The who of South Carolina's Senate runoff stopped being a market on Tuesday night. Within an hour of polls closing at 7 p.m., NBC News' Decision Desk had projected that appointed Senator Darline Graham had beaten Congressman Ralph Norman for the Republican nomination, and Kalshi's head-to-head board went where a settled race goes: 99 cents bid on her contract, a penny offered on his, with just over 7 million contracts traded across the pair in 24 hours. The strange thing is what kept trading afterward. Sitting under the headline race is a quieter board that asks only how many people showed up, priced as a ladder of nine yes-or-no thresholds running from 200,000 votes to 400,000, and on runoff day it did roughly 384,000 contracts, about 64 percent of everything it has traded since it opened. Eight of the nine already read like answers, either bid in the high 90s or offered for pennies. One is still a live argument, and by a coincidence almost too neat to print, the number of contracts the board traded in a day is nearly the same as the number of voters it is arguing about: 375,000. This page reads that board, against the one benchmark that makes it interesting. The first round of this primary drew just under 335,000 voters. The crowd is now paying near-certain prices that the runoff beat it.
The Quick Answer
Kalshi's turnout market prices the South Carolina Senate runoff at more than 350,000 total votes with near-certainty: a 97-cent bid at 8 p.m. Eastern on August 25, on a contract that pays a dollar if the certified count clears the line. More than 375,000 was quoted at 14 to 19 cents and falling as returns arrived, and 400,000 sat at 2 to 3 cents. That central read of 350,000 to 375,000 would clear the first round's turnout of just under 335,000, which is the opposite of how runoffs usually go. The full nine-rung board, the volume spike that repriced it, and why the count still matters after NBC projected Graham the winner are below.
Free: The Weekly PM Market Brief — the 8-model panel's graded record, the week's biggest market-vs-model gaps, and what's spiking next. One email, Sundays. These are market prices and model estimates, not predictions of fact and not financial advice. Kalshi is a CFTC-regulated event-contract exchange (18+; availability varies by state).
The Market
New to this corner of Kalshi? The race itself has been covered on our South Carolina Senate runoff odds page since the field was eleven names, and the market's sharpest moment of the cycle is documented on the price-history page. This piece reads the board that was still trading like an open question after the winner was known: turnout.
| Field | Detail |
|---|---|
| Venue | Kalshi, a CFTC-regulated event-contract exchange (18+; availability varies by state, as of August 2026) |
| The Event | "What will the voter turnout be in the South Carolina Republican Senate special primary runoff?" (ticker KXPRIMARYTURNOUT-SCRSENS26) |
| The Contract | Nine yes/no markets, one per threshold from 200,000 to 400,000; each resolves YES if the total vote count tops that threshold |
| Settles | On the officially certified results from the state's election authority, per the contract rules, so final settlement waits for certification rather than election-night projections |
| Prices Below As Of | 8 p.m. Eastern on Tuesday, August 25, 2026, about an hour after polls closed |
Kalshi structures this as a ladder: a separate yes-or-no contract for each turnout threshold, so a trader only has to judge whether the certified count clears a line, never the exact number. Read all nine rungs together and the board sketches the crowd's full expectation, the way a set of season win totals sketches a team's range. The board is young, and it grew with the election it was measuring: it opened on August 19, the first day of early voting, with six thresholds topping out at 325,000, and Kalshi has been adding higher rungs ever since. For most of its life it traded lightly. Then came runoff day: roughly 384,000 of its 599,000 lifetime contracts changed hands in the final 24 hours, with about 367,000 contracts still open an hour after the polls closed. A board doing two-thirds of its lifetime volume in a day is a crowd showing up to argue right as the real information arrives.
