📘Upside Analysis: Why More Elite Teams May Move From GPS + Biomarkers to Optical Tracking + Biomarkers

Introduction

For more than a decade, GPS/GNSS plus biomarkers has been one of the most common approaches to athlete monitoring in elite team sports.

GPS provides an increasingly sophisticated picture of external load: distance, speed, acceleration, deceleration, high-speed running and other movement metrics.

Biomarkers provide a complementary picture of internal response: inflammation, oxidative stress, muscle damage, nutritional status, hormonal status and other physiological indicators.

Together, they provide something neither technology can provide independently:

What did the athlete do — and how did the athlete respond?

However, a new architecture is emerging.

As optical tracking systems become more accurate, more widely deployed and increasingly capable of capturing not only player position but also movement patterns, skeletal information and contextual game data, teams have a new option:

Optical tracking + biomarkers.

This is particularly relevant in sports where optical tracking infrastructure is already installed in arenas or stadiums and where wearable technologies cannot be used, are inconvenient, or provide only a partial view of competitive demands.

The transition is therefore unlikely to be:

GPS → optical tracking

Instead, the more realistic evolution is:

GPS + biomarkers → optical tracking + GPS + biomarkers → optical tracking + biomarkers + contextual data + AI

The question for teams is no longer simply which tracking technology is more accurate.

The bigger question is:

Which combination of data gives performance, medical and coaching staff the clearest picture of what happened, how the athlete responded, and what should happen next?

---

1. Key Trends

Trend 1 — Optical tracking is moving from “analytics” to “athlete monitoring”

Optical tracking was initially associated primarily with tactical analysis, broadcasting and fan engagement.

That is changing.

Modern optical systems can provide:

• player position

• speed

• acceleration/deceleration

• distance

• spacing

• movement trajectories

• changes of direction

• positional context

• tactical behavior

• skeletal/pose information

• movement signatures

This makes optical tracking increasingly relevant to performance departments rather than only analysts.

The 2026 research literature is also becoming more sophisticated. A recent Journal of Sports Analytics study evaluated commercial computer-vision tracking systems against TRACAB and showed that computer-vision tracking can provide useful positional and speed data, although accuracy varies substantially by provider, camera feed and whether players are successfully detected.

This is important: optical tracking is becoming more viable, but teams should not treat all computer-vision systems as equivalent.

---

Trend 2 — Competition data is increasingly becoming “wearable-free”

One of the strongest arguments for optical tracking is simple:

The athlete does not have to wear anything.

That matters particularly in:

• NBA

• WNBA

• NHL

• basketball

• some football environments

• other indoor sports

The NBA provides a good example. League-wide 3D optical tracking has created a competition environment in which player movement can be captured without relying on individual wearable devices.

Picture: OnTraq’s 3D markerless solution by Qualisys.

The WNBA has also adopted league-wide 3D optical tracking.

This creates an interesting opportunity for performance departments: The competition data can be captured automatically, while biomarkers can provide the athlete-specific physiological response.

---

Trend 3 — Teams increasingly want context, not just workload

GPS answers questions such as:

How far did the player run?

Optical tracking can increasingly answer:

What was the player actually doing?

For example:

• Was the player repeatedly changing direction?

• Was the player operating in unusual areas of the field?

• Did movement patterns change late in the game?

• Did a player’s normal movement signature change?

• Was the player exposed to unusual acceleration/deceleration sequences?

• Was the athlete repeatedly involved in specific tactical situations?

• How did the player’s movement change relative to previous games?

This is a fundamentally different level of information.

GPS produces a physical workload profile.

Optical tracking can potentially produce a movement behavior profile.

That distinction may become one of the most important developments in athlete monitoring over the next five years.

---

Trend 4 — Biomarkers become more valuable when external load becomes richer

Biomarkers are often most useful when they are interpreted in context.

A high CK level, for example, means something different depending on:

• recent training load

• match exposure

• travel

• sleep

• previous injury

• position

• training phase

• individual baseline

The same applies to inflammation or oxidative-stress markers.

Picture: Orreco’s biomarker solution

If the external-load dataset becomes more granular through optical tracking, teams can potentially ask more sophisticated questions:

Did this athlete’s physiology change after a specific type of movement exposure?

Rather than simply:

Did this athlete run a lot?

