Mispricing Scanner
A Bayesian engine scans prediction markets daily for mispricings. Agent A finds the same event on the other platform (LLM-validated) and compares the price — when a validated twin exists. Agent B is Claude NLP estimating probability from the question. When the engine diverges from the market by 5pp+, the contract is flagged with Kelly sizing; 3–5pp near-misses go to a free watchlist band.
Runs every morning; results post by ~10 AM ET. How it works →
Quick Answer
the Mispricing Scanner identifies Kalshi binary markets mispriced by 5 percentage points or more against fair value. A Claude NLP probability estimate scores every candidate market, cross-checked against Polymarket whenever a validated same-event twin exists, and every flag carries Kelly sizing. Near-misses of 3–5pp are published free as a watchlist band. Treat a flag as a shortlist to investigate, not an automatic trade: confirm the resolution terms and book depth before taking a position.
If you're asking whether Kalshi/Polymarket mispricing flags actually pay out: the scanner flags markets where its Bayesian engine diverges from the platform price by 5pp or more. HIGH CONFIDENCE requires 8pp+ and a validated same-event twin on the other platform pulling the same direction — most contracts have no such twin, so many days produce no HIGH flags at all. That scarcity is the design. Everything else lands at MEDIUM or LOW, with a free 3–5pp watchlist band of forming edges below the flag bar. The Kelly fraction shown is quarter-Kelly, sized on a cost-adjusted entry price, not the headline edge.
Caveat: a flag is a signal to investigate, not a guaranteed trade. The engine now validates cross-platform matches with an LLM same-event gate, so the old “same name, different event” pairs are filtered out — but still read both contract pages and confirm the resolution criteria before sizing.
What Is the Mispricing Scanner?
The Mispricing Scanner is a daily automated tool that uses a Bayesian engine to find prediction markets where the listed price is significantly wrong. It scans both Kalshi (the only CFTC-regulated prediction market in the US) and Polymarket. When the engine reads a market as mispriced by 5 percentage points or more, it surfaces with a Kelly-sized position recommendation; 3–5pp near-misses are published free as a watchlist band.
Kalshi is the primary scan — because it's where US traders can actually act on these signals. Polymarket provides the broadest global liquidity. The scanner exploits gaps on both platforms.
How It Works
For each platform: Agent A looks for the same event on the other platform — a title pre-filter, then an LLM gate confirming both contracts resolve on identical criteria before the prices are compared (no validated twin → no cross-reference). Agent B asks Claude for an independent probability estimate. Both feed into a Bayesian weighted average. Divergences of 5pp+ are flagged and 3–5pp near-misses go to the free watchlist band; HIGH CONFIDENCE needs 8pp+ and a validated twin agreeing, which is why it is rare.
When to Use This Tool
Check the scanner in the morning once the day's scan has posted — usually by 10 AM ET. HIGH CONFIDENCE flags are the priority when there are any; otherwise work the widest divergences down. Verify the cross-platform match is the same event, check for overnight news that could explain the gap, then size with the quarter-Kelly fraction shown. This is a signal to investigate, not a guaranteed trade.
MARKET INTELLIGENCE
Find markets where the crowd got it wrong
What is this?
The crowd is often right. But not always. This scanner identifies prediction market contracts where the price seems significantly out of line with real-world data — news, historical base rates, or statistical models.
Think of it as a lie detector for market prices. When a market is way overpriced or underpriced relative to reality, this flags it. You still have to decide — but this gives you the shortlist of where to look.
Real-World Example
→ The Flag
Historical data shows government shutdowns resolve within 14 days 84% of the time. A market asks "Shutdown lasts more than 21 days" — priced at 55¢ (the crowd thinks 55% likely). That's a mispricing signal.
The scanner flags the gap: historical base rate says this should be worth about 16¢, not 55¢. You buy NO at 45¢ (inverse of YES at 55¢) and wait.
✅Action: Use this as your daily deal-finder. Look for anything flagged more than 15 points from the estimated fair value.
