Are Swimming’s Fitness and Competitive Industries Data Fit for AI? – Part 2

Introduction

Welcome back. In Part 1, we highlighted why data quality matters for AI-driven solutions, explored the risks of poor data, and outlined key principles for building AI-ready data structures. Now we shift from theory to practice. In this second installment, we look at the current state of swimming training session data, including gaps, inconsistencies, and missed opportunities. We also explore the potential for a unified framework, with references to Wise Racer's zone-based systems, and address the main question: Is the swimming industry ready to use AI well?

By the end of this post, you will have a clearer picture of the barriers that still exist and actionable insights into how coaches, organizations, and stakeholders can drive the next phase of data-centric innovation across the sport.

Sections Covered in Part 2

• Section 4: The Current State of Swimming Training Session Data Management Evaluates how sessions are currently documented, stored, and interpreted—highlighting the inconsistencies and gaps limiting effective data usage.
• Section 5: So, Is the Swimming Industry Data Fit for AI? Answers the key question driving this series. Emphasizes the role of swimming coaches, administrators, and innovators in fostering collaboration, adopting shared standards, and preparing the sport to use AI responsibly.
• Section 6: Opportunity—Setting the Stage for a Unified Framework Explores how combining technological advancements with training zone frameworks can support standardized, shareable, and actionable data for coaches and athletes.

Section 4: The Current State of Swimming Training Session Data Management

To build useful AI/ML solutions in swimming, we first need to understand the challenges of gathering, storing, and using training session data in real-world settings. This section analyzes the current state of swimming data management through eight core pillars of high-quality data. These pillars are adapted from the broader AI/ML data-quality literature and applied here to swimming practice (Priestley et al., 2023; Zhou et al., 2024; Polyzotis et al., 2018).

1. Intrinsic Quality Intrinsic quality refers to the accuracy, consistency, and completeness of raw data values. In swimming, this quality is often compromised by fragmented record-keeping methods and the complexity of digitizing existing training logs. For example, session plans may be handwritten on notebooks or saved as spreadsheets using coach-specific shorthand, which introduces errors during digitization. Additionally, important metrics like lap times, stroke counts, heart rates, and measurement units for distance, volume, and intensity are sometimes missing or vaguely recorded. Without precise and complete data, AI models may struggle to identify meaningful patterns, weakening their value as decision-support tools (Priestley et al., 2023; Rangineni, 2023).

2. Contextual Quality Contextual quality ensures that data is relevant, timely, and suitable for the specific task at hand. Many swimming session plans are generic and fail to consider individual athlete needs, such as age, gender, injury history, or skill level. This lack of specificity limits an AI system's ability to customize recommendations across swimmer profiles. Ambiguity in descriptions like "build effort" or "as fast as possible" further complicates intensity analysis. Similarly, failure to record the training phase, such as off-season, peak competition, or recovery, removes important temporal context. For AI insights to be useful, data must reflect the athlete's current state, the training goal, and the intended use of the analysis (Priestley et al., 2023; Zhou et al., 2024).

3. Representational Quality Representational quality deals with how well data is formatted and structured for easy interpretation. Inconsistent representation of session details across teams, such as the use of abbreviations like "DKOB", "OUS", "UK", or "choice" among a myriad of others, can lead to confusion. Moreover, session plans often include nested sets or intervals, which are difficult to capture in flat formats like spreadsheets. Without a standardized data schema, important hierarchical relationships between warm-ups, main sets, complementary sets, and cool-downs may be lost. Poor representational quality limits AI's ability to analyze how different components of a workout relate to the intended training stimulus (Priestley et al., 2023; Bompa & Haff, 2009; Riewald & Rodeo, 2015).

4. Accessibility Accessibility ensures that data is easily available to authorized users while maintaining security and privacy. One major challenge in swimming is fragmented data storage, with session logs often spread across personal notebooks, apps, and spreadsheets. This lack of centralization creates data silos, preventing comprehensive analysis. Furthermore, session descriptions are created in different languages, and may contain inconsistent terminology, making it difficult for AI tools to interpret them accurately. Improving accessibility requires secure, shared environments where data can be used by coaches, scientists, and athletes without ignoring privacy, consent, or governance requirements (Zhou et al., 2024; Qayyum et al., 2020).