That creates a much more interesting research opportunity.

---

2. Key Use Cases

2.1 Injury-risk monitoring

The traditional model has often been:

GPS load → algorithm → injury-risk score

The new model can become:

Movement behavior + physiological response + medical history → individualized risk signal

Optical tracking potentially provides information about movement deviations that GPS cannot capture.

For example, an athlete may maintain similar total distance and sprint numbers while subtly changing:

• cutting behavior

• acceleration patterns

• turning patterns

• movement symmetry

• movement smoothness

• preferred movement sequences

This is one of the areas being explored by Orreco’s Motion Signal platform.

Orreco reports that its early research identified 85% of subsequent hamstring injuries with 86% specificity using individualized movement signatures. Those findings should be treated as early/vendor-reported evidence rather than established clinical evidence, but they illustrate the direction of the market.

The important development is not the percentage itself.

It is the concept: movement deviation may contain information that conventional workload metrics miss.

---

2.2 Return to performance

Optical tracking could become particularly valuable during return-to-play/return-to-performance.

Instead of simply asking whether an athlete has completed a certain amount of running, teams could compare:

Current movement behavior vs. historical competition movement behavior.

For example:

• acceleration profile

• deceleration profile

• cutting

• sprint mechanics

• curvilinear running

• movement volume

• movement variability

• position-specific movements

Biomarkers could then provide an additional layer describing the athlete’s physiological response.

This creates a potential three-layer RTP model:

Movement capacity → physiological response → competitive exposure

---

2.3 Post-game recovery decisions

Immediately after a game, performance staff could combine:

External load

Optical tracking

Internal response

Biomarkers

Subjective response

Wellness/RPE

Context

Minutes, travel, previous workload, injury history

The objective is not to create another dashboard.

It is to answer:

Who needs attention tomorrow morning, and why?

---

2.4 Individualized conditioning

Optical tracking can help teams reproduce the actual movement demands of competition.

For example:

“This player needs more high-speed linear running.”

is very different from:

“This player needs to reproduce the specific acceleration, deceleration and change-of-direction profile observed in the final 15 minutes of competition.”

The second approach potentially creates much more sport-specific conditioning.

---

2.5 Tactical-performance integration

One major advantage of optical tracking is that physical data can be connected to tactical context.

A team could potentially analyze:

movement + tactical role + physiological response

For example:

• pressing workload

• defensive coverage

• transition workload

• repeated high-intensity actions

• off-ball movement

• spacing

• changes in tactical role

This is difficult to reproduce using GPS alone.

---

2.6 Scouting and recruitment

Optical tracking also creates an interesting bridge between performance and recruitment.

Teams can potentially compare:

• movement profiles

• peak velocity

• acceleration

• explosiveness

• change of direction

• positional behavior

• tactical fit

• physical output

across leagues and competitions.

This is particularly relevant to platforms such as SkillCorner, which provide computer-vision tracking data across large numbers of competitions.

The future could therefore connect:

Recruitment → performance → medical → biomarkers

rather than treating these as separate datasets.

---

3. Current Studies and Evidence

The evidence base is developing rapidly, but it is important to distinguish between three different categories:

1. Optical tracking validation

2. Biomarker/internal-load research

3. True multimodal studies combining movement and physiological data

The first two categories are relatively mature.

The third is still emerging.

Study 1 — GPS vs Optical Tracking in EPL

A recent study involving 26 professional English Premier League players compared 10-Hz GPS with a 25-Hz optical tracking system from Second Spectrum.

The systems were highly correlated, but optical tracking produced higher values for several metrics, including total distance, high-speed running and sprint distance.

The practical implication is important:

Teams cannot simply assume that GPS and optical numbers are interchangeable.

They need calibration and sport/team-specific conversion models.

---

Study 2 — GPS vs Optical Tracking in Professional Soccer

A 2023 study compared Catapult GNSS and Tracab optical tracking across 16 official matches involving professional players in the Polish Ekstraklasa.

Total distance showed strong agreement, but agreement became weaker for more specific measures such as high-speed and sprint counts.

Again, the conclusion is not that one system automatically replaces the other.

It is that measurement systems can produce materially different numbers even when they are attempting to measure the same physical phenomenon.

