How we model strength of schedule
DraftSOS turns a full season of matchups into floor, median, and ceiling outcomes — so your draft and lineup calls rest on the whole range of what each schedule can produce, position by position.
What we can prove
Loading validation figures…
The model
DraftSOS runs a Monte Carlo simulation over each team's schedule. Every matchup is sampled many times against estimated opponent strength, building a distribution of outcomes rather than a single point estimate. From that distribution we surface the floor, median, and ceiling defensive rank a position is likely to face each week, and we map the median onto the easy / medium / hard tiers shown in the SOS grid.
How we validate it
We measure the model out-of-sample. Using leave-one-season-out cross-validation, we hold an entire NFL season aside, predict it from the remaining seasons, and compare the prediction to what actually happened. We track three measures across positions: rank correlation (how closely the predicted order matches the actual order), mean absolute error (average distance between predicted and actual defensive rank), and directional accuracy (how often we call the easy-versus-hard direction correctly). Holding a full season out keeps the test honest and free of hindsight fitting.
Published results
Our season-by-season validation is in progress. As each season's out-of-sample results are confirmed, we will publish the per-position figures here so you can see them for yourself. In the meantime, the same Monte Carlo engine described above powers every ranking in the free SOS grid — you can put it to work today.
See the model in action.
The same Monte Carlo engine powers the free SOS grid. Try it now — instant, free access.
Our validation uses mean absolute error, rank correlation, and out-of-sample cross-validation, with methodology adapted from open-source fantasy analytics research. Open source credits