Simulate outcomes
The acceptance model scores My student against the sample schools, then Monte Carlo simulates 20,000 admissions seasons.
Per-school admit probability
| School | Base rate | Admit prob. | Band / tag |
|---|---|---|---|
| University of Michigan | 18.0% | 19.5% | reach |
| University of Illinois Urbana-Champaign | 45.0% | 47.5% | target |
| Northwestern University | 7.0% | 10.1% | reach |
| Purdue University | 53.0% | 55.5% | target |
| Michigan State University | 83.0% | 84.4% | safety |
Portfolio simulation
How this worksWhat a Monte Carlo simulation tells you▾
Each school above has a single admit probability — but one number hides what actually matters to a family: the range of ways the season could unfold across your whole list. A Monte Carlo simulation answers that by playing the season out thousands of times. In each simulated season, every school is an independent weighted coin-flip at its admit probability; we tally the results across 20,000 seasons.
How to read the numbers:
- P(at least one admit) — the share of simulated seasons with one or more acceptances. The single best gauge of whether a list is “safe enough.”
- Expected # of admits — the average number of acceptances across all seasons.
- P(affordable admit) — the chance at least one acceptance comes in at or under your budget.
- Distribution of outcomes — below, the full spread: how often you’d expect 0, 1, 2, … admits.
The admit probabilities are themselves estimates, so read these as a calibrated range, not a forecast — and note this baseline treats schools as independent, while strong applicants tend to get several admits at once (a refinement on our roadmap). See the methodology.
Distribution of outcomes
Across 20,000 simulated seasons, how often you land exactly this many admits.
number of admits
Model accept-v0.1 · baseline heuristic, to be retrained on outcome data.
These outputs are estimates from a baseline model — not guarantees of admission, cost, or outcome.