The Evolution of Pitching Analysis in Modern Baseball

Pitching mechanics have long been the subject of intense scrutiny, but the tools available to coaches and analysts today are light-years beyond the naked eye and handheld stopwatch of previous eras. For elite pitchers like Jacob deGrom, who relies on a combination of extreme velocity, sharp movement, and repeatable delivery, advanced video analysis is not a luxury—it is a necessity. High-speed cameras capture motion at thousands of frames per second, while machine learning algorithms parse terabytes of data to identify patterns invisible to even the most experienced coaches. This article explores how these technologies are used to refine deGrom’s mechanics, reduce injury risk, and sustain his performance at the highest level.

DeGrom’s career has been defined by both dominance and time spent on the injured list. Between 2019 and 2023, he made only 68 starts due to forearm, shoulder, and back issues. Advanced video analysis has become central to his rehabilitation and maintenance programs, allowing trainers to detect mechanical drift before it leads to breakdown. By examining every angle of his delivery, from the moment his front foot lifts to the instant of release, analysts can prescribe targeted adjustments that keep his arm in a safe, efficient position.

The Biomechanics of Jacob deGrom’s Delivery

Understanding why video analysis is so valuable begins with appreciating the complexity of deGrom’s motion. He generates immense power from his lower half, driving off the rubber with explosive hip rotation. His arm slot is nearly over-the-top, which helps him create downhill plane and ride on his four-seam fastball. However, this aggressive style also places high stress on the ulnar collateral ligament and rotator cuff.

Using high-speed cameras, analysts have measured deGrom’s maximum elbow flexion at nearly 140 degrees during the late cocking phase. They track his trunk tilt, shoulder external rotation, and the exact timing of his forearm pronation. Every degree matters: a change of even a few degrees in shoulder abduction can alter release point and increase torque on the elbow. Video analysis allows the coaching staff to quantify these angles with sub-millimeter accuracy and compare them across starts, rehab sessions, and even individual bullpen pitches.

The Role of High-Speed Cameras

High-speed cameras such as the Edgertronic and Phantom series are staples of professional baseball analysis. They record at 1,000 to 10,000 frames per second, freezing the ball in flight and capturing the pitcher’s body at the exact moment of maximum stress. For deGrom, cameras placed at multiple angles—side, front, and elevated behind home plate—provide a 360-degree view of his delivery.

These cameras are synchronized with motion-capture systems like KinaTrax or the Xsens suit, which track reflective markers placed on key joints. The combined data is imported into software such as Blast Motion or Rapsodo, which calculates metrics like arm speed, spin rate, and release efficiency. Analysts can then overlay a model of deGrom’s ideal mechanical profile onto his actual performance, highlighting deviations in real time.

Key Metrics from Video Analysis That Drive Improvement

While raw velocity and spin rate are obvious performance indicators, video analysis reveals subtler metrics that are critical for sustained success. Here are the primary parameters monitored for deGrom:

  • Stride length and direction: DeGrom’s stride is typically around 90% of his height. Video tracking determines whether his front foot lands closed, open, or square to home plate, which directly affects hip-shoulder separation and arm path.
  • Arm slot consistency: Even a 2-degree shift in arm slot can change the movement plane of his fastball and slider. Frame-by-frame analysis checks that his hand position at the top of the leg kick matches his established baseline.
  • Hip-shoulder separation: This is the angular difference between the hips and shoulders at foot strike. Greater separation correlates with higher velocity, but excessive separation can increase elbow stress. DeGrom’s optimal separation is around 40 degrees.
  • Elbow flexion at foot strike: Pitchers with a more flexed elbow (above 100 degrees) often generate more velocity but also more valgus torque. Video analysis helps maintain a safe range—typically 80–100 degrees for deGrom.
  • Release point variability: A consistent release point on all pitches (fastball, slider, changeup) is essential for deception. Analysts map his release height, horizontal position, and distance from the rubber across consecutive outings.

These metrics are not just numbers on a screen. They are diagnostic tools that guide every aspect of deGrom’s training. For example, if video shows his stride is tracking too far to the first-base side, his pitching coach can immediately cue him to “stay closed” or “land more towards home.” Without video, such a subtle fault might go unnoticed for weeks.

