{"id":722,"date":"2026-10-21T07:47:00","date_gmt":"2026-10-21T14:47:00","guid":{"rendered":"https:\/\/quantcha.com\/news\/?p=722"},"modified":"2026-10-20T18:17:32","modified_gmt":"2026-10-21T01:17:32","slug":"an-election-night-scorecard-for-prediction-markets","status":"publish","type":"post","link":"https:\/\/quantcha.com\/news\/an-election-night-scorecard-for-prediction-markets\/","title":{"rendered":"An Election Night Scorecard for Prediction Markets"},"content":{"rendered":"\n

Four checks for whether the odds are taming election uncertainty, written down before the results.<\/em><\/p>\n\n\n\n

On October 13, SPX options priced election night at 1.75 normal days, with Kalshi’s odds of a Republican sweep at 10.5%. By the close on October 20, the sweep had slipped to 9.25% and the premium to 1.65. That’s about the move you’d expect if options took their cue from the odds, after two months of mostly going the other way.<\/p>\n\n\n\n

One week doesn’t make a trend. Last week I asked<\/a> whether prediction markets are taking uncertainty out of election night, using what SPX options charge for that night as the measuring stick, and my answer was “not visibly, yet.” With two weeks to go, that “yet” deserves a test written down in advance, with a pass\/fail line for each check, so that when the results come in I’m grading a prediction instead of telling a story about one. Two of the checks play out between now and Election Day, as the odds and the premium move, so there’s something to watch before the votes are counted, and they don’t depend on who wins. The other two get graded the day after, if divided government wins as expected.<\/p>\n\n\n\n

How it’s supposed to work<\/h1>\n\n\n\n

Republicans hold the White House, the House, and the Senate. A sweep keeps it that way, and Kalshi gives it about a 9% chance. Every other result is divided government, at about 91%. That’s a change, but it’s the expected change, so if the odds are doing their job the market is already priced for it. The extra premium options charge for election night pays for the surprise, which this year is one party keeping everything. The more certain the odds, the less surprise is left, and the more of the market’s adjustment happens before the election instead of all at once.<\/p>\n\n\n\n

A quick note on units. I measure the premium in normal days of variance, which is how risk adds up across days. On that scale, 1.65 normal days works out to a move about 1.3 times the size of a normal day’s.<\/p>\n\n\n\n

There are two ways prediction markets could help. They can add information, telling investors something they didn’t already know. Or they can take over some of the hedging, so protection that used to be bought in options moves to event contracts. Information is the use CFTC Chairman Michael Selig had in mind when he said event contracts “provide the public with information about the outcome of future events,”<\/a> so most of what follows tests it.<\/p>\n\n\n\n

Check 1: The premium should follow the odds<\/h1>\n\n\n\n

If options take their cue from the odds, the election-night premium should shrink as a sweep gets less likely and grow as it gets more likely. We can even say by how much. October 20’s prices value election night at 1.65 normal days with the sweep at 9.25%, which implies a sweep would move the S&P about 2% in one direction or the other. Hold that fixed and the premium should be about 2.0 normal days at a 15% sweep, 1.4 at 5%, and 1.15 at 2%.<\/p>\n\n\n\n

For most of the fall, that’s not what happened. In late August, with the sweep at about 15%, options charged between 1.0 and 1.4 normal days for election night. Once the sweep slid below 10%, the premium climbed to between 1.5 and 1.9. In other words, the market’s sense of how much a sweep would matter went up while the odds of one went down.<\/p>\n\n\n\n

You could argue the real question is the Senate, the one chamber still in doubt. It doesn’t help. Kalshi’s Senate odds went from a coin flip in mid-August to about 65% Democratic, so the uncertainty there fell too, and the premium still went up.<\/p>\n\n\n\n

\"\"
Each dot is one trading day from August 17, when SPX listed the election expirations, through October 20. The dashed line is where the October 20 premium would go if it moved with the sweep odds.<\/em><\/figcaption><\/figure>\n\n\n\n

Readings right after the August listing are noisy, and not all of the premium has to be about the election, but neither explains a climb that kept going through September and October.<\/p>\n\n\n\n

My line: average the last three readings before Election Day and compare them with the dashed line at the average sweep odds over those same days.<\/strong> Within 0.15 of the line counts as following the odds. If the sweep moves less than 2 points from 9.25%, the premium’s normal day-to-day wobble is too big to tell anything, and I’ll call it inconclusive.<\/p>\n\n\n\n

Check 2: The expected result should be an ordinary day<\/h1>\n\n\n\n

If divided government wins, the S&P should move about like a normal day on Wednesday, which as of October 20 is about 0.75%. Most of the election premium would go unused, which is what a well-informed market should look like.<\/p>\n\n\n\n

My line: a move of less than 1.0% from Tuesday’s close to Wednesday’s counts as ordinary.<\/strong><\/p>\n\n\n\n

The best comparison is 2018, which started from the same place. Republicans held all three, and Democrats took the House. That was the expected result, and the S&P still rallied 2.1% the next day, about three times a normal day’s move. There are two good explanations, and Kalshi’s balance-of-power market measures neither. Part of the premium pays for uncertainty itself, so when it goes away, even an expected result can lift prices. And a new House majority is still a change. Investors react to how big it is and which seats flipped, while the market only settles on who holds the gavel.<\/p>\n\n\n\n

It’s the check I’m least sure of, which is why it’s worth writing down. A quiet Wednesday says the odds had it priced, and a 2018-style move says options were paying for something the odds can’t see. If the sweep happens instead, this check doesn’t apply.<\/p>\n\n\n\n

