← All posts
·GADUIN Researchflight-delay-event-contractsimplied-probabilityevent-contract-tradingBTS-airline-datatrading-edgeGADUIN

Finding Mispriced Flight Delay Contracts on GADUIN

Identify underpriced flight delay event contracts using BTS data, implied probability math, and a 3-signal framework on GADUIN.

Last updated

Flight delay event contracts aren’t always priced at the real chance of a late arrival. When a contract’s implied probability is far from what public on-time data shows, you may have a trading edge. Below: where to get delay-rate data, how to compare contract prices with it, which route patterns produce repeat gaps on GADUIN, and when the gaps are largest.

What “Mispricing” Means in a Flight Delay Event Contract

A contract is mispriced when the chance the market gives an outcome is far from the chance the public record supports. On GADUIN there’s no math involved. A share pays $1 if its outcome happens and $0 if it doesn’t, so a share at 30¢ means the market gives that outcome a 30% chance. If on-time data for that route, airline and month shows a 42% delay rate, Delayed looks cheap compared with public information.

Two caveats first. A mispricing doesn’t guarantee a payout. Single-flight outcomes swing a lot, and you can lose on any one flight however strong the historical data looks. Also, the price may reflect information you can’t see yet, such as a maintenance advisory, a schedule change or a crew positioning problem. A good method allows for this and doesn’t treat every price gap as a confirmed edge.

For background on how a price maps to a chance, see reading an event contract price as a probability.

Where to Find Reliable Delay Rate Data

An edge starts with better data than the market has already priced in. Two public databases cover U.S. and European routes.

BTS Airline On-Time Statistics (transtats.bts.gov) has on-time performance by carrier, route pair and calendar month for U.S. domestic flights and some international departures. The U.S. Department of Transportation counts a flight as late if it arrives 15 minutes or more after schedule. GADUIN flight markets count an arrival more than 15 minutes after schedule as Delayed, so the two definitions are close, but always check the threshold shown on the market before you use BTS rates as your base figure. A rate built on a different threshold can be far off on the same route.

European air-navigation performance statistics break down European route delay by cause, including ATFM (Air Traffic Flow Management) delay codes. The region’s network managers publish them openly. They’re especially useful for slot-constrained hubs like LHR, CDG, AMS and FRA, where ground delays are well documented and a summer afternoon’s capacity limits show up as hard numbers.

Only draw base-rate conclusions from route-and-month combinations with at least 50 flights. Smaller samples are less certain, and that uncertainty should go into your estimate of how likely you are to lose, without inflating your estimated edge.

Calculating Implied Probability from a Contract Price

You don’t need to convert anything. A share pays $1 if its outcome happens, so the price in cents is the chance in percent. 28¢ is a 28% chance. Any other calculation adds a step that can go wrong.

The price isn’t a GADUIN forecast. It’s where trading has left it, and it keeps moving as people trade. Buying an outcome raises its price, and buying another outcome lowers it. In a market few people watch, a small amount of trading moves the price a lot. That’s why gaps open on quiet routes and close once others act on the same public signals.

An illustrative example: a market on an LGA-to-ORD evening departure in January prices Delayed at 28¢, a 28% chance. BTS data for that route, departure window and month shows a 41% historical delay rate. That’s a raw gap of 13 percentage points.

The gap has to survive your order size. GADUIN adds no commission to trades, so no trading fee eats into it. But a large order fills at a rising average price, so the more you buy, the worse your average entry and the smaller the gap. Small trades barely move it. Work out the position at the average price you expect to pay, and ignore the first price you saw.

Figures in this example are illustrative only. Actual prices vary by route and season.

For more on what a GADUIN price represents and what moves it, see how a GADUIN market price behaves.

Route and Airline Patterns That Create Systematic Mispricings

Mispricings cluster around a few route features that push delay rates away from carrier averages. Public data shows them, but prices often haven’t caught up.

Thin-data routes are the most consistent source. Regional and commuter routes have fewer tracked flights. Traders tend to use broad carrier averages instead of route-specific rates, and the two can be very different.

Cascade routes are less obvious but often offer bigger gaps. When severe weather hits a major hub, attention goes to that hub. Feeder routes from smaller cities into it, like GRR-to-ORD or BMI-to-ORD during Chicago weather, are often underpriced because fewer traders follow them. Delays spread a long way down the network, and prices don’t always follow as quickly.

Time of day is consistently underpriced. The first morning departures carry little built-up delay. The last evening departures absorb a full day of it. Prices often reflect the carrier’s overall on-time average instead of the time-of-day split, which leaves a gap at both ends of the schedule.

Seasonal patterns come back every year. Winter at ORD, EWR and BOS and summer thunderstorm season at ATL, MIA and DFW bring high-delay periods that annual averages hide. Route-and-month combinations in these seasons are more likely to have a base rate well above the annual average.

For why delays cluster at congested hubs, see Airport Slot Controls & Congestion. For how delays spread down the network, see Cascade Flight Delays Explained.

A 3-Signal Framework for Identifying Entry Opportunities

A scored, repeatable checklist cuts down on entries driven by confirmation bias. It also builds your own record, which gets more useful as positions settle.

Signal 1: Base rate gap. Read the market’s chance off the price. Pull the historical record for the exact route, month and departure window. The gap (historical rate minus market chance) must be larger than your minimum. 10 percentage points is a reasonable starting point, but the right minimum depends on how thin the market is and how much of the gap your order will use up as it fills.

