Risk Management Interactive

Stress Testing

Model worst-case scenarios and extreme events. Essential for business continuity, reserve planning, and understanding tail risks.

💪 Stress Testing Questions

🏈

Event Risk

"What if every NFL favorite loses this Sunday?"

🎯

Sharp Attack

"What if a syndicate targets our mispriced lines?"

🔗

Correlation

"What if all same-game parlays hit simultaneously?"

Portfolio Settings

100000 2000000

2 15

Scenario Parameters

5 40

10 60

10 50

⚠️ Summary

Expected Daily P&L +$40,000

Worst Case Loss $-260,000

Loss Multiple 6.5x expected

Scenario Analysis

Base Case

Normal operation

+$40,000

+8.0% margin

Upset Heavy Slate

Multiple favorites lose across all games

$-60,000

-12.0% margin

Correlated Losses

Same-game parlays all hit (QB + WR stacks)

$-96,000

-32.0% margin

Sharp Syndicate Attack

Coordinated sharp money exploits mispriced lines

$-25,500

-17.0% margin

Black Swan Event

Star player injury mid-game, mass cancelations

$-42,000

-42.0% margin

Maximum Adversarial

Perfect storm: upsets + correlations + sharps

$-260,000

-52.0% margin

90-Day P&L Simulation with Stress Events

Simulation includes 5% daily probability of stress event. Stress days show significant drawdowns.

🛡️ Risk Mitigation Strategies

Exposure Limits

→ Cap single-event exposure (e.g., 5% of daily)
→ Limit correlated parlay combinations
→ Set per-player prop maximums

Dynamic Adjustments

→ Move lines to balance exposure
→ Reduce payouts on heavy action
→ Suspend betting on suspicious patterns

Reserve Management

→ Maintain cash reserves = 2-3x worst case
→ Reinsurance for catastrophic events
→ Circuit breakers for extreme losses

📋 Stress Testing Framework

Identify Risks

Brainstorm worst cases

Quantify Impact

Calculate $ exposure

Probability

Estimate likelihood

Mitigate

Implement controls

R Code Equivalent

# Stress testing framework
run_stress_test <- function(portfolio, scenarios) { 
  results <- lapply(scenarios, function(scenario) { 
    # Apply scenario to portfolio
    stressed_pnl <- portfolio$exposure * (portfolio$hold_rate - scenario$shock) / 100
    
    list(
      scenario = scenario$name,
      pnl = stressed_pnl,
      loss_multiple = stressed_pnl / portfolio$expected_pnl
    )
  })
  
  return(do.call(rbind, lapply(results, as.data.frame)))
}

# Define scenarios
scenarios <- list(
  list(name = "Base Case", shock = 0),
  list(name = "Upset Heavy", shock = 20),
  list(name = "Correlated Losses", shock = 40),
  list(name = "Sharp Attack", shock = 25),
  list(name = "Black Swan", shock = 50)
)

# Run analysis
portfolio <- list(exposure = 500000, hold_rate = 8, 
                  expected_pnl = 500000 * 8 / 100)

stress_results <- run_stress_test(portfolio, scenarios)
print(stress_results)

✅ Key Takeaways

• Stress tests answer "what if worst case happens?"
• Model correlation—independent bets rarely all lose
• Reserve = 2-3x worst case scenario loss

• Combine historical and hypothetical scenarios
• Update scenarios as business evolves
• Test mitigations before crisis hits