Bottom line: The crowd panics, overreacts, and misprices markets every day. This helps you catch it.
Full guide →Loading today's scan
Checking Bayesian engine results...
The retuned model is being monitored to confirm it's producing a documented, playable edge over a large enough sample. Every pick is logged and auto-graded.
Read the fine print
Half the edge in prediction markets is knowing exactly how a contract resolves. We read the official resolution rules of the markets traders care about and flag the traps — which data print counts, the exact metric, rounding, and settlement timing.
Today's trapEthereum price at Aug 4, 2026 at 9am EDT?Resolves YES if the simple average of the 60 one-second ERTI prints in the minute immediately before 9:00 AM EDT on Aug 4, 2026 is strictly above $2,534.99.3 traps
- CF Benchmarks ERTI, not spot exchange price — ERTI is a specific real-time index calculated by CF Benchmarks from a defined basket of exchanges — it can diverge meaningfully from prices shown on any single exchange or aggregator a trader might be watching.
- 60-second average, not the 9 AM print — The contract uses the simple average of the 60 seconds *before* 9 AM EDT, so a spike or dip right at 9:00:00 is irrelevant; only the 8:58:60–8:59:59 window (the minute preceding 9 AM) feeds the settlement calculation.
- Strictly above $2,534.99 threshold — An average that lands exactly at $2,534.99 resolves NO — the price must exceed that level, not merely reach it, so positions near the boundary are exposed to precise decimal-level risk.
Ethereum price at Aug 4, 2026 at 9am EDT?Resolves YES if the simple average of the 60 one-second ERTI prints in the minute immediately before 9:00 AM EDT on August 4, 2026 is strictly above $2,539.99.4 traps
- CF Benchmarks ERTI, not spot price — Resolution uses CF Benchmarks' Ethereum Real-Time Index (ERTI), which may diverge from any single exchange's spot price or other index (e.g., CME CF Ether Reference Rate), so traders tracking a different feed could be misled.
- 60-second average, not 9am print — The contract averages the sixty one-second index values in the window *before* 9:00 AM EDT — not the instantaneous price at 9:00 AM — so a sharp spike or drop exactly at 9am does not solely determine resolution.
- Strict 'above' threshold, not 'at or above' — The rule says 'above 2539.99,' meaning an average that lands exactly at $2,539.99 resolves NO, making the precise boundary critical for near-the-money positions.
- EDT timezone, not UTC or EST — The cutoff is 9:00 AM Eastern Daylight Time (UTC−4); traders using UTC or confusing EDT with EST (UTC−5) would reference the wrong 60-second window.
Ethereum price at Aug 4, 2026 at 9am EDT?Resolves YES if the simple average of the 60 one-second ERTI prints in the minute ending at 9:00 AM EDT on August 4, 2026 is strictly above $2,544.99.4 traps
- CF Benchmarks ERTI, not spot price — Resolution uses CF Benchmarks' Ethereum Real-Time Index (ERTI), which may differ from exchange spot prices or other index providers a trader might reference to gauge likelihood.
- 60-second average, not single print — The threshold is applied to a simple average of sixty one-second values in the minute before 9 AM EDT — a momentary spike or dip through $2,544.99 at exactly 9 AM does not determine resolution alone.
- Strict 'above' threshold — $2,544.99 excluded — The contract resolves YES only if the average is strictly above $2,544.99; an average landing exactly at $2,544.99 resolves NO.
- EDT vs. UTC timezone boundary — 9 AM EDT is UTC-4 (13:00 UTC); traders monitoring clocks in UTC or EST (UTC-5) could misidentify the precise settlement window.
Bitcoin price on Aug 4, 2026?Resolves YES if the simple average of the 60 one-second BRTI prints in the 60 seconds before 9:00 AM EDT on Aug 4, 2026 exceeds $71,099.99.4 traps
- Specific index: BRTI, not spot price — Resolution uses CF Benchmarks' Bitcoin Real-Time Index exclusively — not Coinbase, Binance, or any other exchange's spot price, which can diverge meaningfully, especially in volatile conditions.