5. Data Lifecycle Management Data lifecycle management involves tracking data from its creation to its eventual analysis, feedback, and archiving. In many programs, key metrics like in-session and post-session heart rates or rest periods are collected but not consistently fed back into future planning. This feedback gap limits the potential for AI systems and coaching strategies to improve over time. In addition, warm-ups and cooldowns are often tracked with less precision than main sets, creating blind spots in workload and recovery monitoring. A closed-loop lifecycle management system helps preserve traceability and keeps new data connected to future planning decisions (Polyzotis et al., 2018; Priestley et al., 2023; Rangineni, 2023).

6. Ethical and Legal Compliance Athlete privacy, data ownership, and regulatory compliance are critical to maintaining trust in AI applications. Issues often arise when swimmers change teams or when minors are involved, raising questions about who owns the data and how it can be shared. Without clear guidelines, organizations may be hesitant to collaborate or pool their data for AI development. Robust privacy policies and informed consent processes can help mitigate these risks while fostering more responsible data-sharing practices (Qayyum et al., 2020; Zhou et al., 2024).

7. Continuous Monitoring and Improvement Given the dynamic nature of swimming data—new sensors, changing training programs, and evolving goals—continuous monitoring is important. However, many teams lack frameworks for regular data audits and improvements. Incomplete metrics and recurring data gaps go unnoticed, leading to unreliable analyses. Continuous monitoring protocols can help detect anomalies, such as implausibly short lap times or missing rest data, and refine data-collection methods accordingly. This iterative approach helps maintain data quality as conditions evolve (Bangad et al., 2024; Polyzotis et al., 2018; Zhou et al., 2024).

8. Integration of Domain Knowledge Integrating domain expertise helps AI systems interpret ambiguous or complex data more appropriately. Coaches, sports scientists, and athletes provide critical insights that AI alone cannot capture. For example, terms like "build", "feel power", or "cruise" may carry different meanings depending on the swimmer's level or training context. Without expert input, AI models risk misinterpreting such instructions. Collaboration with domain experts helps keep AI-generated recommendations aligned with practical coaching principles, making them more useful as decision support (Priestley et al., 2023; Neutatz et al., 2021; Vec et al., 2024).

By analyzing the current challenges in managing swimming training session data, we can identify where improvements are needed to build AI-ready datasets. From standardizing data formats and contextualizing metrics to centralizing data storage and fostering collaboration, addressing these challenges will help bridge the gap between raw performance data and actionable AI insights.

Section 5: So, Is the Swimming Industry Data Fit for AI?

After exploring the importance of data quality in Part 1 and the current state of swimming session data management in Part 2, we return to the central question: Is the swimming industry ready to use AI well?

The short answer is: not yet, but it has a clear path forward.

The Importance of Coach Leadership Coaches are the gatekeepers of swimming training data. As the primary creators and custodians of training plans, their role is pivotal in driving data-driven advancements. By adopting standardized intensity zones, well-structured session plans, and comprehensive outcome tracking, coaches lay the foundation for more accurate, higher-quality data collection. With this stronger data backbone, AI/ML tools can better support technique feedback, workload interpretation, and long-term planning while still requiring human review and coaching judgment (Vec et al., 2024; Leckey et al., 2025).

Scaling Benefits for All Stakeholders When the swimming industry aligns around high-quality, structured data, the benefits extend across all levels of the sport:

• Athletes: Receive better-contextualized training plans that reflect their individual goals, abilities, and responses, supporting performance development and risk-aware planning.
• Coaches and Clubs: Streamline the session planning process, reduce administrative burdens, and gain access to more useful performance analytics for individuals and teams.
• Organizations and Federations: Can pool appropriately governed and anonymized data across regions to fuel large-scale research, inform national training programs, and develop best practices for all levels of competition—from age-group events to elite international meets.

Keeping It Simple and Universal The key to success is designing simple, intuitive data structures that are easy to adopt while maintaining the depth of expert knowledge. This does not mean oversimplifying or losing valuable insights. Instead, it means making data collection and management accessible to all stakeholders. By using standardized terminology, consistent intensity zones, and well-defined data frameworks, coaches, athletes, and technology developers can collaborate within a common language that bridges expertise and technology.

