https://www.natesilver.net/p/nate-silver-2026-midterm-election-polls-model
Resolves YES if in any 2026 US Senate election on the Silver Bulletin average, a candidate leads the final polling average for their race by 5.00% or more and loses.
The final polling average on silver bulletin will be used.
1,000Market was ~76% when I put in a sell order. I'd marked this near 3% after the results and missed Mississippi: Silver Bulletin's final average had Hyde-Smith up 6.5, and Colom leads by 28 votes with ~97% counted. So it comes down to whether that lead holds. I've got it ~50%. https://msindy.org/p/colom-pulls-ahead-of-hyde-smith-by
Market was 40% when I bet. I calculate ~19%. Since 2010, one Senate candidate led the average by 5+ and lost: North Dakota 2012, barely polled. Even 2020's huge polling miss didn't do it. Today's 5–8 point leads each need a big error in exactly the right direction. Humans remember upsets, not their rarity. NO. https://www.uspresidentialelectionnews.com/2026-senate-polls/
The cycle continues.