ON THE RISE 2026

On the Rise 2026 uses data from the KnowWho Candidate Data Service. KnowWho editors have managed, maintained, and updated these data since November 2025. The information is updated weekly by the organization; the data used in this report were last updated on September 6, 2026.

House of Representatives candidates were identified in all US states and territories; Senate candidates were only collected for states in which a race occurred in the 2026 cycle.

KnowWho identified 3,918 candidates who filed to run for the House and Senate in the current cycle. Of these, 3,428 candidates filed for the House, including delegate seats representing Washington, DC, and overseas territories, and 490 candidates filed for the Senate.

Some candidates appeared multiple times in the dataset. Handling these cases required special scrutiny. When candidates were outright duplicates—with the same name, biography, electoral contest, and other details—the extra entries were removed. This is accounted for in the candidate total of 3,918. Of these candidates, researchers sourced or estimated age data for 2,758.

Candidates could appear more than once in the dataset for a variety of reasons. In some cases, individuals “shop” among districts, looking for a race where they were more likely to find success. This could mean running in two states at once, or losing the primary in one state and then filing for a race in a different state. In other cases, candidates filed to appear on the ballot in one race and then jumped to another, for example abandoning a Senate bid to run for the House. For some statistics used in this report, it was important to tally candidates’ data for all of the races in which they participated. In those cases, multiple observations pertaining to a single individual were left in the dataset and treated as independent. There were 11 candidates to whom this practice was applied; age data were sourced or estimated for 9 of them.

Figures and analysis throughout the report specify the denominator or total set of candidates as necessary for clarity.

Identifying a candidate’s exact age can be difficult, as doing so requires an exact date of birth, with month, day, and year. Neither Federal Election Commission (FEC) nor state filing forms require this information. Occasionally, as in Alaska, official state candidate statements will ask for the individual’s age, but this is rare, and candidates are not required to submit a statement at all, let alone a complete one.

KnowWho provides biographies for some candidates, from which age data could be sourced. Other well-known candidates have Wikipedia pages with source-supported birth dates. In a handful of cases, full birth dates were posted on campaign websites. If none of these sources were available, researchers turned to publicly available databases and candidate social media.

In the absence of age data from the sources above, the team performed internet searches using the candidate’s name, state, and electoral contest, combined with Boolean operators. Search results including specific age data were given priority, but secondary sources citing a candidate’s age were used to estimate birth year when other information was lacking. The secondary sources were mainly dated news articles or candidate surveys.

In the event that full or partial date of birth could not be determined, common life or major historical events were also used to estimate age. These included high school graduation or military enlistment, if there was some indication that the enlistment occurred immediately after high school or college. In some cases, a combination of concrete years and events were combined to generate an estimated age. Similarly, if a candidate mentioned their age at the time of a known historical event, such as the terrorist attacks of September 11, 2001, that was used to estimate their current age.

For some candidates, all that could be determined was the year they graduated college, which posed additional hurdles to estimation. In the event that the college graduation year was identified, candidate biographies, social media posts, and interviews were examined to ensure that no gap year or military service preceded the higher education.

If there was reason to suspect that a candidate’s college enrollment or graduation wasdelayed, college graduation was not used to estimate age.

A final source for estimation was the rare set of cases in which a candidate reported their decade of birth or a news article identified their generational cohort. These data were only used if no other information about the candidate’s age could be found.

When the year of birth was estimated rather than precise, researchers used a conservative estimate. For example, if someone graduated from high school in 2005, it was assumed that they graduated at the age of 18 but had not yet had a birthday that year. Their birth year would thus be estimated at 1986. In the instances for which this methodology was used, the estimation rarely fell along the dividing lines for generational cohorts, only affecting 88 individuals, or 2% of cases.

As stated above, age data were sourced or estimated for 2,758 candidates, or 70.4% of all candidates in the dataset. For 30 of these candidates, the only age data available was birth month and day, but there was not enough information, or information was too contradictory, to estimate a birth year. Therefore, any generational analysis in the report includes 2,725 candidates.

Concrete data (full birth date or exact birth year) were sourced for 1,044 candidates, or 38.3% of those for whom ages were sourced or estimated. A birth year was estimated for another 1,678 candidates. The proportion of candidates with known or estimated ages was roughly the same across chambers of Congress: 70.2% for House candidates and 71.8% for Senate candidates.

It is worth noting that candidates for whom age data could not be sourced were less likely to make it to the primary election ballot. Among candidates without age data, 44% failed to make it to the ballot, compared with 19.6% of candidates whose ages were known or estimated.

Data on party identification was collected by KnowWho for all candidates in the dataset. For some statistical outputs, these data were recoded to simplify analysis and reporting, for example by grouping all third parties and those identified by KnowWho as “other” into a single category. These data are 100% complete.

Data on gender were collected by KnowWho for most candidates. Of the 3,918 candidates in the dataset, 22 were missing gender designations. For those, researchers used Boolean searches to source the information. Transgender candidates were coded according to gender identity, and candidates with a nonbinary gender identity were coded as such. Gender information could not be reliably found for four candidates; age data were also lacking for these individuals. Overall, gender data are 99.9% complete for all candidates and 100% complete for those with age data.

Jamie Smith, PhD, and J.D. Rackey, PhD

The defining narrative of the 2026 congressional midterm election cycle is a call for change. Polling consistently puts congressional approval levels at their lowest point in decades, with recent polls suggesting that only about 10–15% of the country approves of the job Congress is doing. Other indicators, such as public satisfaction with the economy and the direction of the country, all suggest that Americans are unhappy with the status quo. Robust dissatisfaction will of course be associated with a call for a change in the policies and politicians that are in place. It should be no surprise that 2026 has seen the candidate pool grow by 31%, with 3,918 candidates filing to run for Congress, compared with 2,983 candidates who filed to run in 2024.

Even before the first ballot was cast during primary season, 32 House members and 10 senators declared that they would not be seeking reelection. Given additional vacancies caused by death or resignation and incumbents seeking other offices, the 2026 election cycle was set to spur a large shift in how Americans are represented in Congress. Now, after a primary period in which 11 incumbents lost their bids, the freshman class of the 120th Congress is already expected to be one of the largest in the 21st century, regardless of the general election outcome in November.

One of the major factors driving turnover in this cycle has been a call for generational renewal. Recent polling indicates that an overwhelming majority of voters would support age limits on candidates running for Congress, with 78% of Democrats, 83% of Republicans, and 79% of independents saying they would favor the adoption of such a policy.