The Board
Here is the full ladder, quoted bid to ask, at 8 p.m. Eastern on August 25. The bid is what a buyer was actually paying for YES at that moment; a 99-cent bid implies the crowd treats the rung as close to settled.
| Runoff Turnout Above | Quote (Bid To Ask) | Contracts Traded On Runoff Day |
|---|---|---|
| 200,000 | 99¢ to $1.00 | ~9,300 |
| 225,000 | 99¢ to $1.00 | ~900 |
| 250,000 | 99¢ to $1.00 | ~4,100 |
| 275,000 | 99¢ to $1.00 | ~12,600 |
| 300,000 | 99¢ to $1.00 | ~88,600 |
| 325,000 | 99¢ to $1.00 | ~95,000 |
| 350,000 | 97¢ to 99¢ | ~109,800 |
| 375,000 | 14¢ to 19¢ | ~30,800 |
| 400,000 | 2¢ to 3¢ | ~32,600 |
The shape is a cliff. Every rung through 350,000 is bid in the high 90s, then the board falls off a ledge: 375,000 is quoted in the teens, and 400,000 is priced as a curiosity. That puts the market's central expectation between 350,000 and 375,000 certified votes, with the honest disagreement compressed into a single rung.
The row worth narrating is the one the money moved through, and it is not the live one. The 350,000 rung traded about 109,800 contracts on runoff day, the most on the board and nearly everything it has ever traded, because Kalshi only listed that threshold on Saturday, once the early-vote totals made a number fifteen thousand above the first round worth pricing at all. By Tuesday evening the crowd had bid it to 97 cents. Then the exchange stretched the ladder twice more in the middle of runoff day itself, listing 375,000 just after noon Eastern and 400,000 half an hour later, a board being extended upward in real time because the day kept outrunning it. The 375,000 rung, hours old and still trading in the teens, did less than a third of the 350,000 rung's volume. The heaviest trading on this board was the work of turning "did the runoff really out-draw the first round?" from a debate into a price, and the leftover argument is only about the size of the overshoot.
A Worked Example: Reading The 375,000 Rung
One rung is worth walking through slowly, because it shows how much information a two-digit price carries. At 8 p.m. the 375,000 contract was quoted 14 cents bid, 19 cents asked. A YES contract bought at the 19-cent ask pays a full dollar if the certified count tops 375,000 and nothing if it does not, so that price implies somewhere under a one-in-five chance; the 14-cent bid is what a holder could actually sell for at the same moment. The five-cent gap between the two numbers is itself a fact about the market: on a rung this uncertain, with certification days away, nobody was eager to quote it tight.
And the quote was moving while we read it. At 7:53 p.m., minutes before the read above, the same rung was bid at 23 cents against a 26-cent ask; seven minutes later the bid had fallen to 14. The first batches of real returns were doing to this rung what a bad first quarter does to a live point spread, and the direction of the move says the early count was landing toward the low end of the market's range. Both reads are stamped, and the gap between them is the most honest freshness warning this page can give: by the time you read this, the rung has moved again.
The Number To Beat Was 335,000
None of those prices means anything without the baseline. The August 11 first round of this special primary, the eleven-name contest that produced this runoff, drew just under 335,000 ballots, about 11 percent of registered voters. Runoffs in this state are typically the smaller half of the pair; the second date, with a shorter ballot and two weeks of fatigue, historically draws fewer voters than the first. Our own runoff-eve analysis carried that assumption too, and so did the eight-model panel scored on that page, where several seats read the early-vote curve as a smaller, more habitual electorate.
The early signal that the assumption might fail was small enough to miss. Early voting for the runoff ran August 19 through 21 and drew 63,686 ballots, per state Election Commission data, a slight step up from the 61,755 who voted early in the first round. A three percent bump in early ballots is a whisper of a signal, but it left the smaller-election assumption unconfirmed at exactly the moment the market needed it confirmed. What remained was a board forced to price a real range, which is why runoff day itself, the first day of actual precinct information, is when nearly two-thirds of its lifetime volume showed up. By Tuesday evening the crowd's verdict on the baseline was unambiguous: the runoff did not shrink. At a 97-cent bid on 350,000, the market prices this second date at least four to five percent bigger than the first, in an election whose entire premise, a late-August special primary runoff, is the kind of race turnout models expect almost no one to attend.
Why 375,000 Was Worth An Argument
That leftover rung owes its life to what this runoff turned into. The reason the first round's 335,000 stopped being a ceiling is the same reason our runoff coverage ran to five pages: the race got louder every week it existed. Graham, the late Senator Lindsey Graham's sister, was appointed to the seat in July and carried President Trump's endorsement plus a Friday rally in Myrtle Beach. Norman, a sitting congressman, stacked up Nikki Haley and Jim DeMint, among others, and, four days before the vote, Mark Sanford. The only debate produced the soundbite of the cycle when Graham said national security is "not my thing", and her contract gave back 24 cents of a 90-cent close in the sessions that followed. A runoff that repriced that hard on a debate answer was pulling casual attention, and casual attention votes.
