New Blask data shows a sharp divide between brand attention and trading activity, with Kalshi closing the search gap even as it commands substantially more trading volume.
Kalshi may dominate the dollars changing hands in U.S. prediction markets, but Polymarket continues to command considerably more branded search attention according to new data from iGaming analytics firm Blask.
Blask tracked search demand for prediction markets between January and August 2026. The company found that Polymarket retained the largest share of branded prediction-market search demand, as measured by its Brand’s Accumulated Power (BAP) metric, throughout the period, although Kalshi significantly narrowed the gap.
Polymarket lost 16.56 percentage points of BAP through August, from 74.3% to 57.8%. Meanwhile, Kalshi gained 12.7 points from 23.1% to 35.8%. The two companies’ combined share fell from 97.4% to 93.6%. Among smaller competitors, Novig posted the fastest January-to-July increase in Blask Index, rising 64.9%.
The results look very different from the trading volume data. In August, Kalshi recorded $37.17 billion in volume, compared with a combined $8.16 billion across Polymarket and Polymarket US, according to The Block.
Blask says the contrast highlights what the two datasets actually measure.
BAP measures share of branded search demand — people looking up the brand — not share of matched trades or notional volume,” Blask told Gambling Insider.
Polymarket’s edge is largely a name-recognition effect: it’s the more established, more globally-covered brand, so it pulls more search attention per user even where Kalshi is winning more of the actual money.”
As Blask put it: “Trading volume is a liquidity/execution metric; BAP is an attention metric.”
Kalshi Closes the Attention Gap
Blask Index is based on branded search activity rather than transactions, users or revenue.
The company combines geo-tagged search data with contextual filtering designed to isolate betting or trading-related searches and remove irrelevant or negative-intent queries. Its BAP metric measures how the resulting demand is divided between competing brands.
That distinction becomes important when looking at Kalshi’s growth.
During the World Cup, Kalshi’s BAP increased from 18% in May to 29% in June and 36% in July. Polymarket moved in the opposite direction, falling from 80% to 69%, then to 61%. By August, Kalshi’s BAP had reached 35.8%, according to Blask. Polymarket is still ahead, but with a considerably smaller advantage than at the beginning of the year.
Kalshi’s gains were not simply the result of Polymarket losing share. Its Blask Index increased by 22.8% between January and July, even as the overall prediction-market segment contracted by 22%, indicating that Kalshi was attracting more branded search demand in absolute terms while the broader category cooled.
For multi-product operators, Blask said it also isolates searches specifically connected to prediction markets. That helps explain why DraftKings and FanDuel can have enormous sportsbook recognition while registering little prediction-market-specific BAP.
“Kalshi and Polymarket, by contrast, exist only as prediction markets, so effectively all their branded search counts,” the company said.
Record Trading Does Not Necessarily Mean Record Search Demand
One of the biggest examples of the contrast between search and trading activity occurred during the World Cup.
According to Reuters, Kalshi generated about $27 billion in trading volume during the World Cup. Other estimates have been lower, depending on what activity is counted. TickerTracker calculated approximately $13.8 billion across Kalshi’s individual World Cup markets, excluding combo markets.
The broader monthly data underscores the scale of the surge. Pew Research Center, analyzing The Block data, put combined Kalshi and Polymarket trading volume at nearly $53 billion in July. Kalshi accounted for roughly $40.1 billion, compared with about $12.9 billion for Polymarket and Polymarket US combined.
Yet the tournament did not generate Blask’s largest single-day prediction-market demand reading.
Instead, the Blask Index reached its 2026 peak on March 2, the day before the first midterm primaries in Texas, North Carolina and Arkansas. That day’s Index was more than four times a mid-May baseline.
Super Bowl Sunday exceeded the same baseline by more than 3 times, while the highest World Cup reading reached only about half the March 2 level.
Blask cautioned, however, that it cannot establish that the primaries caused the March spike.
“The primaries link is a timing inference — the spike lands exactly on that date — not a query-level breakdown proving people were searching with election intent specifically,” the company said.
That makes the safer conclusion that the year’s largest branded-search spike coincided with the opening of the primary calendar rather than that political markets definitively caused it.
The World Cup instead produced something less dramatic but more sustained.
Median daily prediction-market Index levels during the six-week tournament were 38% above the four weeks before kickoff. Meanwhile, Blask’s traditional U.S. iGaming measure was 2% lower over the same comparison period.
Discovery and Usage May Be Telling Different Stories
Blask believes the discrepancy could reflect different stages of consumer engagement.
A big share of World Cup trading volume likely came from users who were already acquired and active — placing more bets/trades without needing to search the brand again to do it,” the company said.
However, separate mobile app data suggest the tournament also brought substantial numbers of new users into prediction markets.
Sensor Tower found that during the World Cup’s opening three weeks, Kalshi’s average weekly downloads were nearly twice its Super Bowl LX level. Meanwhile, Polymarket’s were nearly four times higher. The analytics firm described prediction markets as the primary driver of growth among U.S. sports-wagering apps during the period.
Apptopia similarly reported continued growth in prediction-market apps during the tournament, with rising daily active users even as sportsbook app growth began to retreat.
That suggests the gap may partly reflect how consumers reached the platforms, rather than simply whether they were new or existing users. New customers can discover Kalshi or Polymarket directly through app stores, advertising, referrals or social media without generating the branded web searches captured by Blask.
A prediction-market user can also repeatedly trade contracts after installing an app without performing another search for the brand. Conversely, someone can search for Kalshi or Polymarket without ever placing a trade.
In that sense, branded search, app acquisition and trading volume may each capture different stages of the customer journey.
Blask described the distinction as the difference between “more people are discovering/considering this” and “existing users are trading more.”
Search Demand Has Not Yet Been Validated Against Prediction Market Activity
There is an important caveat when interpreting the data.
Blask’s methodology is based on Share of Search, which measures a brand’s share of search demand relative to competitors. The company says its own testing in traditional iGaming found a 95% correlation between search share and actual market performance.
However, Blask has not yet tested whether the same relationship holds for prediction markets.
Whether SoS-to-performance holds for prediction-market trading volume specifically is an open question we haven’t run the study on yet, not one we’re claiming a yes or no on,” the company said.
For now, that leaves branded search demand as a useful measure of attention rather than a substitute for trading volume or prediction-market share.
The post Report: Polymarket Leads US Prediction Market Search Demand Despite Kalshi’s Trading Dominance appeared first on Gambling Insider.