AFOS Analytics
Real-time political pricing
Prediction markets × polls × news
no smoothed averages
United StatesIn the week of August 24 to 28 the Senate's two books swapped roles. Last issue the one that moved was the seat distribution, with the 51-seat bucket giving up 2.50pp while the binary contract stood still. This time it was the reverse: the Senate binary rose 1.00pp on both sides, from 49.50% to 50.50% on the Republican contract and from 49.50% to 50.50% on the Democratic one, and the seat distribution barely moved, with the largest band change at 0.50pp.
The week had a ballot, and it was a Senate one. South Carolina's special Republican runoff was held on August 25, and Darline Graham defeated Ralph Norman for the nomination to the seat left vacant by Lindsey Graham's death in July. The Senate binary moved two days later, at the August 27 close. AFOS records both dates and asserts no link between them.
In the House the Monday close and the Friday close are identical for the second straight week, 88.50% for the Democrats and 11.50% for the Republicans, 0.00pp on both sides. Within the week the contract swung one 1.00pp band and returned to where it started.
What happened after the covered week changes this Monday's picture. The Republican side of the Senate gave back the 1.00pp it had gained and the Democratic side did not, so the Democrats are 1.00pp ahead, at 50.50% against 49.50%. Both binary pairs sum to 100.00% in today's reading, and both passed through 101.00% on the preceding days: the Senate pair at the August 29 close and the House pair at the August 30 close.
With that, the distance between the two prices for the same outcome in the Senate, the binary and the one implied by the seat distribution, stands at 3.13pp in this Monday's confirmed reading, against 5.09pp at Friday's close. The 1.96pp narrowing splits almost evenly: the binary contract gave up 1.00pp, from 50.50% to 49.50%, and the normalized distribution rose 0.96pp, from 45.41% to 46.37%. The two moved toward each other, which had not happened in the previous readings.
⚠️ The house average on the generic ballot stands at D+5.66pp, over 15 polls from 11 pollsters, with field through August 17, and it is against that base that this issue's price is read.
1.Executive Summary
For a trading desk, the week comes down to this: the Senate control price moved and the seat structure behind it did not. That is the exact inversion of the previous week, when the distribution reorganized internally and the binary did not budge. Two straight weeks in which the two books on the same outcome tell different stories, and in neither did the two move together.
The House remains the still contract on the grid. That is two straight weeks, ten trading sessions, in which the Monday close and the Friday close coincide, with a one-band 1.00pp swing inside the week that always returns. This is not an absence of trading: the Democratic House contract is the most liquid of the eight, with USD 5.64M accumulated. It is a price the market reaffirms, not a price nobody touched.
What actually changed sits in the Senate's arithmetic. The binary and the distribution had been drifting apart, and this Monday the distance between them fell to 3.13pp. Anyone reading the two books as a pair has, this time, both moving toward each other in almost equal measure: the binary gave up 1.00pp and the normalized distribution rose 0.96pp. In the previous readings the binary was the number that never moved, and all of the movement came from the other side.
2.Why this brief does not subtract the market from the polls
This section exists because the most tempting calculation on this page is the one that cannot be made. The control contract and the generic ballot measure different things, and the difference is not one of scale or of precision: it is one of quantity.
<strong>Why it matters:</strong> 88.50 minus 5.66 is 82.84, and that number does not exist. It has no unit, it would change size if the poll were expressed on another scale, and it corresponds to nothing that can be checked on November 3. Translating votes into seats runs through 435 districts with different shapes, and in 2012 the Democrats won more House votes and fewer seats. The AFOS panel leaves that edge MUTE on purpose, and this brief does the same.
The legitimate question is not the subtraction, it is coherence: is a market paying 88.50% on Democratic control compatible with a 5.66-point advantage at the ballot box? The honest answer is that it depends entirely on geography, and the market publishes its own answer in the seat distribution contracts, which is where votes become seats. In the Senate, whose eleven bands sum to 97.90% in this reading, that answer is explicit: 52.50% on the Republicans ending with 49 seats or fewer.
There is one contract that would measure exactly the SAME quantity as the poll, the House popular vote margin, and it is worth explaining why it does not appear on this page. Its fourteen bands sum to 151.40%, far outside the 95 to 105 gate AFOS requires, because the contract's structure includes a bucket capturing any other outcome and counted alongside the rest. This is not a one-day fluctuation: it is how the contract was built. It is collected every day to keep the series and is never published as if it held.
