: the live graded record below.

2. The ideas our own testing killed

A validation program that only ever confirms its own ideas is a marketing department. Ours rejects things, and we think the rejections are worth publishing:

  • "Wait for the pullback" was reversed. A backtest of 12,805 events found no evidence that waiting for a moving-average pullback improved results under any statistical weighting, and the advised limit orders only ever filled about 37% of the time while the names that never pulled back kept running. The live advice was reversed the same day.
  • Sector-rotation and seasonality entry gates were rejected. Plausible-sounding defensive filters failed holdout validation, and several signals pointed the wrong way entirely. Nothing shipped.
  • A downgrade-on-dip hypothesis was ruled inconclusive. In-sample support did not hold up in the holdout period, so nothing was deployed and the finding was recorded as unresolved.

Every rejection above exists internally as a committed script, a raw results file, and a findings document, so each can be re-run and re-checked at any time.

3. Forward-tested daily: the live graded record

Since August 6, 2026, every rating MELANY publishes is captured at the moment of publication and graded automatically against subsequent market moves at fixed horizons. Captures are immutable and grading is mechanical, so the record cannot be edited, trimmed, or cherry-picked after the fact. It went public before any results existed, which is the point: a record you start publishing only after it looks good is not a record.

50.9%

of favorable calls (Favorable and Highest Conviction together) beat more than half of the other core stocks rated the same day, over the following month.

3,288 graded calls. 50% is the no skill line. Read 2026-11-11 from the public feed, captures through 2026-11-10.

27,583
ratings captured
1,660
tickers covered
68
capture days
57,392
graded outcomes

Record live since 2026-08-06. Read 2026-11-11 from the public feed, captures through 2026-11-10. Raw feed: marketeyeslive.com/api/validation-status.

The headline benchmark: the same day list

A rating claims that a stock is a stronger choice right now than the other names rated that day. So the figures on this page grade each call against the other core stocks rated the same day, over the same window: a call counts as a beat when its return over the window is higher than the returns of more than half of those stocks. Grading against the same day list cancels the market's direction, which a comparison with the market as a whole does not; beating the market is the job of a portfolio, not of a single rating. Fifty percent is the no skill line. The public feed also grades every call against a market benchmark, funds included, and publishes those figures beside these. The figures are plain hit rates with the number of calls behind each one; same day calls move together, so read a figure a point or two from the line as a figure, not a verdict.

The record by tier at a day, a week and a month

TierWindowCallsBeat the same day listBeat same industry peers
Favorable + Highest Convictionnext day7,06649.8%49.5%
following week6,46049.8%48.9%
following month3,28850.9%46.8%
Holdnext day8,85950.4%48.9%
following week7,98550.8%48.9%
following month4,91650.1%48.9%
Very Promisingnext day1,38047.3%46.6%
following week1,22046.3%47.4%
following month72748.7%49.1%
Promisingnext day3,50842.1%41.6%
following week3,05935.3%34.4%
following month1,66131.0%29.1%
Runner!next day52850.6%53.6%
following week47251.9%54.7%
following month31850.0%57.6%
Catalyst Watchnext day3652.8%50.0%
following week2528.0%40.0%
Constructive (REIT label)next day2817.9%16.0%
following week2842.9%36.0%
following month2650.0%11.5%
Steady (REIT label)next day26649.3%47.7%
following week22652.7%50.9%
following month15122.5%25.8%
Neutral (REIT label)next day41950.6%53.2%
following week38150.4%58.3%
following month18641.9%53.8%
Lagging (REIT label)next day10756.1%61.7%
following week10464.4%74.0%
following month5770.2%80.7%

As of the 2026-11-11 read, captures through 2026-11-10. A cell is published only once it holds 20 graded outcomes across at least 8 distinct tickers, so a dramatic figure can never rest on a handful of names; a word with no cell above that floor is not shown. Green is above the 50% line, red is below it; a three month window is published in the raw feed once its cells clear the same floors. REITs are recorded under the app's own labels and are listed where they have a same day grade; funds (ETFs) are not graded against the same day list, only against a market benchmark, and those figures are in the raw feed. Promising is the early rung of the separate speculative model, for companies the main model cannot anchor (no or negative trailing earnings, a price far ahead of trailing or expected earnings, too little company data, or a strong sector theme while the main score is under 60), and is printed as it is.

One company, two products

The Market Eyes Live app (iOS and web) and the free public Market Eyes Live MCP server for AI assistants are the same company and the same MELANY engine. The MCP serves the free public data tier; the app carries the full research depth (entry zones, alert levels, portfolio tools, Lock Radar).

Try Market Eyes Live free Read the raw record feed

The 19-year stress test measures the engine's rules against real historical market data, including companies that were later delisted, so it is not curve-fit to the survivors. Backtested figures are computed with the benefit of hindsight and do not reflect trading in live accounts, so they are not a promise of future results. The live graded record is different: it is captured in real time, out of sample, and published before the results exist, so it is a genuine forward record, not a simulation, and it measures the engine's published calls rather than any individual's returns. All figures on this page trace to committed test scripts and results files. Market Eyes Live provides algorithmic research and education, not individualized financial advice. Markets involve risk, including loss of principal. Past performance does not guarantee future results.

Last updated 2026-11-11 · What is Market Eyes Live? · All rated stocks · Mortgage rates · Privacy · Terms

Market Eyes Live

Every stock rated. Every published rating graded.

One stop research for stocks and mortgage rates. MELANY scores every U.S. stock and ETF 0 to 100 across eight sourced factors, PRISM ranks conviction and APEX writes the plan with one entry zone. Every published rating is graded in public. Built by MEL Technologies, the verification layer for AI trading. Free to start on iOS and the web.

Get the free iOS app Create a free account on the web Free account, no card. Pro: 30 days free, then $9.99 a month.