- 60-second average, not single snapshot — The contract averages 60 individual one-second BRTI prints rather than taking a single price at 9 AM, so a brief spike or dip exactly at 9 AM does not solely determine resolution.
- Strict threshold: must exceed $71,099.99 — The rule is 'above 71099.99,' meaning a final average of exactly $71,099.99 resolves NO — the price must be at least $71,100.00 (to the cent) to resolve YES.
- Timezone is EDT, not UTC or EST — The cutoff is 9 AM Eastern Daylight Time (UTC-4); confusing this with EST (UTC-5) or UTC shifts the effective snapshot window by one hour, which matters for scheduling or hedging around the fixing.
Bitcoin price on Aug 4, 2026?Resolves YES if the simple average of the 60 one-second BRTI prints in the 60 seconds before 9:00 AM EDT on Aug 4, 2026 exceeds $71,199.99.4 traps
- BRTI, not spot exchange price — Resolution uses CF Benchmarks' Bitcoin Real-Time Index specifically — not Coinbase, Binance, or any other exchange's spot price, which can diverge meaningfully around the threshold.
- 60-second average, not single snapshot — The settling price is a 60-second simple mean of per-second BRTI prints ending at 9:00 AM EDT, so a brief spike or dip at 9:00 AM alone does not determine resolution — the full preceding minute matters.
- Strict greater-than threshold ($71,199.99) — The rule says 'above 71199.99,' meaning the average must exceed that figure; an average of exactly $71,199.99 resolves NO, making the precise cent-level boundary critical near the threshold.
- EDT vs. EST timezone shift risk — The cutoff is explicitly 9 AM EDT (UTC-4); if clocks have changed or are misread as EST (UTC-5), traders could misidentify the exact settlement moment by a full hour.
Bitcoin price on Aug 4, 2026?Resolves YES if the simple average of the 60 one-second BRTI prints in the minute immediately before 9:00 AM EDT on Aug 4, 2026 exceeds $71,299.99.4 traps
- Specific index: BRTI, not spot price — Resolution uses CF Benchmarks' Bitcoin Real-Time Index exclusively — not Coinbase, Binance, or any other exchange's spot price, which can diverge meaningfully from BRTI.
- 60-second average, not single snapshot — The contract settles on the simple average of 60 individual second-level prints leading up to 9 AM EDT, so a brief spike or dip right at 9 AM alone won't determine the outcome — the entire preceding minute matters.
- Strict greater-than threshold ($71,299.99) — The rule says 'above 71299.99,' meaning the average must be at least $71,300.00 to resolve YES — an average of exactly $71,299.99 resolves NO.
- EDT timezone boundary (not UTC or EST) — The 9 AM cutoff is Eastern Daylight Time; traders in other timezones or those confusing EDT with EST (a 1-hour difference) could misjudge the exact settlement moment.
Bitcoin price range on Aug 4, 2026?Resolves YES if the simple average of the 60 one-second BRTI prints in the 60 seconds before 9:00 AM EDT on Aug 4, 2026 falls within the range $71,200.00–$71,299.99.4 traps
- Specific data source: CF Benchmarks BRTI only — Only CF Benchmarks' Bitcoin Real-Time Index (BRTI) is used — not Coinbase, Binance, CME settlement, or any other price feed — so a price in-range on other platforms is irrelevant.
- 60-second average, not spot price at 9 AM — Resolution uses the simple average of the 60 seconds *before* 9 AM EDT, not the instantaneous spot price at 9:00:00 AM, so a single in-range print at 9 AM does not guarantee resolution YES.
- Upper bound is $71,299.99, not $71,300 — The range is strictly capped at $71,299.99, meaning an average of $71,300.00 or higher resolves NO — the $0.01 boundary is a hard cutoff.
