The Mensor's price composites are depth-aware: a contributing market's influence is derived from persistent two-sided order-book depth near its midpoint, after applicable fees. The other instruments use declared transforms suited to what they measure: an expectation for the rate ladder, a mixed-input macro composite, and volume share for attention.
Depth is closer to current executability than trailing volume, but it is not committed capital: displayed orders can be cancelled, fragmented, or manipulated. The band, persistence rule, concentration, and achieved venue coverage therefore travel with the figure.
Where a question is priced on more than one venue, contributing markets are mapped only when their resolution criteria are substantively identical. Mappings are published alongside each index and revised as criteria change.
Two markets that appear to price the same event often resolve on measurably different criteria. We treat resolution criteria as primary data and document differences before any aggregation.
Selection is versioned separately from aggregation. The implemented rule is MIS-SELECT-3.0, proposed September 29, 2026 and awaiting formal approval. A constituent is a market that is deep enough to mean something, tradable enough to be a price, and relevant enough to belong. Each of the three is measured, scored, and disclosed. Floors alone select for whatever happens to be liquid; a scored model selects for what the index is about.
Nine eligibility gates come first, hard and in order, each failure logged with its code. Category: inside the family's lexicon and outside the exclusion list (sports and props, novelty, tweet counts, celebrity behaviour, single-person trivia), the only place editorial judgement enters. Contract type: binary yes/no, with ladders entering only through a declared transform. Resolution rules present and an end date parseable. Tenor: between 14 and 400 days to resolution at entry; a member exits at 7. Listed on the venue at this run's fetch. A fee-adjusted mid between 3¢ and 97¢, since a pinned contract carries no information. A bid and an ask both present. A risk direction assignable by lexicon, by model classifier at 0.8 confidence or above, or by founder decision. And for a cross-venue merge, resolution criteria that are exact or compatible; divergent criteria stay separate events.
What clears the gates is scored on three dimensions, each a percentile within the family universe. Depth, weight 0.40: two-sided book depth within 2¢ of the mid, the same within 5¢, and the seven-day median of the first, so a single deep print does not qualify a market. Depth is read from the order book, never inferred from volume. Tradability, weight 0.25: the fee-adjusted spread, the size fillable within 2¢on each side, the share of the week's readings with both sides quoted, the venue's taker fee, and the shape of the tenor; a spread wider than 6¢ fails outright. Relevancy, weight 0.35: subject fit against the lexicon, the week's coincident coverage, the thirty-day price range and variance, open interest and volume, and a second venue. A market can be deep and tradable and still not belong; relevancy is what keeps a basket about its subject.
The composite is 0.40 depth plus 0.25 tradability plus 0.35 relevancy, computed hourly. A market enters at 0.40 and exits below 0.30, an asymmetric buffer so a name near the line does not flap; resolved, delisted, and gate failures exit at once. One canonical event per subject and outcome; other tenors of the same question are its legs, and a resolving tenor proposes its successor as a roll.
Every basket has a minimum: 15 constituents for a flagship, 5 for a corridor. When fewer clear the entry score at the standard depth tier of $5,000, the scan steps down through $2,500, $1,000, and $500, and the entry score to 0.30, only as far as needed. The hard floor is $250 of band depth and a score of 0.25; below that the basket publishes short and is labelled thin. Every constituent carries its admission tier, and the ledger shows the tier and floor the scan settled on.
Membership is decided by score; weight is decided by depth, through the unchanged capped waterfall. The two are kept separate so that widening a basket never lets a thin contract move the figure: a $500 leg beside a $60,000 leg carries under 1% of the reading.
Every index page carries, beneath its constituents, the selection ledger from the latest scan: universe, eligible, scored, and admitted; the effective tier and entry score; the related markets the scan is holding, each with its 24-hour volume where the scan publishes one, its gate code or score, and its depth; and every exit in the last thirty days with its reason. Each constituent shows its depth, tradability, relevancy, and composite scores. A reader can see that a corridor is thin because the market is thin, not because the scan is blind.
One sheet per published index: what it measures, its scale, how it is constructed, and what its coverage does and does not yet include. The methodology version on each sheet is the one stamped on every published tick, so a reading is always traceable to the rules that produced it.
