Somewhere in a data centre in Frankfurt or Amsterdam, a monitoring dashboard is refreshing every hour with a single line of numbers: total bytes transferred, Iran, last four weeks. Most days the line is boring. On some days it is the difference between a Polymarket contract paying out and expiring worthless.
That dashboard is Cloudflare Radar, and it is the sole arbiter of the total internet blackout in Iran market on Polymarket. The contract does not care about protests, cabinet statements, or reports from digital-rights groups. It cares about one chart, one threshold, and six consecutive hours. Everything else is atmosphere.
Deposit $20, get $50 to trade
Welcome bonus on the world's largest prediction market.
18+ · New users only · Availability and terms vary by region · Trading involves risk of loss · BeGambleAware.org
What the contract is really asking
Strip away the headline and the market poses a technical question dressed as a political one. Has Iran's total bytes-transferred figure, as recorded on Cloudflare Radar's traffic-trends chart, fallen to 1.0% or less of the previous four weeks' maximum, and stayed there for at least six consecutive hourly data points, by the specified deadline?
That is it. No qualitative judgement about whether Iranians can reach a VPN. No exception for partial regional shutdowns. No credit for a dramatic degradation that hovers at, say, 3% of peak. The rung either trips or it does not.
The deadlines are laddered. July 31 was the first rung and has already resolved No; that outcome is settled fact, off the board, no longer part of the live question. What remains are three deadlines: August 31, September 30, and December 31. Each is its own binary bet on the same measurement, just with more calendar to work with.
As of the 12 August snapshot, the market's implied lean is clear. The December 31 leg trades around 40%, September 30 around 19%, August 31 around 7%. In plain terms, traders price the December rung as the most plausible of the three but still see it as more likely than not to fail. The near-term August window is priced as a long shot; the year-end window sits closer to a coin flip without quite getting there. Those figures are a snapshot; they will look different by the time you read this. The durable claim is the shape: nearer deadlines cheaper, further deadlines meaningfully more expensive, none of them favoured to resolve Yes.
Why the Cloudflare anchor matters
There is a reason Polymarket wrote the rule this way, and it says something worth pausing on about how prediction markets handle contested political facts. A market that resolves on "was there a blackout" invites endless argument. A market that resolves on "did this specific chart cross this specific line for this specific number of hours" invites almost none.
That is the appeal of an oracle-friendly criterion. Cloudflare Radar publishes hourly bytes-transferred figures publicly, the four-week window is a rolling calculation, the 1.0% threshold is unambiguous, and the six-hour minimum rules out momentary blips. Someone flicking a router in Tehran does not resolve this contract. Neither does a two-hour outage during a protest. The bar is a nationwide, sustained, near-total floor in traffic, and the evidence is a public URL.
Worth understanding: this is a good example of the design choices behind how prediction market resolution actually works. The messier the underlying real-world question, the more the contract writers lean on a single, auditable, public data source. It is the same instinct that pushes contracts to reference official government press releases, ETF approval documents, or exchange filings.
What the July rung's expiry tells you
The fact that July 31 resolved No is data in its own right, not just an eliminated leg. It tells you that across the whole of July, Cloudflare's Iran chart did not spend six consecutive hours below 1% of peak. Not during any protest, not during any state security operation, not during any deliberate throttling event that might otherwise have counted.
That is context, not prophecy. The absence of a July blackout does not make an August or December one less likely in some mechanical sense; the Iranian authorities' willingness to use total shutdowns is a political decision, not a mean-reverting process. But it does anchor the market's current lean. Traders are pricing three remaining rungs having just watched the first one fail, and the shape of the ladder (nearer legs cheap, further legs less cheap) reflects that.
The sensitivity here is worth flagging. On a ladder with legs this far apart in price, a single confirmed six-hour dip below the threshold on a given date collapses the whole structure. There is no partial credit. That is why these markets tend to look calm for weeks and then reprice violently when a specific event lands. Understanding how prediction market odds move as new information arrives is most of the game.
What to actually watch
If you are following this contract as a reader rather than a trader, the useful discipline is to bookmark Cloudflare Radar's Iran page directly and check the traffic-trends chart yourself. The market resolves on that chart. Everything downstream, including this article, is commentary on numbers you can see with one click.
The less useful discipline is to trade the noise. Reports of connectivity problems in specific Iranian cities, statements from ministers about "security measures", social media posts claiming a blackout has started: none of these resolve the contract. Only the chart does. Contracts anchored to a single public data source reward patience and punish reflex.
iPredicta tracks markets like this precisely because they are the cleanest example of what prediction markets do well: turn a fuzzy political question into a testable measurement with a specific settlement rule. The Iran blackout ladder is on the watch list for exactly that reason. It will not tell you what Iranian authorities are thinking. It will tell you, hour by hour, what one dashboard is measuring, and that is the only question it was ever asking.
Frequently asked questions
What counts as a total internet blackout for this market?
The contract resolves Yes only if Cloudflare Radar's total bytes-transferred figure for Iran falls to 1.0% or less of the previous four weeks' maximum, and stays there for at least six consecutive hourly data points, by the deadline. Partial shutdowns, regional throttling, or brief dips that do not last six hours do not count. The reference chart is the traffic-trends bytes-transferred view on Cloudflare Radar's Iran page.
What does the July 31 leg resolving No mean for the remaining rungs?
It means that during July, Iran's Cloudflare Radar bytes-transferred figure did not spend six consecutive hours below the 1% threshold. That is settled fact, so the July rung is off the board. The August, September, and December deadlines are independent binary bets on the same measurement, each with more calendar to work with; the July result does not mechanically raise or lower their probability, but it does set the current baseline against which traders are pricing the remaining ladder.
