Prompt
Design a schema (no SQL — just table names, columns, and arrows) for tracking the Lions’ 2025 draft scouting board. Requirements:
- One row per prospect with biographical info.
- Multiple scouting grades per prospect (multiple scouts).
- Combine measurables (40-time, vertical, etc.) if measured.
- Pre-draft visits (date + attendees).
- Post-draft career AV by season.
Three to five tables. Identify primary keys and foreign keys.
Expected output
A schema sketch — Markdown table or a small diagram. Each table should have a name, columns (with types), a primary key, and any foreign keys.
Hint
The pattern: prospects (one row per prospect), and child tables that
reference prospects.prospect_id for each multi-valued concept. The
combine and AV tables are one-to-one with prospects; grades and visits
are one-to-many.
Solution
prospects prospect_id PK TEXT full_name TEXT position TEXT college TEXT birthdate DATE
combine_results prospect_id PK,FK→prospects TEXT forty_yd_dash REAL vertical_inches REAL broad_jump_inches REAL bench_reps SMALLINT
scouting_grades grade_id PK SERIAL prospect_id FK→prospects TEXT scout_name TEXT grade_value REAL -- pick a scale and stick to it (PFF-style 0-99, BLESTO-style 0.0-9.0, or your team's own) grade_date DATE notes TEXT
prospect_visits visit_id PK SERIAL prospect_id FK→prospects TEXT visit_date DATE visit_type TEXT -- 'top30', 'pro_day', 'private_workout'
career_av prospect_id PK,FK→prospects TEXT season PK SMALLINT approximate_value SMALLINTThe key shapes:
combine_resultsandcareer_avuseprospect_idas PK (one row per prospect, or one row per prospect-season for AV).scouting_gradesandprospect_visitsuse aSERIALsurrogate key because you can have many per prospect.