---

Study 3 — Computer Vision Using Broadcast Footage

A 2026 Journal of Sports Analytics study evaluated three commercial computer-vision providers using FIFA World Cup footage.

The study found meaningful variation between providers and configurations.

Position RMSE ranged from 1.68 to 16.39 meters and speed RMSE from 0.34 to 2.38 m/s depending on provider and configuration.

The tactical camera feed generally produced better detection.

This is an important warning for teams:

“AI optical tracking” is not a single technology category.

Teams need to evaluate:

• camera infrastructure

• detection rate

• positional accuracy

• speed accuracy

• occlusion handling

• extrapolation

• sampling rate

• latency

• skeletal accuracy

• consistency over time

---

Study 4 — External Load + Biomarkers

A 2026 Scientific Reports study followed 25 elite youth soccer players for three months and compared external load measured with a local positioning system with RPE, sleep, CK, LDH, CRP, cortisol, transferrin and other markers.

The results showed that relationships existed between external load and biomarkers, but subjective measures were generally more consistent than biomarkers.

This is extremely important.

It reinforces that biomarkers should not become another automatic red/yellow/green score. They should be interpreted in context.

---

Study 5 — Individualized Biomarker Reference Ranges

A recent study involving English Premier League players examined individualized adaptive reference ranges for longitudinal blood monitoring.

The research supports the idea that athlete-specific baselines can be more useful than simply comparing athletes with generic population reference ranges.

This is directly relevant to the optical + biomarker model:

The future is likely to be athlete-specific rather than population-based.

---

Study 6 — Internal + External Load and Neuromuscular Performance

A 2025 study of professional soccer players examined internal and external training load in relation to neuromuscular performance.

The authors concluded that practitioners should use a multi-metric approach rather than relying on a single load variable.

This supports the broader shift toward multimodal athlete monitoring.

---

Study 7 — Barcelona Multimodal Research

A 2026 preprint examined 26 FC Barcelona first-team players using:

• GPS training data

• blood-based inflammation profiling

• immune-age measurements

• injury records

The study is particularly interesting because it demonstrates where the market is heading: multiple physiological and workload data streams analyzed together.

However, because this is currently a preprint rather than a peer-reviewed publication, teams should treat its findings as preliminary.

---

4. GPS + Biomarkers vs. Optical Tracking + Biomarkers

The key conclusion is that optical tracking is not necessarily a better GPS.

It is a different type of measurement system.

GPS is particularly strong for portable, repeatable training monitoring.

Optical tracking is particularly strong for competition, spatial context and movement behavior.

Biomarkers then provide the internal physiological layer.

---

5. The Emerging Architecture

The most sophisticated teams may eventually operate something closer to:

Layer 1 — Competition movement

Optical tracking

Position
Speed
Acceleration
Deceleration
Movement signatures
Tactical context
Skeletal/pose data

Layer 2 — Training movement

GPS/GNSS / LPS / IMU

Portable training load
Rehabilitation
Field sessions
Travel sessions

Layer 3 — Internal response

Biomarkers

Inflammation
Oxidative stress
Muscle damage
Nutrition
Hormonal/physiological status

Layer 4 — Human response

Wellness + RPE

Sleep
Fatigue
Soreness
Stress
Readiness

Layer 5 — Context

Medical + competition + travel

Injury history
Minutes
Schedule
Travel
Return-to-play status

Layer 6 — AI

Individualized decision support

The ultimate objective should be:

What happened → How did the athlete respond → What does it mean → What should we do?

---

6. Future Trends

6.1 From player location to player movement signatures

The market is moving from:

Where was the player?

toward:

How does this player normally move?

This could become a major new category in sports medicine and performance.

---

6.2 From population thresholds to individual baselines

The future athlete monitoring system will increasingly understand:

“This is normal for this athlete.”

rather than:

“This is normal for athletes.”

That shift is particularly important for biomarkers and movement data.

---

6.3 From separate systems to multimodal platforms

Teams increasingly want one athlete record rather than five disconnected dashboards.

Optical tracking, GPS, biomarkers, force plates, wellness, sleep, medical records and video should eventually feed the same decision environment.

---

6.4 AI will increasingly connect movement and physiology

The next generation of models will not simply ask:

Does high workload increase injury risk?

They will ask:

Which combinations of movement behavior, physiological response, recovery status and context are associated with meaningful changes in this specific athlete?

