How well did Kalshi predict the midterms? We crunched the data.
Qué dice
A CBS News piece analyzes how accurately the prediction market Kalshi forecast the 2026 midterm elections. According to the summary, a CBS News analysis found that the candidate favored by Kalshi traders won 95% of congressional and gubernatorial races.
Por qué este veredicto
Only an RSS feed summary is available, so there is very little text to evaluate. What is present is consistent with routine post-election data journalism: a national outlet with an established data/elections unit running an accuracy check on a prediction market shortly after the vote. Kalshi is a real, CFTC-regulated exchange that has offered election contracts since late 2024, so the premise of the story is plausible rather than invented. The single statistic (95% of congressional and governors' races) is attributed to an in-house CBS News analysis, which is the expected form of attribution for this kind of piece, and prediction markets routinely post high hit rates because most House and Senate seats are uncompetitive. There is no hype language, no shouting headline, and nothing contradicting established fact. The missing byline and the lack of matching stories elsewhere are weak signals: data pieces are often lightly bylined in feeds, and a proprietary in-house analysis would not be expected to be duplicated by other outlets. I am overriding the neutral specificity and AI-marker priors, which the scorer itself flags as unmeasurable on this amount of text. Confidence is capped by the summary-only evidence and my inability to verify the 95% figure or confirm the article exists at the stated URL.
Too little text to assess style meaningfully. The one sentence reads like a standard CBS News deck line summarizing an internal analysis; nothing in it suggests template-generated filler, but a single sentence cannot rule it out either.
Afirmaciones examinadas
-
unverifiable
Kalshi traders' favored candidate won 95% of congressional and governors' races in the 2026 midterms
Specific proprietary finding from a CBS News analysis published after my knowledge cutoff. The magnitude is plausible given that most House seats are safe, but I cannot check the number.
-
supported
Kalshi operated prediction markets on 2026 U.S. midterm races
Kalshi is a real CFTC-regulated event exchange that won the legal right to list U.S. election contracts in 2024; offering midterm markets follows directly.
-
unverifiable
CBS News conducted its own analysis of prediction market accuracy
In-house attribution is the normal form for this kind of data piece, but the underlying methodology is not visible in the feed summary.
Señales
| Signal | Strength | Score |
|---|---|---|
|
Confianza en la fuente
Estimación previa de reputación del dominio que publica. Un punto de partida, no un veredicto.
|
|
0.72 |
|
Corroboración
Cuántos medios independientes publicaron una noticia coincidente, con descuento cuando todos son de nivel bajo.
|
|
0.32 |
|
Especificidad
Densidad de personas nombradas, citas directas, declaraciones atribuidas, cifras y fechas.
|
too little text to assess | — |
|
Marcadores de texto de IA
Marcadores estilométricos de relleno generado. Más alto significa más mecánico. Un indicio débil, nunca una prueba.
|
too short to assess style | — |
|
Carga emocional
Lenguaje sensacionalista, indignado o evasivo. Más alto significa más cargado.
|
|
0.00 |
|
Autoría
Si se acredita a una persona concreta como autora.
|
|
0.00 |
Advertencias
- Assessment is based solely on a one-sentence RSS summary; the full article could not be retrieved, so confidence is capped.
- No independent corroboration was found, but an in-house proprietary analysis would not be expected to appear elsewhere.
- The verdict leans partly on the outlet's reputation and on the plausibility of the premise rather than on verified details of the finding.
- There was too little text to assess writing style or sourcing density, so this verdict rests on the outlet and on corroboration.
- Judged from the feed summary only; the full article could not be retrieved.
Procedencia
| Evaluado por | claude-opus-5 |
| Texto disponible | Solo el resumen del canal |
| Encontrado en | CBS News Politics (outlet canal) |
| Visto por primera vez | Nov 10, 14:05 EST |
| Estimación previa combinada | 0.49 |