I want a straightforward system that awards points to every harness driver and gallops jockey after each race and then identifies the best-value pick to win the day’s challenge. Here’s the flow I have in mind: • Pull each meeting’s fields, odds, finished positions, and participant names directly from Tab.com.au via web scraping. • Apply the fixed points structure (3 pts = 1st, 2 pts = 2nd, 1 pt = 3rd) race by race. • Design an AI-based formula that converts the official tote price (or starting price) into a “challenge price” showing which driver or jockey is most likely to top the leaderboard. I’m open to statistical or machine-learning methods as long as they are transparent and can be tweaked later. • Output the rolling leaderboard in a clean CSV or simple webpage so I can review the standings live or race-by-race. Key notes – Both harness and gallops meetings must be handled in the same workflow. – Web scraping is preferred; no manual data entry. – The code should run on a typical Windows environment and be reasonably documented so I can update selectors if Tab.com.au changes. – Include a brief outline of the model logic and any libraries you intend to use (e.g., Python, BeautifulSoup, Pandas, scikit-learn). When you submit, please attach a concise proposal describing your approach, timeline, and a relevant example of past data-driven racing or sports work if you have one.