Both candidates also brought documented records the electorate has had years to weigh, which belongs on any page pricing them. Norman's file is old news: the 2018 meeting where he set his loaded handgun on the table in front of constituents, and the January 2021 text in which he urged Mark Meadows toward "Marshall Law" to keep Trump in office, a message he later attributed to frustration. Graham's file holds no lawsuit or ethics action we could find; the documented knock is the seat itself, an appointment framed by critics as a famous name inherited, plus the debate answer above. None of it moved prices this week, and that is the point of listing it: the late repricing on this race traced to the debate stage and the count, while both records had been priced in for years.
All of that noise is the mechanism behind the ladder's cliff. A sleepy runoff lands under its first round and the whole board settles at 99 cents by breakfast. A loud one drags new voters in, and the crowd has to find the line where the growth stops. On Tuesday night it put that line between 350,000 and 375,000, and the fading teens quote on the higher rung is the market saying the overshoot probably stays under forty thousand votes.
The Winner Market Died At 99 Cents. This One Gets To Keep Trading
By the time the turnout board was doing its heaviest volume, the headline race had stopped paying attention. NBC News projected Graham the winner on runoff night, her contract was bid at 99 cents against Norman's penny, and the roughly 7 million contracts that crossed the pair in 24 hours were mostly traders collecting or cutting positions on a question with a known answer. She goes on to face Democrat Annie Andrews on November 3, a race the 2026 Senate control board will fold into the chamber math, and the two runoff verdict pages, the case that was made for Graham and the one for Norman, now belong to the graded record.
The board under the headline race is built differently, and the difference is worth understanding if you read these markets regularly. The nominee contracts could settle off a projection: Kalshi's rules for that market make it "eligible for accelerated determination after a consensus of media organizations project the winner." The turnout contracts settle on the officially certified count, and only that; per the contract's own rules, provisional and absentee ballots count exactly as the certification reflects them, and preliminary counts and projections settle nothing. Certified results land days after election night, on the state canvassing calendar rather than the networks'. So the board carrying the night's biggest leftover question is also one that cannot be graded on election night, and anyone still buying the 375,000 rung in the teens is making a specific claim: that late-counted ballots and final precinct totals push the certified number past a line the election-night count leaves in doubt. Where that claim exhausts itself is the 400,000 rung: at 2 to 3 cents, nobody seriously argues this election found seventy thousand voters the first round never met.
One caution before anyone reads too much into a single rung: the cliff rungs are the deep part of this board, and the live rung is not. The 375,000 contract traded about 30,800 contracts on runoff day against the 350,000 rung's 109,800, so the market's confident statement, that the runoff out-drew the first round, rests on much more money than its open question does. Prices here can move with every batch of returns; treat the quotes on this page as a stamped photograph of runoff night, not a standing offer, and re-read the board before acting on anything.
The Bottom Line
Strip the night down to its two numbers and the story tells itself. Just under 335,000 South Carolinians voted in the first round of this special primary, and an hour after the runoff's polls closed the crowd on Kalshi was paying 97 cents on the dollar that the second round beat that. A late-August special runoff, the format that is supposed to empty the electorate, instead out-drew its own first round behind a Trump-backed appointee, a defiant congressman, and a debate answer that repriced the race twice. The winner market has already given its verdict and gone quiet. Turnout stays an open market until the certified count lands in the days ahead, which makes it the sharpest question left on the South Carolina board, and the certified number will grade this page's snapshot either way. Every market this cycle we have priced in public is tracked on the 2026 Senate races board, and if you would rather sweat something graded by a final score tonight than by a canvassing board days from now, our free expert picks today are open to everyone, no account needed.
- South Carolina Senate Runoff Odds: Graham Vs Norman, Kalshi
- South Carolina Senate Runoff Odds: Kalshi Beat The AP Call
- When Is The South Carolina Senate Runoff? Date, Matchup And Kalshi Odds
- 2026 Senate Races: Every Kalshi Board We Track
Kalshi is a CFTC-regulated event-contract exchange (18+, availability varies by state). This article is informational and does not constitute financial, investment, betting, or voting advice.