3.Weighted scenarios for the week
The three scenarios below are not narrative: they are a direct reading of the Senate seat distribution, the contract in which the market publishes the probability of each band. The eleven bands sum to 97.90%, inside the 95 to 105 gate AFOS requires to publish a distribution, and the three scenarios below cover that entire sum, with no overlap and no hole. The percentages are the market's, not the writer's, and do not constitute a recommendation.
The Republicans end with 49 seats or fewer. It is the sum of three bands: 25.50% on 47 or fewer, 13.50% on exactly 48 and 13.50% on exactly 49. Over the covered week this aggregate gained 0.50pp, all of it in the 48-seat bucket. In this case the Senate changes hands and the House, which the market already gives to the Democrats at 88.50%, follows.
The Republicans end with 51 seats or more and govern without depending on the tiebreak. It is the sum of seven bands, from the 51-seat one at 10.50% to the 57-or-more one at 0.90%. Over the covered week this aggregate gave up 0.60pp, with the largest loss in the exactly-53-seat bucket, which is the number they hold today.
The Republicans end with exactly 50 seats and the tiebreak goes to the vice presidency. This is the scenario that makes the binary and the distribution disagree, because it counts as Republican control in the binary and shows up as an isolated band in the distribution. It gained 0.50pp over the covered week. More than one in four Republican control scenarios runs through here, not through an outright majority: 12.50 out of 45.40.
4.Indicator Grid
| Contract | Current | Δ week | Vol USD acc. | Implied reading |
|---|---|---|---|---|
| House, Democratic control | 88.50% | 0.00pp | USD 5.64M | Second straight week closing where it opened. It is the most liquid of the eight contracts, and the price is reaffirmed, not forgotten. |
| House, Republican control | 11.50% | 0.00pp | USD 4.30M | Mirror of the previous one. Rose to 12.50% mid-week and came back at Friday's close. |
| Senate, Democratic control | 50.50% | +1.00pp | USD 2.39M | Rose at the August 26 close and did not come back. Carries more volume than the Republican contract in the same market. |
| Senate, Republican control | 49.50% | +1.00pp | USD 1.66M | Rose to 50.50% at the August 27 close and gave the gain back on August 30, after the covered week. |
| Senate, 47 Republican seats or fewer | 25.50% | 0.00pp | USD 0.18M | Most probable band of the Senate distribution and the one that moved least: zero variation across the five sessions. |
| Senate, 50-50 tie | 12.50% | +0.50pp | USD 0.13M | The band the binary counts as Republican control and the distribution isolates. It is the arithmetic source of the divergence between the two books. |
| Senate, exactly 53 Republican seats | 5.25% | -0.50pp | USD 0.09M | Largest band drop of the week. It is exactly the number of seats the Republicans hold today, that is, the scenario of neither losing nor gaining any. |
| Senate, exactly 54 Republican seats | 3.95% | 0.00pp | USD 0.78M | Pays 3.95% and carries more volume than any other band in the distribution, four times that of the most probable band. |
| Election happens as scheduled | 96.10% | +0.20pp | USD 0.38M | Closed the week at 96.30% and returned to 96.10%. It is the institutional floor against which the rest of the grid is read. |
| Generic ballot, AFOS average | D+5.66pp | n/a | 15 polls, 11 pollsters | Simple unweighted average over a 30-day rolling window, with field through August 17. Five houses closed field on that same August 17 and three of them landed on the same number, D+5, with Emerson College at D+8. |
5.Liquidity and market structure
6.Electoral calendar through November
| Date | Sample | Why it matters | |
|---|---|---|---|
| Sep 1, Tuesday | Massachusetts primary | 1 Senate seat | Opens the September block, which is the last of the cycle's primary calendar. In the Democratic Senate primary, Ed Markey is seeking a third full term and his challengers include Representative Seth Moulton. |
| Sep 8, Tuesday | New Hampshire primary | 1 Senate seat, OPEN | The only open seat of the four: Jeanne Shaheen announced she is not seeking re-election. A seat with no incumbent is the one that moves the Senate's seat distribution most. |
| Sep 9, Wednesday | Rhode Island primary | 1 Senate seat | Jack Reed is seeking re-election. Second straight day of voting, and the state votes on a Wednesday, outside the Tuesday pattern of the rest of the calendar. |
| Sep 15, Tuesday | Delaware closes the primary calendar | 1 Senate seat, final state | Chris Coons's seat and the final primary of the cycle. From then on the candidate field is closed in every state, and the market starts pricing a settled race. |
| Nov 3 | General election | 435 House seats and 1/3 of the Senate | The date on which every contract on this grid resolves. Until then AFOS measures the distance between the price and the ruler, without arbitrating between them. |
There is no mandatory poll registry in the United States, so this calendar is one of ELECTORAL EVENTS and not of release dates. Each pollster sets its own cadence. South Carolina's special Republican runoff, which appeared on the previous issue's calendar, was held on August 25 and leaves this list: Darline Graham won and faces Democrat Annie Andrews on November 3 for the seat left vacant by Lindsey Graham's death in July.