A composite index of the macro stress prediction markets and selected fiat anchors are pricing, recession risk, the rate path, inflation pressure, growth, and financial conditions, on a base-100-neutral scale.
A weighted composite of seven constituents. Where a prediction market is deep enough it supplies the value (recession risk, the rate path) and the matching official series serves as its anchor; where it is not (inflation, growth, the financial-conditions cluster of the 10-year yield, the broad dollar, and WTI crude), the official series enters as a minority constituent. Aggregate non-market weight is capped at 25%, so the market view always carries the majority. Each value is normalized against a documented neutral and composed through the shared engine: component caps, an equal-weight floor, and a short smoothing window. Composition changes are logged and the seam is stamped.
Scale: base 100 = neutral · higher = more macro stress priced (recession risk, inflation pressure, tighter financial conditions) · bands: Calm / Easing / Neutral / Strained / Stressed · version MIS-MACRO-1.1.0
Coverage confidence: low. Prediction-market majority: 75%; fiat anchors capped at 25%; 9/9 live constituents.
A depth-weighted measure of the geopolitical risk the prediction markets are pricing: 0-100, higher means more conflict and instability. Reads across venues where resolution criteria and liquidity permit; coverage today is Polymarket-led, disclosed per tick in the coverage badge.
Each constituent is a canonical event with a declared risk direction: an escalation contract feeds its probability into the meter directly, a de-escalation contract (a ceasefire, a deal, a normalization) feeds the inverse, so good news lowers the reading. Constituents are depth-weighted with component caps, and events whose venue listings resolve on different criteria are flagged or excluded rather than blended. Coverage is currently Polymarket-led (the coverage badge on each page states the venue concentration plainly), and Kalshi listings join as their depth and resolution criteria qualify.
Scale: 0–100 · higher = more geopolitical escalation / instability · bands: Calm / Low / Elevated / High / Severe · version MGRI-1.1.0
Coverage confidence: low. Single-venue coverage; cross-venue weight below 20%. 22/22 live constituents.
A market-implied index of AI progress, adoption, commercialization, and optimism, capability milestones, revenue and adoption, against regulation, backlash, and compute bottlenecks. 50 = neutral outlook.
In research: the constituent framework (capability, commercialization, regulation and backlash, infrastructure) is built and tested on fixtures, and the index publishes a live level once real market mappings replace them.
Scale: 0–100 · higher = greater market-implied AI progress / adoption / optimism · bands: Stalling / Skeptical / Mixed / Constructive / Accelerating · version MIS-MAIO-0.1.0
A market-implied meter of priced political disruption, chamber control flips, contested processes, irregular exits, read across venues. Not a partisan thermometer: it measures how much change the markets price, not which side benefits.
Composed through the shared engine: de-vigged midpoints, fee adjustment, depth weighting, component caps. Composition changes are logged and the seam is stamped.
Scale: 0–100 · higher = more change-of-control and contested-process priced · bands: Settled / Steady / Contested / Charged / Upended · version MPDI-1.0.0
Coverage confidence: medium. 1/1 live constituents; cross-venue weight 100%.
Where prediction-market dollars actually sit: each market's 24h volume is partitioned by category, and the published level is the leading category's share. A sociology instrument, it measures what people bet on, not what the prices say.
Each market's 24-hour traded volume is assigned to one category by an ordered, versioned lexicon, and the level is the leading category's share. Volume is the measurement object here, not a proxy for conviction. Incentive-driven or artificial activity still affects that object, so per-venue shares, category breadth, and volume-to-depth ratios are required cross-checks. Version 1.1.0 (October 3, 2026) added a sports title grammar for the per-game market names Kalshi introduced in the third quarter, a named weather category, and a read of the full eight-hour window. The series from September 28, 2026 relabels under it; earlier readings are unchanged and keep their 1.0.0 stamp.
Scale: natural unit · higher = a larger share of prediction-market dollars on the leading category · version MATI-1.1.0
Expected 2026 Fed path derived from prediction-market cut-count ladders, published in its natural unit, expected cuts. Horizon: calendar 2026 cut count; this ladder cannot express a hike. MRPI-2 (rolling eight meetings, signed) is awaiting approval. A Macro subindex and a Macro constituent, not a top-level flagship.