Case Study: Correcting DeGrom’s Mechanical Drift During the 2022 Season

After returning from a shoulder injury in 2022, deGrom’s fastball velocity was still elite (98–101 mph), but his command was erratic. Video analysis revealed that his front shoulder was flying open early, causing his arm to drag and his release point to rise. High-speed clips from his pre-injury outings were overlaid on current footage, showing a visible difference in his trunk lean.

The coaching team used software to create a side-by-side comparison with colored tracking lines. They identified that deGrom’s lead hip was not rotating fully, which forced his upper body to compensate. The fix was a series of drills emphasizing hip drive and staying stacked over the back leg through the stride phase. After three weeks of targeted work, his command returned to All-Star levels, and his slider’s horizontal movement increased by two inches.

This case illustrates the power of video analysis to not only detect problems but also to prescribe specific, measurable corrections. Without the side-by-side overlay and quantified angles, the mechanical drift might have been misattributed to loss of arm strength rather than a flaw in the lower half.

Injury Prevention Through Motion Capture and Wearable Tech

Preventing injuries is arguably more important than optimizing performance for a pitcher with deGrom’s injury history. Video analysis is now integrated with wearable sensors that measure joint angles and muscle activation during bullpen sessions. The combination provides a holistic picture of stress distribution throughout the kinetic chain.

For example, motion capture systems can calculate the valgus torque on his elbow at ball release. If that number exceeds a safe threshold (typically above 70 Nm for elite pitchers), the staff will adjust his workload or modify his mechanics. Research from the American Sports Medicine Institute (ASMI) has shown a direct link between elevated elbow torque and UCL tears, and video analysis is the primary means of monitoring that torque in real time. Research from ASMI underscores that early detection of torque spikes can prevent catastrophic injuries.

Another key application is monitoring shoulder horizontal abduction during the early arm acceleration phase. If deGrom’s arm drifts horizontally beyond a safe range (commonly 10–15 degrees of hyperabduction), it increases the risk of impingement and labral tears. Video analysis flags these deviations instantly, allowing the strength and conditioning staff to implement corrective exercises such as external rotation pulls and scapular stabilization drills.

The Role of Artificial Intelligence in Predictive Analysis

Artificial intelligence (AI) has moved beyond detecting patterns to predicting injury likelihood. Machine learning models trained on thousands of pitcher deliveries can identify movement signatures that precede injury. For deGrom’s team, AI tools scan his video data for warning signs like increased trunk tilt, reduced knee flexion at foot strike, or delayed arm acceleration.

A 2023 study published in the Journal of Orthopaedic & Sports Physical Therapy found that AI models could predict elbow injuries with over 85% accuracy based on video-derived metrics alone. Read the study here. By incorporating these predictions into daily training decisions, deGrom’s team can proactively reduce his pitch count, alter his off-day routine, or address minor mechanical issues before they become serious.

Integrating Video Analysis with Strength and Conditioning

Video analysis does not exist in a vacuum; its findings must be translated into actionable training modifications. For deGrom, this means close coordination between the pitching coach, the strength coach, and the analytics staff. When video reveals a loss of hip-shoulder separation, the strength team might add rotational core exercises like cable chops and landmine rotations. If the arm slot is dropping, they may focus on shoulder mobility drills to maintain external rotation range.

One cutting-edge method is real-time biofeedback. Using a system like Motus, which measures arm speed and elbow torque via a sleeve sensor, deGrom can see his metrics on a tablet immediately after each pitch. If the torque is too high, he can adjust his mechanics before the next pitch. This instant feedback loop accelerates learning and reduces the time needed to ingrain new movement patterns.

Furthermore, video analysis is used to evaluate the effectiveness of new training equipment. For example, weighted ball protocols or plyometric drills can be analyzed frame-by-frame to ensure they are not causing compensation patterns. If a drill induces a sidearm release or a shortened stride, it is modified or dropped.

