Dimension scores are derived from public data and fields; weighted into the composite. Reference only.
MoneyPuck is an advanced analytics and game prediction platform focused on the NHL (National Hockey League). Developed by Peter Tanner using Python and AWS, it provides in-depth game insights for hockey fans, data analysts, and bettors through multidimensional statistical models.
Features and Use Cases: The platform’s core features include pre-game win probability forecasts, playoff probability simulations (100,000 simulations per season), projected starting goalie predictions, and real-time in-game win probability modeling. Its standout value lies in proprietary advanced metrics such as “Expected Goals Adjusted for Shot Sequence,” “Expected Goals Adjusted for Shooting Talent,” and “Expected Goals Created.” These metrics help correct biases in traditional data and more accurately reflect the true performance of players and teams.
Data Sources and Scale: The models are built on a solid data foundation. The expected goals model was trained on more than 800,000 shots and 50,000 goals from the 2007–2015 seasons, while the pre-game prediction model uses data from the 2017–2024 seasons. Model weighting is clearly disclosed: scoring chances account for 54%, goaltending for 29%, and win ability for 17%. Historical results show that its pre-game prediction accuracy has remained stable at 60%–64%, making it a useful reference.
Pricing and Trial: The source text does not mention any pricing information, subscription plans, or API access. It is likely positioned as a free public-interest or hobbyist tool, so there is no concept of a free trial.
Support Channels and Integrations: Access is available only via the web. Support channels are limited to Twitter and the developer’s personal email, with no enterprise-grade customer support. No third-party integrations or API support are mentioned.
Pros: Very high model transparency, with variables and weights disclosed in detail; innovative metrics that fill gaps in the industry; stable prediction accuracy.
Cons: Extremely niche audience, limited to the NHL; advanced metrics have a high learning curve; lacks commercial data interfaces.
Best-Fit Users: Hardcore NHL fans, sports bettors, and hockey data analysts. It is mainly suitable for individuals or small research teams.
Access from China: Unknown. Since it is hosted on AWS and contains no sensitive content, it can usually be accessed directly, but there is no confirmed network availability information. Payments are not relevant.
Alternatives: Hockey data sites such as Natural Stat Trick and Evolving Hockey.
⚠ This review is compiled from public sources and does not constitute a purchase recommendation. Verify all facts on the vendor's official site. Verify on moneypuck.com official site.
moneypuck.com is an United States Marketing & SEO provider. TG4G tracks its product information, an overall rating of 7.0/10, and a China-accessibility score of China direct-connect friendly. Click "Visit Official Site" to reach moneypuck.com directly.