PLONN is a machine learning model designed to optimize sports betting parlays for maximum Return on Investment (ROI). By analyzing historical and in-season data, PLONN ranks potential parlay combinations based on their likelihood of winning and their potential payout.
Implement a neural network that takes a series of sports bets and produces a ranked list of different parlays for maximum ROI.
Given the bets:
Bruins (-135) vs. Avalanche (+ 200)
Ducks (-120) vs. Leafs (+100)
Lighting (+136) vs. Oilers (-162)
[Bruins, Ducks, Lighting] (+ 461),
[Bruins, Ducks, Oilers] (+ 240),
[Avalanche, Ducks, Oilers] (+156),
Where the first parlay has the best balance of win probability and ROI, the second has a higher win probability but lower ROI, and the third has an equal win probability to the second but an even lower ROI.
Understand the intricacies of neural networks, their applications, advantages, and limitations. This project serves as a hands-on learning experience in the realm of ML/AI.
After rigorous testing and achieving satisfactory accuracy, the plan is to launch a Discord bot offering this service on a subscription basis.
Data Collection: Target bets based on in-season and past-season data. Some teams may exhibit trends based on historical matchups, coaching strategies, or player rivalries.
Date | TeamA | TeamB | Total Goals | O/U | Dataset A Pred.(in-season) | Predicted # of Goals A | Dataset B Pred. (historical) | Predicted # of Goals B | Notes |
---|---|---|---|---|---|---|---|---|---|
4-Oct | Detroit Red Wings | Pittsburgh Penguins | 3 | 6.5 | 0 | ||||
4-Oct | New York Rangers | New Jersey Devils | 7 | 6.5 | 1 | ||||
4-Oct | Calgary Flames | Edmonton Oilers | 9 | 6.5 | 1 | ||||
4-Oct | Seattle Kraken | Vancouver Canucks | 3 | 6.5 | 1 | ||||
Acc % | 75% | 0 | |||||||
5-Oct | Washington Capitals | Columbus Blue Jackets | 6 | 6.5 | 0 | 5.7 | 1 | 6.6239 | |
5-Oct | Columbus Blue Jackets | Toronto Maple Leafs | 7 | 6.5 | 0 | 5.99 | 0 | 5.9103 | OT for 7th Goal |
5-Oct | Florida Panthers | Tampa Bay Lightning | 9 | 6.5 | 0 | 6.9 | 0 | 6.26 | |
5-Oct | Boston Bruins | New York Rangers | 4 | 6.5 | 0 | 6.47 | 1 | 6.6 | |
5-Oct | New York Islanders | Philadelphia Flyers | 7 | 6.5 | 0 | 5.63 | 0 | ||
5-Oct | Carolina Hurricanes | Nashville Predators | 6 | 6.5 | 0 | 5.9 | 0 | 5.18 | |
5-Oct | Dallas Stars | St. Louis Blues | 4 | 6.5 | 1 | 6.6 | 0 | 6.35 | |
5-Oct | Winnipeg Jets | Ottawa Senators | 3 | 6.5 | 0 | 6.16 | 0 | 5.4 | |
5-Oct | Minnesota Wild | Chicago Blackhawks | 3 | 6.5 | 0 | 5.37 | 0 | 4.83 | |
5-Oct | San Jose Sharks | Los Angeles Kings | 7 | 6.5 | 0 | 6.18 | 0 | 4.7686 | OT for 7th Goal |
5-Oct | Arizona Coyotes | Anaheim Ducks | 0 | 6.5 | 0 | 5.25 | 0 | 5.7548 | |
5-Oct | Colorado Avalanche | Vegas Golden Knights | 1 | 6.5 | 1 | 6.6 | 0 | 6.0907 | |
Acc % excl. OT | 84% | Acc % incl. OT | 66.6% |
Predicted 11/16 games accurately, 2/16 games were predicted accurately however when into over time for their 7th goal.
/Data/accuracy.xlsx
will be updated along the course of the 2023-2024 season for more up-to-date accuracy