Technologies Powering the Analysis

The following tools are central to the advanced video analysis pipeline for deGrom and other elite pitchers:

  • Edgertronic Cameras: Capable of 1,000–10,000 fps, these capture the minutiae of delivery mechanics, including ball spin axis and seam orientation.
  • KinaTrax Optical Motion Capture: Uses reflective markers and multiple infrared cameras to build a 3D skeleton model of the pitcher, providing joint angles and segment velocities.
  • Rapsodo: Combines radar and camera data to measure spin rate, spin axis, velocity, and release point. It is often used in conjunction with high-speed video for a complete picture.
  • Blast Motion: A wearable sensor attached to the back of the hand that tracks wrist angle, bat speed (for hitters), and arm whip. For pitchers, it measures arm speed and acceleration patterns.
  • EyeTracking Systems: While not yet standard, eye-tracking cameras can measure a pitcher’s visual focus during delivery and help optimize head position for better command.

These technologies produce massive datasets. A single bullpen session for deGrom might generate 50 GB of video and sensor data. Dedicated analysts use custom software developed by teams like Driveline Baseball to sort, visualize, and interpret the information. Driveline’s biomechanics lab has worked with numerous MLB pitchers and publishes research that informs best practices across the league.

Challenges and Limitations of Video Analysis

Despite its power, video analysis is not without pitfalls. Overreliance on data can lead to “paralysis by analysis,” where a pitcher becomes overly focused on mechanical cues and loses the natural feel of his delivery. For deGrom, maintaining a balance between data-driven adjustments and trusting his innate talent is essential.

Another challenge is the variability of in-game conditions. Stadium lighting, camera angles, and the pressure of a live game can affect movement patterns. Therefore, video analysis from bullpen sessions must be validated with game-day footage. Differences in mound slope, rubber height, and weather (wind, humidity) can also influence mechanics, and analysts must account for these factors when drawing conclusions.

Moreover, video analysis is only as good as the expertise of the interpreter. A poorly calibrated camera or an incorrect marker placement can lead to misleading data. The coaching staff must be trained to distinguish between significant mechanical drift and normal variation. The adage “correlation does not imply causation” applies here: a change in arm slot might cause a drop in velocity, or it might be a symptom of fatigue from a previous outing. Contextual understanding is crucial.

Future Directions: What’s Next for deGrom and Baseball?

As deGrom continues his career into his late 30s, video analysis will play an even greater role in managing his workload and extending his prime. Emerging technologies such as inertial motion capture (using IMUs without external markers) will allow for analysis in any environment, including during games. This could provide real-time feedback to the dugout without the need for a lab setup.

Machine learning will become more sophisticated, potentially predicting the ideal pitch mix based on an opponent’s swing weaknesses and a pitcher’s current mechanical state. For example, if deGrom’s arm slot is slightly lower on a given day, the system might suggest throwing more sliders because the pitch will have a different plane than usual—and the model will know that swing-and-miss rates increase under those conditions.

Virtual reality (VR) also holds promise. Pitchers could wear VR headsets to simulate hitters from specific teams while their mechanics are recorded. This would allow for safe, repetitive practice of new mechanical adjustments without the stress of a live arm. Win Reality is already used by many MLB players for hitting, and similar systems for pitching are in development.

Finally, the integration of video analysis with biometric data—heart rate, cortisol levels, sleep patterns—will create a truly holistic picture of an athlete’s readiness. For an aging elite pitcher like deGrom, the goal is to train smarter, not harder. Advanced video analysis is the cornerstone of that smarter training.

Conclusion

Jacob deGrom’s ability to throw 100 mph with pinpoint control is the result of a finely tuned mechanical engine. Advanced video analysis gives his team the ability to maintain that engine, diagnose problems before they cause breakdowns, and make targeted improvements year after year. From high-speed cameras that capture the subtle flexion of his elbow to AI models that predict injury risk days in advance, the technologies described here are transforming how elite pitchers are developed and maintained.

While no system can eliminate injury entirely, the combination of video analysis, wearable sensors, and machine learning offers the best defense against the physical demands of professional pitching. For deGrom, this means more time on the mound, more dominant performances, and a legacy that continues to grow. As the tools evolve, the relationship between pitcher and data will only deepen—ushering in a new era of performance optimization in baseball.