Skip to content

    Under active development. New calculators and features ship every week. Everything currently runs free while the site matures — pricing pages are a preview, checkout is not open yet.

    Search ⌘K
    Calculates locally in your browser · scenario-weighted provision
    Banking / provisioning

    Scenario provision calculator

    A single expected-loss number hides the range. This calculator weights base, downturn and severe scenarios by their probability, producing an allowance that reflects the whole distribution rather than one forecast, and shows how much of the tail the weighted figure ignores.

    Exposure$5,000,000Scenarios3Spread$220,000
    Weighted allowance
    $58,500
    Probability-weighted expected loss
    Worst scenario
    $240,000
    Severe case, not weighted
    Best scenario
    $20,000
    Base case
    Weighted PD
    2.30%
    Across the scenario set
    Scenario table
    Each scenario’s loss and its contribution to the allowance.
    $58,500
    Scenario table
    ScenarioPDLGDLossWeighted
    Base1.00%40%$20,000$12,000
    Downturn3.00%50%$75,000$22,500
    Severe8.00%60%$240,000$24,000

    Weights are normalised, so they need not sum to exactly 1.

    Inspect the model
    Choose a quick result view, a scaled comparison, or the complete diagnostic set.
    Loss by scenario
    The severe case is several times the base case.
    Loss by scenario$0$56.6k$113.3k$169.9k$226.6k$283.2k$20.0kBaseBase$75.0kDownturnDownturnSevereSevereLoss
    Contribution to the allowance
    Weights decide which scenario actually drives the provision.
    Contribution to the allowance$0$5.7k$11.3k$17.0k$22.7k$28.3k$12.0kBaseBase$22.5kDownturnDownturn$24.0kSevereSevereWeighted
    Canonical equations
    The exact relationships implemented by this engine.

    PD × LGD × EAD · Per scenario.

    Σ w_i · EL_i · Weights normalised to sum to one.

    worst − best · How much the outcome depends on the scenario.

    finance methodology

    A clear path from inputs to results.

    A single expected-loss number hides the range. This calculator weights base, downturn and severe scenarios by their probability, producing an allowance that reflects the whole distribution rather than one forecast, and shows how much of the tail the weighted figure ignores.

    Step-by-step methodWorked exampleAssumptions explained

    How to use

    1. Enter the exposure.
    2. For each scenario, set PD, LGD and its probability weight.
    3. Read the weighted allowance.
    4. Compare it with the worst case.

    The formula

    EL=∑iwi⋅PDi⋅LGDi⋅EADEL = \sum_i w_i \cdot PD_i \cdot LGD_i \cdot EAD

    Reference: IFRS 9 / CECL multi-scenario expected credit loss.

    Worked example

    $5M exposure, three scenarios

    Inputs
    • Base: PD 1%, LGD 40%, weight 0.6.
    • Downturn: PD 3%, LGD 50%, weight 0.3.
    • Severe: PD 8%, LGD 60%, weight 0.1.
    Calculation path
    1. 1

      Base loss: 5M × 1% × 40% = $20,000.

    2. 2

      Downturn: $75,000; severe: $240,000.

    3. 3

      Weighted: 20,000×0.6 + 75,000×0.3 + 240,000×0.1 = $58,500.

    4. 4

      The weighted figure is a quarter of the severe case.

    Result

    Weighted allowance $58,500 · severe case $240,000.

    Interpreting the result

    The gap between the weighted allowance and the severe case is the tail the accounting number does not cover. That is deliberate — the allowance is a probability-weighted estimate, while capital and stress testing cover the tail. Reading only one of the two numbers is how banks get surprised.

    Assumptions & limitations

    • Three scenarios with user-supplied probabilities.
    • Losses are PD × LGD × EAD within each scenario.
    • No discounting or lifetime PD term structure.
    • Weights are normalised.

    Frequently asked questions

    Why weight scenarios?

    A single forecast embeds one view of the future; weighting shows the range and its likelihood.

    Should weights sum to one?

    They are normalised, so they need not, but the ratio between them is what matters.

    Does this replace a model?

    No. It is a screening calculation; production models use many scenarios and macro overlays.

    Methodology last reviewed 2026-10-21. Results are informational and not financial or engineering advice — see our disclaimer.