Analytics Disclaimer

1. What the Service Is (and Is Not)

The Service is a sports intelligence and research platform. It ingests historical and live sports data (team, player, and game statistics) from third-party data feeds, applies computational models — including machine-learning ensembles, quantum-inspired simulations, Bayesian inference, and other statistical methods — and produces analytical artifacts such as probabilities, factor analyses, projections, rankings, and narrative explanations (“Model Output”).

The Service is intended for:

• Research into sports analytics and computational methods;
• Educational and informational use by subscribers;
• Commentary, discussion, and quantitative exploration of sporting events.

The Service does not:

• Accept, facilitate, broker, settle, or process any wager, contest entry, or real-money transaction tied to the outcome of any sporting event;
• Offer any guarantee of outcome, financial return, or performance;
• Provide advice of any kind.

2. Not Advice of Any Kind

Any reliance you place on Model Output is strictly at your own risk. You are solely and exclusively responsible for any decision you make, and you agree that Neural Flow Dynamics LLC is not liable for any decision, action, or omission based in whole or in part on the Service.

3. No Guarantee of Accuracy or Outcome

Model Output represents a statistical estimate of likelihood derived from historical patterns, current data, and model structure. It is not a prediction in the colloquial sense, it is not a certainty, and it is not a guarantee.

We make no representation or warranty, express or implied, regarding:

• The accuracy of any specific Model Output;
• The completeness or timeliness of any data displayed;
• The likelihood that any sporting event will occur as the model estimates;
• The continued availability or reliability of any Service feature.

Sports are inherently uncertain and are affected by factors that no model can fully capture, including injuries, weather, rest, motivation, coaching decisions, officiating, and random variation. Events that a model estimates as highly unlikely will still occur, and events the model estimates as highly likely will still fail to occur. That is the nature of probability.

4. Methodology and Data Sources

Model Output is generated by an ensemble of computational methods that may include, without limitation:

• Deep neural networks (TensorFlow.js) trained per-league on historical game-level data;
• Classical machine-learning classifiers (support vector machines, random forests, gradient-boosted trees, logistic regression);
• Quantum-inspired variational circuits and tensor-network simulations;
• Bayesian inference, Gaussian processes, and time-series methods;
• Domain-specific feature engineering from team, player, and game-level statistics.

Source data is obtained from third-party sports data providers. While we perform reasonable validation and quality-assurance, we do not warrant the accuracy or completeness of any third-party data. Errors, delays, or gaps in third-party data will affect Model Output.

5. How to Read Accuracy Figures

Where the Service, our marketing pages, or our research documentation reference accuracy percentages, hit rates, calibration statistics, or similar figures, you should read them as follows:

• Unless explicitly stated otherwise, reported figures reflect historical backtesting on validation data drawn from completed games. They describe how the model performed on that specific dataset, under those specific conditions, at that specific time.
• Backtested results are not predictive of future performance. Live accuracy varies with sample size, data availability, league conditions, schedule, injury patterns, and model drift.
• Accuracy is reported at the granularity of the underlying claim. A “home wins” accuracy is not interchangeable with “home wins by 2+” accuracy, which is not interchangeable with joint or compound claim accuracy.
• The Service may display separate confidence, calibration, or sample-size indicators. These are diagnostic fields and are not themselves accuracy figures.

6. How to Read Probabilities

A model probability (for example, “62%”) is the model’s estimate of the likelihood of an event given the inputs and the model’s structure. It is not a statement that the event will occur. A well-calibrated 62% estimate means that, over a long sequence of similar 62% estimates, the event is expected to occur about 62% of the time — and therefore not occur about 38% of the time. Individual events are not a test of the model.

7. Known Limitations

Users should be aware of the following material limitations inherent in any statistical sports model:

• Data gaps. Injury reports, line-up information, weather, and officiating data are not always available, are sometimes late, and sometimes conflict across sources.
• Regime changes. Rule changes, coaching changes, roster turnover, and structural changes in a league can invalidate historical patterns.
• Sample size. Early-season, expansion, international, and low-volume markets have materially less data and therefore materially wider error bands.
• Non-stationarity. Player performance, team chemistry, and competitive balance evolve over time. Models trained on older data may be less informative about the current period.
• Tail events. Rare, high-impact events (major injuries, weather cancellations, unusual officiating) are by definition under-represented in training data.
• Model risk. All models are simplifications of reality. Structural misspecification, numerical instability, and training artifacts can all affect Model Output.

8. User Responsibility

You are responsible for how you use the Service and for any decisions you make. In particular:

• You should treat Model Output as one input among many. Do not treat any single probability, projection, or factor as dispositive.
• If a decision based on Model Output could materially affect your finances, your health, or your legal rights, you should seek qualified professional advice before acting.
• You must comply with all laws applicable to you in your jurisdiction. The availability of the Service in your jurisdiction does not imply that a given use of the Service is lawful there.

9. Trademarks and Non-Affiliation

NBA, NFL, MLB, NHL, Premier League, La Liga, Bundesliga, Serie A, and all league, team, player, and broadcaster names, logos, and related marks are registered trademarks of their respective owners. We use these marks only to identify the subject of our research and commentary, under the doctrine of nominative fair use.

Neural Flow Dynamics LLC is an independent research company. The Service is not affiliated with, endorsed by, sponsored by, or approved by any league, team, player, coach, broadcaster, data provider, or any of their affiliates.

10. Third-Party Data Quality

The Service depends on data supplied by third-party sports data providers. We do not control the timing, accuracy, or completeness of that data. We are not responsible for errors, delays, interruptions, or omissions in third-party data, or for any resulting inaccuracies in Model Output. Where we detect material data issues, we may suppress, annotate, or withdraw the affected output.

11. Forward-Looking Statements

Any statement on the Service or in our communications about future events, future performance, roadmap, or expected results of the Service itself (as opposed to the sporting events it analyzes) is a forward-looking statement that reflects our current views. Actual outcomes may differ materially, and we undertake no obligation to update any forward-looking statement except as required by law.

12. Changes to This Disclaimer

We may update this Analytics Disclaimer from time to time. When we make material changes we will revise the “Effective date” above and provide notice through the Service. Continued use of the Service after the effective date of a change constitutes acceptance of the updated Disclaimer.

13. Contact

For questions about methodology, data sources, or specific Model Output, please contact us at the address above.

This Analytics Disclaimer is provided for review and reflects enterprise-grade defaults for a statistical sports research platform. Before relying on this Disclaimer for customer-facing use, it should be reviewed and finalized by qualified legal counsel licensed in your jurisdiction.