themortgagellm™Rate distribution across FICO × LTV cells
Prompt
“Give me the median lock rate for conforming 30-year purchase locks in the last 14 days, broken out by FICO band and LTV band.”
Response
Trailing 14-day median lock rate per FICO × LTV cell (Conforming 30YR Fixed Purchase). Cells with fewer than 5 locks are suppressed to avoid noise.
| FICO ↓ / LTV → | ≤60 | 75-80 | 80-85 | 90-95 | >95 |
|---|---|---|---|---|---|
| 780+ | 6.375% | 6.490% | 6.490% | 6.375% | 6.500% |
| 740-759 | 6.375% | 6.500% | 6.500% | 6.500% | 6.500% |
| 720-739 | 6.375% | 6.500% | 6.500% | 6.625% | 6.500% |
| 700-719 | 6.375% | 6.625% | 6.625% | 6.625% | 6.500% |
| 660-679 | 6.490% | 6.750% | — | 6.875% | — |
| 620-659 | 6.500% | 6.625% | — | 6.750% | — |
Note-rate spread is narrow (~50 bps top-to-bottom), LLPA carries the risk price. Most of the credit-and-LTV differentiation moves through the LLPA overlay rather than the note rate itself. This is why the Market Snapshot page shows median rate + median LLPA in each cell — the rate alone under-states how differently a 780/60 loan is priced vs a 640/95 loan.