Signal 2: A near-term catalyst. A high historical rate without a current signal leaves more doubt about whether it will repeat. Look for NWS Watches or Warnings, FAA operational advisories or airline ground stops that affect the route. If the market hasn’t priced in a catalyst yet, there may be a lag: it knows the base rate but hasn’t reacted to the near-term signal.

Signal 3: The timing window is open. Prices tend to move toward the true rate in the 24–48 hours before departure as more people act on available information. The biggest mispricings come before that, when public data has updated and the price hasn’t.

Composite scoring (illustrative): score each signal from 0 to 2. Only enter when the total is 4 or more out of 6. Log the score with every entry. After 20 or more positions, you’ll see which signals best predict correct outcomes.

For the expected-value math behind threshold-based entry, see Expected Value (EV) for Event Contract Traders.

Position Sizing for Identified Mispricings

Even a real mispricing can lose on any single contract. Position size decides how much of your edge turns into results over many positions, and how much one bad outcome sets you back.

A Kelly-style approach sizes each position in proportion to your estimated edge divided by the implied chance of losing. Full Kelly is usually too aggressive for single flights, which swing a lot. Half-Kelly or quarter-Kelly is a better starting point. You give up some theoretical growth for much smaller drawdowns.

Two rules apply whatever your estimated edge:

  • Cap each contract at a set share of your total capital. A ceiling of 2–5% per position stops one bad result from causing an outsized drawdown.
  • Contracts on different flights through the same hub on the same day are correlated. A weather event at ORD delays all the feeders at once. Size the whole group as one position.

GADUIN charges no commission on trades, so nothing comes off the position itself. Moving USDT in and out does cost money. A deposit pays the Ethereum network fee, which is a flat amount and so hits small positions hardest. A withdrawal carries a GADUIN commission, shown on the withdraw form before you confirm. Include both in your minimum-edge calculation, and don’t run a strategy whose edge is smaller than the cost of funding it.

For the full sizing math, see Kelly Criterion for Prediction Markets: Position Sizing.

When Information Asymmetry Is Greatest: Timing Entry

The biggest mispricings show up when new public data is out and the price hasn’t adjusted yet. If you know where that lag usually appears, you can time entries more consistently.

After the National Weather Service issues a Watch or Warning for a relevant corridor, there’s often a window before the FAA publishes a Ground Delay Program (GDP) advisory. The storm is public, but the GDP isn’t out yet. Once it is, many more traders arrive at once and the price moves fast. That window often has the biggest gap between the public signal and the contract price.

Overnight weather model runs (GFS and ECMWF, issued around 00:00 UTC and 12:00 UTC) can change route-level delay chances before traders in U.S. time zones react. Traders in Europe who watch the overnight runs may see the signal before U.S. activity resumes and moves the price.

Before you enter, look at the price history for the past 6–12 hours. A big move means the window has probably closed and the edge is partly or fully priced in. A flat price after a major model update or NWS advisory suggests the window may still be open.

For seasonal timing around winter storms, see Winter Storm Flight Delay Event Contract Strategy. For live GDP advisories, see FAA National Airspace System Status.

Pre-Entry Validation Checklist

Run these checks before each entry so your edge calculation uses matching data. A mismatch in the settlement rule can turn a correct read of the market into a loss.

  1. Settlement criterion match. Check the threshold on the GADUIN market (flight markets count more than 15 minutes late as Delayed) and filter your historical data to the same threshold. Data built on a different delay definition gives a different rate and shouldn’t be used in its place.

  2. Settlement record check. Read what the market settles on. A flight market settles on the destination airport operator’s published arrival time against its published schedule, and the rule is worded the same before and after you trade. Knowing which record decides the outcome avoids surprises at settlement.

  3. Clean base rates. If your sample is big enough, leave out the COVID years (2020–2021). Operations were unusual then, and those years add uncertainty to any base rate.

  4. Carrier-route sanity check. If the market’s chance is far from the carrier’s overall on-time average, find the reason first: a late-day slot, a feeder route prone to cascades, or a market almost nobody trades. A gap you can’t explain is a reason to dig further. Don’t treat it as an automatic entry.

  5. Read the market’s own terms. The threshold, the scheduled time and the destination whose record counts are set when the market opens and don’t change. Cancelled is a separate outcome, so a Delayed position pays $0 if the flight is cancelled. Check the terms before you trade.

For how GADUIN verifies outcomes, see How GADUIN Settles Flight Delay Contracts in USDT.

Executing on GADUIN and Tracking Results

Analysis only pays off with clean execution and a log that lets you check your method over time.

After the 3-signal check and validation, buy at the current price. At entry, record the price and the chance it implies, your historical rate, the signal scores and the position size as a share of your balance. At settlement, record the outcome next to your pre-entry edge estimate. After 20 or more positions, patterns appear. You might find you overestimate your edge on thin-data routes, or spot weather windows correctly but enter after the price has already started moving.

A GADUIN market settles once the destination airport publishes the arrival, and each share of the outcome that happened pays $1 into your balance. Plan for this: a position ties up money until the market settles. If the record is unclear, the market pauses for review. How long that takes depends on the evidence, and there’s no guaranteed deadline.

The log has a second use. It’s the only reliable way to tell whether your data-based approach has a real edge on your routes or is just noise. A small sample isn’t enough. Wait for 30–50 settled positions before drawing conclusions.

For full settlement details and USDT timing, see How GADUIN Settles Flight Delay Contracts in USDT.


Event contracts involve risk of loss. Past historical delay rates do not guarantee future contract settlement outcomes. GADUIN event contracts are not available to U.S. persons. This article is for informational purposes only and does not constitute financial advice.

Look up flight status by flight number. The methodology explains scheduled, estimated and actual times.