That is a much more difficult problem — but potentially much more valuable.

---

6.5 Optical tracking will move closer to real-time

Today much of the value is retrospective or post-game.

The next evolution is:

live movement → live deviation detection → live intervention

However, this requires extremely reliable tracking and very low false-positive rates.

A risk signal that produces ten unnecessary interventions will quickly lose credibility with coaches.

---

6.6 Cameras will become shared infrastructure

This could be particularly important economically.

If a league, venue or arena already installs optical tracking infrastructure, multiple departments can use the same data:

• coaching

• performance

• medical

• scouting

• analytics

• broadcasting

• officiating

• fan engagement

That creates a very different ROI equation from buying individual GPS devices for every athlete.

---

7. Key Vendors

The market should be viewed as an ecosystem rather than a single vendor category.

Integrated optical + biomarker / athlete intelligence

Orreco

Orreco is currently one of the clearest examples of the convergence.

Its platform combines:

• biomarkers

• internal/external load

• GPS

• computer vision

• Motion Signal

• wellness

• medical information

• recovery

• women’s performance

• AI

Video: Orreco Motion Signal

Its Motion Signal product uses computer vision to analyze movement deviations, while its biomarker platform provides individualized physiological monitoring.

This makes Orreco particularly relevant to the emerging optical/motion + biomarker category.

---

Optical tracking

Genius Sports / Second Spectrum

Second Spectrum is deeply established in:

• NBA

• WNBA

• EPL

• MLS

• basketball

• soccer

Its technology provides optical player and ball tracking and increasingly skeletal/3D data.

Picture: Back in 2020, the MLS signed a partnership with Second Spectrum, which provided its proprietary optical tracking system to work for every MLS match.

The NBA and WNBA are particularly relevant examples of league-level optical infrastructure.

Here is an interview we conducted with Michael D’Auria EVP Partnerships at Genius Sports:

Hawk-Eye

Hawk-Eye provides optical tracking, skeletal tracking and biomechanics across multiple sports.

Its capabilities extend into:

• basketball

• football

• tennis

• rugby

• hockey

• biomechanics

Its markerless skeletal capabilities could become particularly interesting as teams move from simple positional tracking toward movement analysis.

TRACAB / EA

TRACAB remains a major optical tracking technology, particularly in football.

Its technology has been used as a reference system in validation studies comparing optical tracking with GPS.

Video: TRACAB’s Automatic Player Tracking in the NFL

SkillCorner

SkillCorner is particularly interesting for soccer because its AI-based tracking can derive player tracking from broadcast, tactical and wide-angle video.

This creates an important opportunity beyond stadium-based systems: optical tracking without requiring every club to install a dedicated camera infrastructure.

Video: SkillCorner

Sportlogiq

Sportlogiq uses computer vision and machine learning to derive tracking and contextual data from video.

It is particularly established in hockey and also operates across soccer and American football.

Video: Sportlogiq

---

Athlete-management / integration layer

ActionApps

Action Apps is important because it represents a different approach to the traditional athlete management system: giving teams greater ownership, flexibility and control over their data infrastructure. Built using Microsoft Power Platform, Azure and Microsoft 365, its AMS is designed to integrate information from multiple performance and medical systems rather than forcing teams into a rigid, standalone data environment.

Its platform can bring together:

• GPS and tracking data

• Wearables

• Video analysis

• Performance and testing data

• Injury and medical records

• Player availability

• Planning and scheduling

• Scouting and athlete development data

The platform also uses Power BI for analytics and visualization and includes AAVA, an AI agent that allows practitioners to query their data using natural language. Importantly, organizations can manage their data within their own Microsoft environment rather than having it reside solely within a traditional shared SaaS AMS architecture.

This model could become increasingly relevant as teams add optical tracking, biomarkers, wearables and other high-frequency data streams. The value may increasingly shift from simply collecting more athlete data to creating an infrastructure where teams can own, integrate, analyze and query those different datasets in one environment.

In an optical + biomarker model, platforms such as Action Apps could therefore serve as the integration and intelligence layer connecting external workload with internal physiological response—and helping practitioners turn those combined datasets into decisions rather than simply another dashboard.