Section 6: Opportunity—Setting the Stage for a Unified Framework

Our exploration of data quality challenges reveals a crucial insight: creating high-quality AI/ML solutions in swimming is not only about better sensors, computer vision, more detailed spreadsheets, or digitizing the workouts of the best swimmers in the world. The real opportunity lies in establishing a unified framework: a shared blueprint that standardizes how training sessions are planned, recorded, and analyzed. When swimming professionals and technology experts collaborate around common standards, they can make data richer, more reliable, and more reusable. This can benefit everyone from elite athletes chasing records to fitness enthusiasts seeking steady improvements.

A Shared Vision Despite the diversity in coaching methods and swimmer skill levels, high-quality data is a practical requirement for tracking progress, supporting risk-aware load monitoring, and improving training interpretation. Harmonizing the way key metrics are captured can reduce many of the data issues discussed in this series. These metrics include stroke counts, rest intervals, and intensity zones, and the issues include inconsistent terminology, lack of individualization, and incomplete rest and recovery data (Fernandes et al., 2024; Borresen & Lambert, 2009; Dudley et al., 2023).

This is not only a technology initiative. It is a bridge between sports science and data science. Coaches, sports scientists, and software developers each bring valuable expertise, helping the framework reflect the practical realities of daily training sessions while remaining technically sound and scalable.

Building on a Training Zones Framework Wise Racer has already taken steps toward standardization by introducing two key models:

1. The 9-Zone Performance Swimming Training Framework Designed for competitive athletes, this framework categorizes effort into nine zones, covering everything from low-intensity technique work to high-intensity sprints.
2. The 5-Zone Fitness Swimming Training Framework Built for fitness and recreational swimmers, this simplified system focuses on core intensity ranges, making it accessible to those who prioritize fitness improvements over competition.

These zone-based frameworks help swimmers, coaches, and stakeholders communicate effectively about intensity and effort. However, zones alone are not enough. For these frameworks to support consistent interpretation, they must be paired with standardized data-collection protocols. This means clear definitions of each zone, uniform methods for logging sets and intervals, and a consistent approach to capturing athlete-specific contexts such as injury history or training phases. With this structure, an athlete training in Zone 3 should represent the same intended stimulus category in the data. The athlete's actual response still needs individual testing, monitoring, and coaching interpretation (Fernandes et al., 2024; Borresen & Lambert, 2009).

The Path Forward In upcoming blog posts, we will provide guidelines for implementing a unified framework. This includes structuring training plans, logging session outcomes, and using data to adjust coaching decisions. We will also explore how consistent, high-quality data can strengthen AI/ML tools in swimming by enabling:

• Feedback Loops: Analysis of rest intervals, stroke efficiency, and heart-rate data to help coaches fine-tune training decisions.
• Predictive Analytics: AI models that may help flag patterns related to plateau risk, fatigue, or poor load decisions, provided the outputs are interpreted with appropriate caution.
• Individualized Recommendations: Decision-support systems that adapt planning suggestions to an athlete's personal thresholds, whether they are a youth swimmer or a triathlete focused on long-distance open-water events.

Benefits for Institutions, Parents, Coaches, and Swimmers A structured, technology-driven approach to swimming training benefits every stakeholder in the ecosystem:

• Enhanced Personalization: By combining standardized zones with precise, athlete-specific data, coaches can tailor training sets and intensities to individual needs while monitoring response and recovery.
• Efficient Workload Management: Improved tracking of rest, recovery, and workload data helps coaches reduce blind spots in cumulative load management.
• Easier Progress Tracking: With a unified data format, tracking a swimmer's progress over weeks or seasons becomes more straightforward, offering both coaches and parents a clearer view of performance trends.
• Collaborative Advancement: When multiple clubs, regions, or federations adopt similar frameworks, they can share and compare appropriately governed, aggregated insights. This collaboration can spur innovation and support better standards across the sport.

A Modernized Swim Culture A shared vision for data management, combined with standardized frameworks like Wise Racer's 9-Zone and 5-Zone models, can improve how swimming is taught, trained, and evaluated. By adopting a common data structure and aligning around effective training principles, the swimming community can create a more informed, inclusive, and dynamic environment. This can support performance development while also fostering long-term engagement at all levels—from grassroots programs to elite international competition.

Summary

Part 2 offers both a reality check and a roadmap for progress. We examine the fragmented state of current session data management and show how this disarray limits effective AI adoption. However, the outlook is not bleak. We outline a hopeful path forward through harmonized data collection protocols, the integration of domain knowledge, and the application of Wise Racer's intensity zone frameworks. By addressing these gaps, the swimming community can move toward more useful AI/ML-assisted insights while preserving the role of coaches, athletes, and sport scientists.