Published in its natural unit: the expected number of 2026 Fed rate cuts implied by cut-count ladder markets, read across venues and depth-weighted where both venues clear a per-leg depth floor. A subindex of the Macro index and one of its constituents.
Scale: natural unit · higher = more expected 2026 Fed cuts (more easing priced) · version MIS-RATEPATH-1.0.0
The Iran corridor of the geopolitical risk meter, strike, escalation, and nuclear-timing markets read across venues.
Each constituent is a canonical event with a declared risk direction: an escalation contract feeds its probability into the meter directly, a de-escalation contract (a ceasefire, a deal, a normalization) feeds the inverse, so good news lowers the reading. Constituents are depth-weighted with component caps, and events whose venue listings resolve on different criteria are flagged or excluded rather than blended. Coverage is currently Polymarket-led (the coverage badge on each page states the venue concentration plainly), and Kalshi listings join as their depth and resolution criteria qualify.
Scale: 0–100 · higher = more geopolitical escalation / instability · bands: Calm / Low / Elevated / High / Severe · version MGRI-1.1.0
Coverage confidence: low. Single-venue coverage; cross-venue weight below 20%. 8/8 live constituents.
The Taiwan corridor, strait, blockade, and cross-strait election markets read across venues.
Each constituent is a canonical event with a declared risk direction: an escalation contract feeds its probability into the meter directly, a de-escalation contract (a ceasefire, a deal, a normalization) feeds the inverse, so good news lowers the reading. Constituents are depth-weighted with component caps, and events whose venue listings resolve on different criteria are flagged or excluded rather than blended. Coverage is currently Polymarket-led (the coverage badge on each page states the venue concentration plainly), and Kalshi listings join as their depth and resolution criteria qualify.
Scale: 0–100 · higher = more geopolitical escalation / instability · bands: Calm / Low / Elevated / High / Severe · version MGRI-1.1.0
Coverage confidence: low. Single-venue coverage; cross-venue weight below 20%. 2/2 live constituents.
The Ukraine corridor, territory, ceasefire-timing, and escalation markets read across venues.
Each constituent is a canonical event with a declared risk direction: an escalation contract feeds its probability into the meter directly, a de-escalation contract (a ceasefire, a deal, a normalization) feeds the inverse, so good news lowers the reading. Constituents are depth-weighted with component caps, and events whose venue listings resolve on different criteria are flagged or excluded rather than blended. Coverage is currently Polymarket-led (the coverage badge on each page states the venue concentration plainly), and Kalshi listings join as their depth and resolution criteria qualify.
Scale: 0–100 · higher = more geopolitical escalation / instability · bands: Calm / Low / Elevated / High / Severe · version MGRI-1.1.0
Coverage confidence: low. Single-venue coverage; cross-venue weight below 20%. 4/4 live constituents.
The Gaza corridor, ceasefire and escalation markets read across venues (Polymarket's book is typically deeper than Kalshi's here).
Each constituent is a canonical event with a declared risk direction: an escalation contract feeds its probability into the meter directly, a de-escalation contract (a ceasefire, a deal, a normalization) feeds the inverse, so good news lowers the reading. Constituents are depth-weighted with component caps, and events whose venue listings resolve on different criteria are flagged or excluded rather than blended. Coverage is currently Polymarket-led (the coverage badge on each page states the venue concentration plainly), and Kalshi listings join as their depth and resolution criteria qualify.
Scale: 0–100 · higher = more geopolitical escalation / instability · bands: Calm / Low / Elevated / High / Severe · version MGRI-1.1.0
Coverage confidence: low. Single-venue coverage; cross-venue weight below 20%. 4/4 live constituents.
A calibration white paper, scoring the cross-venue aggregate against single-venue prices on resolved markets, follows once enough mapped markets have resolved to make the comparison meaningful. The harness is built; the evidence accrues.
Read the working paper: Reading across the market. It separates the proposed standard, the current implementation, and the validation still required before any figure can be treated as a professional benchmark.
We can share the methodology with you over email. Leave your address below.
Your email