Polling, Prop Bets, and Political Truths: Carl Allen Returns

What do you get when a data scientist with a knack for sports betting and political forecasting returns to the pod? A masterclass in what polling can and can't do, how bad assumptions skew our democracy, and why Carl Allen thinks we need to stop blaming the camera for the race result.

In this fascinating, far-ranging convo, Corey and Carl Allen (author of The Polls Weren’t Wrong) break down why political polling is misunderstood, how data intersects with integrity, and where we go from here in the 2026 election cycle.

They also take a surprising detour into MLB match-fixing, sports betting strategies, and how the “edge of the bell curve” reveals more than most talking heads on TV.

📣 Calls to Action

✅ TELL A FRIEND ABOUT TP&R!!! Help spread the message of meaningful conversation.

✅ Subscribe to Talkin' Politics & Religion Without Killin' Each Other on your favorite podcast platform.

✅ Leave a review on Apple Podcasts, Spotify, or anywhere you listen

✅ Join the community on Substack: coreysnathan.substack.com

✅ Watch & subscribe on YouTube: youtube.com/@politicsandreligion

⏱️ Timestamps & Key Topics

• [00:00] Welcome & Carl’s update on his Substack and data research

• [00:04] MLB pitch-fixing scandal, ethics in sports betting, and data detection

• [00:10] How Carl became a professional sports bettor—accidentally

• [00:14] Finding value in rare events: betting baseball home runs and NFL TDs

• [00:20] Political forecasting vs. polling: "Polls are not predictions"

• [00:28] Nate Silver, spread fallacies, and the misuse of polling data

• [00:34] Why undecided and third-party voters break the math

• [00:43] 2024 election surprises and the power of high voter turnout

• [00:50] What to watch in 2026: the Senate map, gerrymandering, and candidate quality

• [00:56] The TP&R question: Building bridges by talking about anything but politics

🧠 Key Takeaways

• Polls ≠ Predictions: A poll is a snapshot, not a forecast. Yet many analysts treat it like a crystal ball.

• Margins lie: Spread (the difference between candidates in polls) is often misleading—especially when undecided voters are high.

• Integrity matters: Carl warns that analysts must be consistent—praising predictions that turn out right and taking accountability when wrong.

• High turnout shifts everything: New Jersey’s 2025 elections showed that even strong vote counts can get dwarfed by record-breaking turnout.

• Quantifying uncertainty is revolutionary: Understanding what we don’t know is just as important as what we do.

💬 Notable Quotes

“Polls are not predictions of election outcomes. That simple statement contradicts the global consensus.”

“We don't beat up the photographer if the runner in the lead ends up losing. So why do we do that to pollsters?”

“Some two-point leads are better than six-point leads—depending on where you are in the race.”

“Being wrong is fine. Being inconsistent is not.”

“Gerrymandering means leaders get to pick their voters, instead of voters picking their leaders.”

🔗 Connect

• Carl Allen’s Substack: realcarlallen.substack.com

• Carl’s Book: The Polls Weren’t Wrong

Corey is @coreysnathan on...

• Substack
• LinkedIn
• Facebook
• Instagram
• Twitter
• Threads
• Bluesky
• TikTok

🙌 Our Sponsors

• Meza Wealth Management: www.mezawealth.com

• The Village Square: villagesquare.us

Proud members of The Democracy Group

🎧 Thanks for listening! Take a moment to name the good stuff in your life. And as always, go talk politics and religion—with gentleness and respect. 🎙️✨