But how do we design training sessions that meet the demands of the AI era? In the next installment, we will introduce our comprehensive training session framework, including the key considerations and design choices needed for AI-ready data. Then, in the final installment, we will show examples of how to apply this framework in practice and how training zones and session structures can work together to support meaningful improvements.

Call to action

This cannot be a solo effort. Swimming needs your support.

If you are a coach, athlete, data scientist, sports scientist, sports director, or simply passionate about swimming and your trade, and you would like to contribute to this conversation, please reach out. Your insights and expertise can help drive meaningful change.

If you care about the future of swimming, you can support this initiative by sharing this post and following Wise Racer on LinkedIn, Facebook, or Instagram. Together, we can build a smarter, data-driven future for the sport.

Note: This article was originally written in English and translated into other languages using automated AI tools so we can share this information with more people. We do our best to keep translations accurate and easy to understand, and we welcome help from the community to improve them. If anything in a translated version is unclear, incorrect, or differs from the English version, the original English text should be considered the official version.

Sources

Bangad, N., Jayaram, V., Sughaturu Krishnappa, M., Banarse, A., Bidkar, D., Nagpal, A., & Parlapalli, V. (2024). A theoretical framework for AI-driven data quality monitoring in high-volume data environments. International Journal of Computer Engineering & Technology, 15, 618-636. https://doi.org/10.5281/zenodo.13878755

Bompa, T. O., & Haff, G. G. (2009). Periodization: Theory and methodology of training. Human Kinetics.

Borresen, J., & Lambert, M. I. (2009). The quantification of training load, the training response and the effect on performance. Sports Medicine, 39, 779-795. https://doi.org/10.2165/11317780-000000000-00000

Dudley, C., Johnston, R., Jones, B., Till, K., Westbrook, H., & Weakley, J. (2023). Methods of monitoring internal and external loads and their relationships with physical qualities, injury, or illness in adolescent athletes: A systematic review and best-evidence synthesis. Sports Medicine. https://doi.org/10.1007/s40279-023-01844-x

Fernandes, R. J., Carvalho, D. D., & Figueiredo, P. (2024). Training zones in competitive swimming: A biophysical approach. Frontiers in Sports and Active Living. https://doi.org/10.3389/fspor.2024.1363730

Leckey, C., van Dyk, N., Doherty, C., Lawlor, A., & Delahunt, E. (2025). Machine learning approaches to injury risk prediction in sport: A scoping review with evidence synthesis. British Journal of Sports Medicine. https://doi.org/10.1136/bjsports-2024-108576

Neutatz, F., Chen, B., Abedjan, Z., & Wu, E. (2021). From cleaning before ML to cleaning for ML. Bulletin of the IEEE Computer Society Technical Committee on Data Engineering.

Polyzotis, N., Roy, S., Whang, S., & Zinkevich, M. (2018). Data lifecycle challenges in production machine learning: A survey. ACM SIGMOD Record, 47, 17-28. https://doi.org/10.1145/3299887.3299891

Priestley, M., O'Donnell, F., & Simperl, E. (2023). A survey of data quality requirements that matter in ML development pipelines. Journal of Data and Information Quality, 15. https://doi.org/10.1145/3592616

Qayyum, A., Qadir, J., Bilal, M., & Al-Fuqaha, A. (2020). Secure and robust machine learning for healthcare: A survey. IEEE Reviews in Biomedical Engineering. https://doi.org/10.1109/RBME.2020.3013489

Rangineni, S. (2023). An analysis of data quality requirements for machine learning development pipelines frameworks. International Journal of Computer Trends and Technology, 71, 16-27. https://doi.org/10.14445/22312803/IJCTT-V71I8P103

Riewald, S. A., & Rodeo, S. A. (Eds.). (2015). Science of swimming faster. Human Kinetics.

Vec, V., Tomazic, S., Kos, A., & Umek, A. (2024). Trends in real-time artificial intelligence methods in sports: A systematic review. Journal of Big Data. https://doi.org/10.1186/s40537-024-01026-0

Zhou, Y., Tu, F., Sha, K., Ding, J., & Chen, H. (2024). A survey on data quality dimensions and tools for machine learning. 2024 IEEE International Conference on Artificial Intelligence Testing, 120-131. https://doi.org/10.1109/AITest62860.2024.00023