Check 3: It should come in below 2022<\/h1>\n\n\n\n

2022 was the last midterm without Kalshi’s congressional markets, so it’s the closest thing to a control group. Two weeks out, 2022 was pricing election night at about 2.05 normal days. 2026 is at 1.65, which is the direction taming would point.<\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

2022’s election expirations listed 29 days before the vote, so its line starts later. 2026’s listed on August 17. Readings through October 20, 2026.<\/em><\/p>\n\n\n\n

I wouldn’t call it yet, though. Over its final five trading days, 2022 swung between 1.34 and 1.81 and averaged 1.57, so a gap this size can come and go in a week. <\/p>\n\n\n\n

My line: 2026 has to average 1.2 or less over its final five readings to count as clearly below.<\/strong> It has averaged 1.61 over the past five, so that’s a high bar.<\/p>\n\n\n\n

Check 4: The sectors should split, not the index<\/h1>\n\n\n\n

There’s a version of taming the S&P alone can’t show. With a Republican president, the veto limits how much a Democratic Congress can change, so the index effect of divided government should be modest. But a Democratic sweep and a split Congress could still treat healthcare, defense, energy, and banks differently, with sectors moving against each other while the S&P nets out.<\/p>\n\n\n\n

If the odds really have done their job ahead of time, that’s what election night should look like: a quiet index and a reshuffle underneath. It isn’t what the last two midterms looked like. The day after 2018, all 11 sector funds rose, by between 0.6% and 2.9%. The day after 2022, all 11 fell, by between 0.8% and 4.9% (that was also the day FTX collapsed, so it’s not the cleanest read).<\/p>\n\n\n\n

My line: if divided government wins, the S&P moves less than 1.0% and at least 3 of the 11 sector funds move the opposite way by more than 0.5%.<\/strong> Neither 2018 nor 2022 had a single one. On an ordinary day it happens roughly one time in ten.<\/p>\n\n\n\n

I’m not going to predict which sectors go which way. Sector options only expire on Fridays, so they can’t isolate election night.<\/p>\n\n\n\n

How much money is behind the odds<\/h1>\n\n\n\n

In May I wrote about how cheap thin markets are to push around<\/a>, where $30,000 would have cleared Polymarket’s entire Seahawks book.<\/p>\n\n\n\n

Kalshi’s sweep market is thinner than you’d expect for one of the most-watched elections anywhere. As of the close on October 20, about $238,000 of resting orders stood between the sweep’s 9.25% and half that. That’s order-book depth, the money it takes to move the price. The most the whole balance-of-power market could pay out, across all four outcomes, is about $35 million.<\/p>\n\n\n\n

Thin isn’t the same as easy to fool, since this number has the world watching and plenty of capital ready to buy it back if it looks wrong. A smaller foreign election that draws less attention from big money is another story.<\/p>\n\n\n\n

What matters here is the thinness. Open interest in the SPX options expiring the day after the election covers about $51 billion of the index, so an options trader has little reason to take cues from a number backed by a rounding error by comparison.<\/p>\n\n\n\n

Hedging can’t close the gap either. All of Kalshi’s Congress markets together could pay out about $79 million as of October 14, enough to cover about $4 billion of stock against a 2% move, so protection isn’t moving out of options at this size or in this shape.<\/p>\n\n\n\n

Where I land<\/h1>\n\n\n\n

I expect this week’s step along the dashed line to turn out to be noise, with the premium finishing well above 1.2 and Wednesday moving more than 1% even if divided government wins, sectors mostly together. That would be “not visibly” again.<\/p>\n\n\n\n

Here’s what would change my mind: the premium following the odds into the final week, a finish clearly below 2022, a quiet Wednesday after divided government, or a calm S&P with sectors pulling in opposite directions.<\/strong> I’ll be checking all four once the results are in. I’m also curious whether the same test tells a different story for the Fed, which decides on October 28. What would you need to see on November 4 to believe the odds made election night calmer?<\/p>\n","protected":false},"excerpt":{"rendered":"

Four checks for whether the odds are taming election uncertainty, written down before the results. On October 13, SPX options priced election night at 1.75 normal days, with Kalshi’s odds of a Republican sweep at 10.5%. By the close on October 20, the sweep had slipped to 9.25% and the premium to 1.65. That’s about … Continue reading “An Election Night Scorecard for Prediction Markets”<\/span><\/a><\/p>\n","protected":false},"author":1,"featured_media":723,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":[],"categories":[8],"tags":[],"_links":{"self":[{"href":"https:\/\/quantcha.com\/news\/wp-json\/wp\/v2\/posts\/722"}],"collection":[{"href":"https:\/\/quantcha.com\/news\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/quantcha.com\/news\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/quantcha.com\/news\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/quantcha.com\/news\/wp-json\/wp\/v2\/comments?post=722"}],"version-history":[{"count":2,"href":"https:\/\/quantcha.com\/news\/wp-json\/wp\/v2\/posts\/722\/revisions"}],"predecessor-version":[{"id":727,"href":"https:\/\/quantcha.com\/news\/wp-json\/wp\/v2\/posts\/722\/revisions\/727"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/quantcha.com\/news\/wp-json\/wp\/v2\/media\/723"}],"wp:attachment":[{"href":"https:\/\/quantcha.com\/news\/wp-json\/wp\/v2\/media?parent=722"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/quantcha.com\/news\/wp-json\/wp\/v2\/categories?post=722"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/quantcha.com\/news\/wp-json\/wp\/v2\/tags?post=722"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}