25 Key Races Get Real-Time Breakdowns in New CNN Election Tool

Twenty-five major Senate and gubernatorial races will receive real-time analysis from a new tool called Sightline as Election Day concludes. The system, developed by CNN’s Decision Desk, aims to clarify the status of contests that will determine Senate control and state leadership.

The tool evaluates data through five distinct signals. The first is the outstanding vote. This metric calculates how many unreported votes a trailing candidate requires to win. It determines if that margin is possible based on current performance and the location of remaining ballots. This specific indicator activates once 25% of a state’s expected total vote is reported.

Geography serves as the second signal. It checks if reported votes are evenly distributed across the state, both by location and partisan lean. If the distribution is unbalanced, the model assesses whether rebalancing would alter the final outcome. This calculation begins when 10% of the expected state vote is in.

The third component compares current results to past elections and pre-election expectations. Before polls close, the system relies on race ratings, polling data, and national environment signals from early-reporting states. As the night progresses, the model shifts focus to direct comparisons with 2024 county-level performance. State-specific estimators contribute to this metric once 10% of the expected vote is reported.

The fourth signal measures the size of a candidate’s lead. It assesses the confidence level that a lead will hold, factoring in the volume of reported votes. Like the geography and comparison metrics, this indicator begins calculating at the 10% reporting threshold. The fifth signal analyzes vote methods, such as mail, early in-person, or Election Day voting.

In states where all counties report by method, the model estimates the representation of each voting type. It identifies which candidate benefits when underrepresented methods catch up and extrapolates final results. This component also activates at the 10% mark. The relative weight of these five signals changes throughout the night. For instance, early on, the system places higher weight on historical voting patterns by method. As data becomes more precise, the current vote method data drives the forecast more heavily.

Sightline provides an overall rating on whether a race leans toward a specific candidate. These forecasts are distinct from formal CNN projections. The underlying indicators may align or send opposing signals. In either case, the tool reflects what Decision Desk analysts see in the data. The assessment updates as knowledge of the race grows.