The Risk Takers Podcast Failed Betting Projects & Kalshi Parlays | Ep 124
13 snips
Oct 8, 2025 Dive into the world of failed betting projects, where insightful lessons abound. Discover the complexities behind building college football and NASCAR models amid biases and market assumptions. Learn about Kalshi's new parlay product and its implications for prediction markets. The hosts discuss regulatory challenges in the betting landscape and how competition can outshine restrictive rules. Plus, listener questions tackle success metrics in golf models and automation in trading.
AI Snips
Chapters
Transcript
Episode notes
Laps-Gained NASCAR Experiment
- John tested a NASCAR 'laps gained' metric by comparing lap times against median race pace and translating seconds to laps gained.
- The metric handled mid-race failures but failed when simulation needed crashes, pit stops, and starting-grid correlation.
Beware Tail Failures In Models
- Regression models break down at tails and extreme cases, so naive models overstate edges on very bad teams.
- Adjust models for tail behavior or you'll chase false edges like betting on the 'worst' team every week.
Pick Markets With Real Execution Upside
- Don't build strategies in tiny markets unless the upside justifies the effort and execution complexity.
- Prioritize markets where you can actually execute scalably, not just where modeling looks interesting.