Here is a video overview of ActionApps’ latest AI rehab tool:

Genetrainer

Genetrainer is particularly relevant to this shift because its platform is designed around individualization and the integration of multiple athlete data streams, rather than treating workload as a standalone metric. The platform can bring together GPS and video analysis, RPE and subjective data, wellness, strength & conditioning, medical and injury information, lifestyle/body metrics, third-party devices and genetics, while using AI to identify correlations, trends, outliers and potential areas requiring attention. Genetrainer also says it can create individualized load models by combining genetic information, age and training stress.

This type of approach becomes increasingly relevant in an optical tracking + biomarker ecosystem. Instead of asking only “How much load did the athlete accumulate?”, the objective becomes understanding how that specific athlete is responding to the load, using multiple sources of information and individualized baselines. As optical, biomarker, genomic, wellness and medical datasets expand, platforms such as Genetrainer could increasingly serve as an integration and intelligence layer, helping practitioners connect external workload with internal and individual characteristics and turn those relationships into more personalized performance and recovery decisions.

Catapult

Catapult remains a major GPS/athlete-monitoring platform, but its strategy also illustrates the convergence.

OpenField can now ingest optical tracking data such as TRACAB and NBA Hawk-Eye data, allowing teams to bring optical competition data into an existing athlete-monitoring environment.

This is potentially more important than a simple “GPS vs optical” comparison.

The future may be GPS platform + optical data + biomarkers, rather than replacing the GPS platform.

---

Biomarker specialists

Orreco

Orreco is particularly relevant to the optical tracking + biomarker model because its approach focuses on understanding an athlete’s internal physiological response, complementing the external workload captured by tracking systems. Orreco’s Advanced Biomarker Analytics uses blood testing and longitudinal athlete profiling to assess areas such as recovery, inflammation, immune function, iron status and other physiological factors, with the goal of translating laboratory data into individualized recommendations for practitioners.

Orreco As teams increasingly capture external load through optical tracking, platforms such as Orreco could provide the other side of the equation: optical tracking shows what the athlete did, while biomarkers help indicate how the athlete is responding internally. The potential value therefore lies not in another isolated dataset, but in combining external workload, internal physiology and an athlete’s individual baseline to inform recovery, training and return-to-play decisions. The most sports-specific integrated biomarker provider in this category.

InsideTracker

InsideTracker provides blood biomarker analysis and personalized health/performance recommendations and has offerings for professional and collegiate sports.

Its role is more focused on biomarker intelligence than optical tracking, making it a potential component of a broader multimodal ecosystem.

---

8. Recommendations to Teams

1. Do not replace GPS simply because optical tracking is available

The strongest architecture may be complementary.

Use:

GPS/GNSS for training

and

optical tracking for competition

when the environment supports it.

---

2. Start with the competition environment

Ask:

Can optical tracking give us information during competition that GPS cannot?

If the answer is yes, start there.

This is especially relevant in basketball and other sports where wearables may not be available during competition.

---

3. Calibrate before comparing

Teams should not assume:

10 km GPS = 10 km optical

Recent research shows that different systems can generate materially different values.

Establish team-specific conversion and benchmarking procedures before changing historical datasets.

---

4. Keep biomarkers individualized

Avoid generic:

“CK above X = bad.”

Instead:

“What is abnormal for this athlete?”

Longitudinal athlete-specific reference ranges are likely to be much more valuable.

---

5. Demand multimodal integration

When evaluating vendors, ask whether they can combine:

• optical tracking

• GPS

• biomarkers

• wellness

• sleep

• force plates

• medical history

• injury history

• video

• training plans

The value increasingly comes from the relationship between datasets, not the individual dataset.

---

6. Focus on decisions rather than dashboards

A team should be able to identify several concrete decisions that the system improves.

For example:

• Should this player train fully tomorrow?

• Should conditioning be modified?

• Does this athlete require medical follow-up?

• Is the player ready for return-to-performance?

• Did the player receive enough competition exposure?

• Does the next training session need to replicate a specific movement demand?

If the answer is simply:

“We get another dashboard”

the investment case is weak.

---

7. Run controlled pilots

Rather than replacing an existing GPS system immediately, teams should run a 3–6 month comparison.

Measure:

• GPS vs optical agreement

• data availability

• staff time

• false alerts

• actionable findings

• injury/availability outcomes

• coach adoption

• medical adoption

• athlete acceptance

The key KPI should not be the number of data points.