- Timezone is EDT, not UTC or EST — The 9 AM reference is Eastern Daylight Time (UTC−4); traders in other timezones or tracking UTC-based feeds must account for the correct offset to identify the exact 60-second window.
Bitcoin price range on Aug 4, 2026?Resolves YES if the simple average of the 60 one-second BRTI prints in the 60 seconds immediately before 9:00 AM EDT on August 4, 2026 is strictly below $52,700.4 traps
- Specific data source: CF Benchmarks BRTI only — Only CF Benchmarks' Bitcoin Real-Time Index qualifies — prices from Coinbase, Binance, CMC, or any other index are irrelevant and could show a different value at the same moment.
- 60-second average, not spot price — Resolution uses the simple average of 60 individual one-second BRTI prints (the 60 seconds before 9 AM EDT), so a single-second spike or dip above/below $52,700 does not determine the outcome — the mean of all 60 prints does.
- Strict 'below' threshold, not at-or-below — An average exactly equal to $52,700.00 resolves NO; the price must be strictly less than $52,700, so positions near the threshold carry boundary risk.
- EDT timezone, not UTC or EST — The snapshot is anchored to Eastern Daylight Time (UTC−4); August 4, 2026 falls in daylight-saving time, so confusing EDT with EST (UTC−5) would mean monitoring the wrong 60-second window.
Bitcoin price range on Aug 4, 2026?Resolves YES if the simple average of the 60 one-second BRTI prints in the 60 seconds before 9:00 AM EDT on Aug 4, 2026 exceeds $71,299.99.4 traps
- Specific data source: CF Benchmarks BRTI only — Only CF Benchmarks' Bitcoin Real-Time Index is used — no other exchange price, index, or aggregator (e.g., Coinbase spot, CME reference rate) is relevant, even if they diverge significantly at the moment of resolution.
- 60-second average, not a spot price — Resolution uses the simple average of 60 individual one-second BRTI values in the minute before 9 AM EDT, so a brief spike or dip through $71,300 at exactly 9 AM does not determine the outcome — the full 60-second window matters.
- Strict threshold: must be ABOVE $71,299.99 — The condition is strictly greater than $71,299.99, meaning an average of exactly $71,299.99 resolves NO; the price must print at least $71,300.00 (or one cent above) to resolve YES.
- Timezone: EDT, not UTC or EST — The cutoff is 9 AM Eastern Daylight Time (UTC-4); traders in other timezones or those confusing EDT with EST (UTC-5) could misidentify the exact 60-second window being measured.
Ethereum price at Aug 3, 2026 at 9am EDT?Resolves YES if the simple average of the 60 one-second ERTI prints in the minute before 9:00 AM EDT on Aug 3, 2026 exceeds $2,584.99.3 traps
- CF Benchmarks ERTI, not spot exchange price — ERTI is a specific real-time index from CF Benchmarks aggregating multiple venues; it can diverge meaningfully from prices shown on any single exchange or data feed a trader might reference.
- 60-second average, not the 9am print — Resolution uses the simple average of the 60 seconds *before* 9:00 AM EDT, so a brief spike or dip exactly at 9:00 AM is irrelevant — only the preceding minute's average determines settlement.
- Strict 'above' threshold, not 'at or above' — The price must be *above* $2,584.99, meaning an average landing exactly at $2,584.99 resolves NO, making the penny distinction critical near the boundary.
Related Tools
Frequently Asked Questions
Which Kalshi binary markets are mispriced right now?
PredictionMarketsPicks runs a daily Bayesian scan that flags Kalshi binary markets trading 5 or more percentage points away from fair value, cross-checked against Polymarket and an independent probability model. Each flag carries a tier — HIGH CONFIDENCE (8pp+ divergence with a validated cross-platform twin agreeing), MEDIUM, or LOW — plus a quarter-Kelly position size. Below the flag bar we publish a free 3–5pp watchlist band of forming edges. Read HIGH CONFIDENCE flags first, confirm both contracts resolve on the same criteria, then size with the Kelly fraction shown. The board updates every morning, with results posting by roughly 10 AM ET; treat each flag as a shortlist to investigate, not an automatic trade.
