Python Forecasting / Simulation Engineer for forecasting prototype

I need the data/modeling built for a forecasting prototype, with outputs structured so they can connect to a Supabase-backed web app. The dataset comes from a public application/draw system with data that requires cleaning, normalization, and validation. It is a preference-point-based lottery where higher point holders generally have priority, but some winners may be selected through random allocation depending on the rules. The initial milestone is focused on one state / one application category and should produce: Current-year odds by point level Forecasted draw probability by point level Estimated draw window by point level Estimated year to reach a 90% cumulative draw probability Estimated cumulative cost until drawn Clean output tables structured for Supabase/Postgres Documentation of assumptions, methodology, limitations, and update process The model should account for point creep, applicant demand changes, quota changes, cost changes, and inconsistent public data sources. Preferred implementation is Python using pandas/numpy or similar. I’m open to alternatives if there’s a strong reason, but the workflow must be reproducible, documented, and structured for Supabase/Postgres. Strong fit if you have experience with forecasting, simulation, probability modeling, Monte Carlo, public data cleaning, reproducible workflows, and model validation. I’d like to start with a small paid test milestone. If the prototype goes well, there is potential for a much larger pipeline covering additional states, application categories, and a more complete forecasting system. Please include how you would approach forecasting a preference-point lottery with point creep and inconsistent public data.