A risk model is only as good as its track record. If you don’t check whether its warning signals match reality, you’re flying blind. This chapter gives you the statistical tools to stress‑test VaR and Expected Shortfall models, showing whether they are trustworthy or dangerously flawed.