A sharpshooter can hit 40% from deep and still miss his makes over because the offense never gave him shots. That is why NBA three-point props need two models: one for attempts, one for makes. You will start with a binomial baseline, find the chance of four or more makes from eight attempts, then let attempts bounce between six, eight and ten and see how the tails change even when the average stays put. You will also treat shooting percentage as uncertain instead of fixed, so a hot week does not run your projection. Finish with a price check against the market. Volume first, accuracy second, and the number you bet gets more honest.
Learning objectives
- Explain why a three-point prop needs separate models for attempts and makes.
- Build a baseline count model and let attempts vary.
- Check a modeled price against the market and validate the workflow.