What is the Kalshi mispricing scanner?
The Kalshi mispricing scanner is a daily automated scan that uses a Bayesian engine to flag Kalshi binary markets trading 5 or more percentage points away from fair value (it scans Polymarket the same way). Agent A finds the same event on the other platform — LLM-validated — and compares the price when a validated twin exists; Agent B asks Claude for an independent probability estimate. Every flag carries a quarter-Kelly position size, and 3–5pp near-misses are published free as a watchlist band.
Do Kalshi vs Polymarket mispricing flags actually pay out?
The scanner flags markets where the Bayesian engine (Agent A: cross-platform price match; Agent B: Claude NLP probability estimate) diverges from the platform price by 5pp or more. HIGH CONFIDENCE requires a validated same-event twin on the other platform pulling the same direction, so days without a validated twin produce no HIGH flags — that scarcity is the design, not a malfunction. The rest of the board is MEDIUM or LOW, with a free 3–5pp watchlist band below it. The Kelly fraction shown is quarter-Kelly, sized on a cost-adjusted entry price rather than the headline edge. Caveat: a flag is a signal to investigate, not a guaranteed trade. Agent A validates that any cross-platform match resolves on the same event via an LLM gate, so the old "same name, different event" pairs are filtered out — but still read both contracts' resolution rules yourself before sizing.
How does the scanner engine detect mispricings?
For the Kalshi scan: Agent A looks for the same event on Polymarket — a title pre-filter narrows candidates, then an LLM gate confirms both markets resolve on the same event and criterion before their prices are compared (no validated twin → no cross-reference). Agent B asks Claude for an independent probability estimate. Both signals feed into a Bayesian weighted average. For the Polymarket scan: the agents are reversed — Agent A uses Kalshi as the cross-reference. When the engine disagrees with the platform price by 5pp+, the market is flagged.
When should I act on a HIGH CONFIDENCE mispricing flag?
HIGH CONFIDENCE means the Bayesian engine diverges from the market by 8 or more percentage points AND a validated same-event twin on the other platform pulls the same direction. Before acting, verify that the cross-platform match really is the same event, check for any news that could explain the gap, and use the quarter-Kelly fraction shown to size the position. This is a signal to investigate, not a guaranteed trade.
How many mispricing flags clear HIGH CONFIDENCE each day?
Often none. HIGH CONFIDENCE requires a validated same-event twin on the other platform pulling the same direction as our own probability estimate, and most Kalshi contracts simply have no Polymarket twin that resolves on identical criteria. Days without a validated twin produce no HIGH flags at all — scarcity is the design, not a malfunction. The alternative, promoting flags to HIGH on divergence alone, is what the engine used to do, and it labelled the large majority of flags HIGH while adding nothing to their accuracy.
Is the Mispricing Scanner free to use?
Partly. The 3–5pp watchlist band of forming edges is free to everyone, as is one full "taste" flag each day. The complete flagged board — every 5pp+ contract with its Kelly sizing — is a Pro tool at $14.99 per month, and it surfaces in full during promotional preview windows. The scan runs every morning and stores results in Supabase, so every Pro trader sees the same flagged contracts. The free tools — EV Calculator, Probability Converter, and Combo Edge Builder — cover the core math if you want to vet a single contract before subscribing.
What is the difference between the Kalshi scan and the Polymarket scan?
Both scans flag divergences of 5 percentage points or more using the same Bayesian engine, swapping only which platform anchors the price check. On the Kalshi scan, Agent A uses Polymarket as the anchor; on the Polymarket scan, Agent A uses Kalshi. Agent B always asks Claude for an independent estimate. Kalshi is the primary scan because it is the only CFTC-regulated venue where US traders can act on the signal.