Podcast
Gaming the system: Paul Mitchell on the impact of prediction markets and bad polls
Image created by Andrii Yalanskyi, iStock Images. CAPITOL WEEKLY PODCAST: A poll from Median Strategies roiled the Los Angeles mayoral race earlier this month, showing Mayor Karen Bass with an 11 point lead over challenger Nithya Raman. The results were cited by several media outlets, and were touted by the Bass campaign before the poll was exposed as fake. When contacted, Median Strategies posted a statement that their polling had been “a short-term social experiment examining how purported polling information could enter and spread through the political information ecosystem without independent verification.”
Social experiment or not, the poll did impact the race – at least in the short term – and certainly impacted election prediction markets like Kalshi and Polymarket. Political data expert Paul Mitchell joined us to talk about how junk polls, paid influencers and strategic gaming of prediction markets can distort perceptions about a campaign. Plus, who had the Worst Week in California Politics?
2:01 “That’s the Pacific Ocean”
4:35 Capitol Weekly Internship
7:30 Death penalty
9:04 Paul Mitchell
10:23 Bogus polling
13:37 Picked up by media
17:02 “Everybody’s a pundit”
20:48 Prediction markets
22:13 $7M bet on Spencer Pratt AFTER the election
23:56 Could prediction markets replace polling?
27:15 How is this legal?
30:05 Predictit
31:35 A note about gaming
34:01 Duane Dichiara debt
34:48 Scott Lay’s election pool
35:33 Junk polls designed to generate coverage
38:28 “Traditionally, your campaign’s internal polling is some of the highest quality polling”
41:03 What California race should people be watching more closely?
43:15 The split in the Democratic Party
47:14 #WWCA
Want to see more stories like this? Sign up for The Roundup, the free daily newsletter about California politics from the editors of Capitol Weekly. Stay up to date on the news you need to know.
Sign up below, then look for a confirmation email in your inbox.

Leave a Reply