It should be: number of better decisions enabled by the system.

---

8. Evaluate the vendor’s evidence carefully

Ask vendors for:

• independent validation

• prospective validation

• sample size

• sport and population

• sensitivity

• specificity

• false-positive rate

• external validation

• data missingness

• camera occlusion performance

• model drift

• explainability

Teams should be particularly careful with injury-prediction claims.

A technically impressive model is not automatically a clinically useful decision-support tool.

---

Conclusion

Elite sports monitoring is entering a new phase.

The first generation was largely about collecting workload.

The second generation connected workload with internal response.

The next generation will increasingly combine:

movement + physiology + context + AI.

This makes optical tracking + biomarkers particularly interesting.

Optical tracking can provide a richer description of what happened during competition, while biomarkers can provide information about how the individual athlete responded internally.

But the future is unlikely to be a simple replacement of GPS.

The more likely model is:

**GPS for portable training and rehabilitation

• optical tracking for competition

• biomarkers for internal response

• wellness and medical data for context

• AI for individualized interpretation.**

The strategic opportunity for teams is therefore not to ask:

“Should we replace GPS?”

The better question is:

“What information are we currently missing that optical tracking can provide — and can we combine it with physiological data to make better decisions?”

That is where the next generation of athlete monitoring is likely to create value.

The winners in this market will not necessarily be the companies collecting the most data.

They will be the companies — and teams — that can transform movement and physiological data into better decisions, with fewer false alarms and a clear connection to player availability, performance and return-to-play.

Current studies / source URLs

These are the key sources I would attach to the published Upside version:

• EPL GPS vs. Optical Tracking: Interchangeability of external player load variables from different athlete tracking systems in English Premier League soccer players — directly compares GPS with Second Spectrum optical tracking. Study / full text

• 2023 GPS vs. Tracab: Assessing the agreement between a GNSS and an optical-tracking system for measuring total, high-speed running, and sprint distances in official soccer matches. Study / full text

• 2026 computer-vision validation: Concurrent validity of computer-vision artificial intelligence player tracking software using broadcast footage — particularly useful because it demonstrates that accuracy varies significantly by provider and video configuration. Journal of Sports Analytics study

• 2026 external load + biomarkers: On the relationship between external and internal load variables in elite youth soccer players — LPS combined with CK, LDH, CRP, cortisol, transferrin, CMJ, RPE and sleep. Scientific Reports study

• 2025 individualized biomarkers: Early warning system for player recovery? — longitudinal individualized biomarker reference ranges in EPL players. PubMed study

• 2025 internal/external load: Associations between internal and external training load measures and neuromuscular performance in elite soccer players. Springer study

• 2024 athlete monitoring review: Athlete Monitoring Systems in Elite Men’s Basketball — specifically discusses optical tracking, biomarkers and the need to integrate objective and subjective measures. Open-access review

• 2026 Barcelona multimodal preprint: Beyond External Load: Integrative Immune Monitoring Reveals Injury-Predictive Signals in the Athlete’s Internal State. This is a preprint, so I would clearly label it as preliminary evidence. Preprint

• Orreco Motion Signal research: Orreco’s published description of its movement-signature research, including the reported 85% sensitivity / 86% specificity result. I would label this company-reported early research, rather than peer-reviewed evidence. Orreco Motion Data Collective research

• Biomarker evidence: Blood Biomarker Profiling and Monitoring for High-Performance Physiology and Nutrition— useful for the limitations, sampling and interpretation section. PubMed / full text

Key vendor links

• Orreco — biomarker + motion/computer vision + athlete intelligence.

• Genius Sports / Second Spectrum — optical tracking across NBA, WNBA and other sports.

• Hawk-Eye Innovations — optical tracking, skeletal tracking and markerless biomechanics.

• TRACAB / EA — established optical tracking infrastructure.

• SkillCorner — scalable AI tracking from broadcast/tactical video.

• Sportlogiq — computer vision, tracking and contextual analytics.

• Kitman Labs — athlete monitoring/performance optimization and data integration.

• Catapult — OpenField integration of NBA Hawk-Eye and TRACAB optical data.

• InsideTracker — biomarker platform used across professional and collegiate sports.

You may also like: