Does your forecast actually beat a fair naive benchmark, out-of-sample? Upload your forecast-vs-actuals history and get an honest verdict — scored against the tougher of two naive baselines using standard published scoring rules. Even when the answer is no.
or drag & drop a .csv / .tsv file here
Load a built-in example: a monthly demand model whose forecast looks busy but actually loses to the naive baseline. A good illustration of a FAIL.
Need a template? Download a blank CSV. Required columns: actual and forecast. Optional: date (for ordering), and sigma or lo/hi (to also score calibration & CRPS).