The News, Explained

Cloudflare published its report on internet traffic and attacks during the U.S. midterm elections on November 19, 2026. It examines election-day access alongside different attacks observed beforehand. Each number needs its population, period and measurement unit. Source

The observations come from three Cloudflare protection programs: the Athenian Project, Cloudflare for Campaigns and Project Galileo. The company says they protected more than 1,400 election-related websites, including state and local governments, candidates, parties, journalists and civil society. These are the provider’s observations and classifications within its protected population. Source

Ordinary traffic to protected state and local government election websites rose from about 130 million requests on November 1 to 602 million on November 3. The report gives a 363% increase; comparing those approximate values yields about 4.6 times the original volume. The body measures web requests. One person can generate multiple requests, so requests and people need to be distinguished. Source

Cloudflare reports no significant DDoS attacks on its protected election-related sites on November 3. DDoS, or distributed denial of service, involves requests from multiple sources burdening a service. The scope of this statement is one election day, the provider’s protected election-related websites and attacks it classifies as significant DDoS. Source

Over the 12 months before the election, the company says it mitigated 392 million vulnerability attacks on government election websites. Such attacks try to exploit weaknesses in website code to access internal systems or data. The scope here is government election websites, a 12-month period and attacks defended against. Those labels should remain attached when this figure is read beside the one-day DDoS observation. Source

Email offers a denominator example. In the preceding 90 days, 3.9% of email sent to Campaigns-protected organizations was classified as malicious. Nearly 42% of that malicious email passed basic authentication checks, such as sender verification or checks for tampering in transit, but was identified by the company’s additional phishing detection. The first percentage uses all incoming email; the second uses the malicious subset. Authentication and malicious-content detection also have distinct roles. Source

OYOPICK’s Take

The useful perspective is to consider high election-day demand alongside attacks observed in the preceding period. Availability of essential information and protection of the code, data and email behind services deserve attention together. Keeping the time window and organization group attached to each figure helps avoid compressing different measurements into one risk ranking.

We would welcome follow-up reporting that further separates defended attempts from confirmed harm for the same period and population. That would help readers assess both why many threats were detected and what protection achieved. We hope better observation and defense technology will support reliable public access to essential information, with its effects evaluated alongside the scope and duration of the evidence.