7.Watch list, week triggers
- The distance between the Senate's two books, today at 3.13pp is the first reading in which both sides move toward each other, the binary giving up 1.00pp and the distribution rising 0.96pp since Friday's close. In the previous ones the binary stood still and the distribution did all the moving. If the next one confirms, it changes which side is correcting which.
- The 50-50 tie band, today at 12.50% is the band that sustains the entire divergence, because the binary counts it as Republican control and the distribution isolates it. It gained 0.50pp over the week. As long as it rises without the binary following, the two books keep disagreeing.
- The four September primaries, starting September 1 Massachusetts, New Hampshire, Rhode Island and Delaware close the cycle's calendar, and all four settle a SENATE candidacy, which is the chamber in play on this grid. New Hampshire's is the only open seat.
- The money that is not where the probability is the four bands at 50 Republican seats or fewer sum to 65.00% of probability and carry USD 0.41M. The seven bands at 51 or more sum to 32.90% and carry USD 2.46M. Half the probability with six times the volume.
- The popular vote margin contract, which remains structurally outside the gate its fourteen bands sum to 151.40%, because the contract includes a bucket capturing any other outcome. It is the only market that would measure the same quantity as the poll, which is why it is collected daily even so, to keep the series. It is never published as if it held.
8.Methodology
AFOS reports the three signals separately: prediction market price, generic ballot polls with sample size and field date, and press coverage from a fixed list of outlets. It builds no composite index and does not subtract one quantity from the other.
Every published price passes through two independent readings eight minutes apart, and is released only when they agree within 0.20pp. This issue uses the August 31 reading, 11:44 BRT, with an approved verdict. Weekly variations use each day's CLOSE, not the first and last point of the window, because the second choice changes the number without changing the fact. The House series begins on July 29, 2026, when collection was switched on, so any superlative about it is always 'of the series', never 'of the cycle'.
A distribution is published only if its bands sum between 95% and 105%. In this issue the Senate's sums to 97.90%, inside the gate, and the three scenarios in section 3 cover that entire sum. The generic ballot average is a simple arithmetic mean, unweighted and with no pollster excluded; when the same house publishes more than one screen from the same round, only one enters, by the hierarchy of likely voters, registered voters, adults. Accumulated USD volume is cited as context and is not guidance; book depth is not cited, by editorial decision of May 21, 2026. Code and method are open under Apache 2.0.
9.Additional reading · macro coverage
AFOS is the primary source of this analysis. The references below are complementary context and support none of the measurements above.
- The Associated Press · Coverage of South Carolina's special Republican Senate runoff, held on August 25, and of the shape of the November races., apnews.com/hub/us-elections
- Bolts · Guide to the September 2026 elections, with the four primaries that close the cycle's calendar and what is at stake in each state., boltsmag.org/whats-on-the-ballot/september-2026-elections-guide/
- The Hill · Coverage of the congressional campaigns, including the New Hampshire Senate race, which holds its primary on September 8., thehill.com/homenews/campaign
Get it by email
AFOS Tradeoff is the weekly read on Brazilian political risk, for people who decide on data rather than headlines.
No spam. Unsubscribe anytime.
After signing up you choose whether to receive it in English, Portuguese or Spanish.