Adverse Selection
When informed bettors have an edge, they bet more. The book's customer pool becomes adversely selected—dominated by those who beat the line.
🎯 The Adverse Selection Problem
If your prices are wrong, the people who know they're wrong will bet more. You end up trading disproportionately with informed counterparties.
The Lemons Problem
Like used car markets: only owners of bad cars want to sell at average price. In betting: only bettors with edge want to bet against you.
Who Bets?
- Sharp (+EV): Bets when edge > cost
- Public (-EV): Bets for entertainment
If sharps see +EV, they load up. Book's mix shifts toward losers (for the book).
Market Parameters
📊 Book's Expected Profit
Betting Pool Composition
✓ Healthy mix. Public volume covers sharp losses.
Common Adverse Selection Scenarios
Early Lines
⚠️ Sharps bet first, get best prices
✓ Lower limits on openers
Injury News
⚠️ Informed bettors exploit slow line moves
✓ Faster line updates, API monitoring
Prop Bets
⚠️ DFS players have research edge
✓ Wider spreads on props
Live Betting
⚠️ Bettors watching game vs book with delay
✓ Faster feeds, wider spreads
🛡️ Mitigation Strategies
Screen
Limit/ban sharp bettors
Price
Widen spreads on vulnerable markets
Speed
Faster line updates reduce info edge
Pool
Attract more recreational bettors
R Code Equivalent
# Adverse selection model
calculate_book_profit <- function(sharp_pct, sharp_edge, public_edge, hold) {
sharp_ev <- sharp_edge - hold
public_ev <- public_edge - hold
# Book's profit is negative of bettor EV
sharp_loss <- sharp_pct / 100 * sharp_ev * -1
public_win <- (1 - sharp_pct / 100) * public_ev * -1
return(sharp_loss + public_win)
}
# Find break-even sharp %
break_even_sharp <- function(sharp_edge, public_edge, hold) {
# At break-even: sharp_loss = public_win
# s * (sharp_edge - hold) = (1-s) * (hold - public_edge)
s <- (hold - public_edge) / (sharp_edge - public_edge)
return(s * 100)
}
profit <- calculate_book_profit(10, 5, -2, 5)
cat(sprintf("Book profit: %+.2f%%\n", profit))✅ Key Takeaways
- • Adverse selection: informed bettors bet more
- • Book's pool skews toward those with edge
- • Must win enough from public to cover sharp losses
- • Screen: limit sharp bettors
- • Price: widen spreads on vulnerable markets
- • Speed: faster updates reduce info asymmetry