Archinisis Blog: Mastering Acceleration Data: From Raw Data to Insights

Imagine having a tool that goes beyond what your eyes can see and your instincts can sense. A tool that gives you a clear, objective window into exactly how your athletes move the boat. Acceleration data does just that. It pinpoints where acceleration is created, where it's lost, and how consistent each stroke truly is. This means you can diagnose technical issues with precision, track progress with hard numbers, and deliver feedback that directly translates to faster, more efficient rowing. With acceleration analytics, you’re not just collecting data, you’re unlocking a new level of coaching.

Your crew just completed the Basel Head regatta, a 6 km race, the only thing you have left, is exhaustion. What if, you also had access to a wealth of performance data? That's where Archinisis' sensor comes into play. Just mount the sensor on the boat, so that while you row, the sensor collects. You can then interpret different metrics to improve your performance, one of them: acceleration.

The acceleration is directly measured with the accelerometer integrated in the inertial measurement unit (IMU), 200 times per second. It is crucial to have a high sampling frequency to ensure to have sufficient time resolution to see small and rapid changes in the boat movement.

This metric highlights the most critical phases of the stroke, particularly around the catch and during the drive phase, where the boat experiences the highest acceleration. By focusing on acceleration data, you can break down your rowing data into manageable, actionable insights. By the end of this article, you'll be able to interpret three fundamental acceleration graphs: continuous boat acceleration, acceleration pattern, stroke pattern. Understanding these will allow you to make sense of your data and turn numbers into real improvements on the water.

This 6 km regatta is distinguishable in three parts: an upstream start, a turn, and a downstream finish (more details). As opposed to the analysis of speed data, the stream direction does not affect the acceleration. The upstream and downstream part showcase the exact same acceleration pattern as can be seen in the example figure below. If pattern change, something went "wrong".

Continuous Boat Acceleration Explained

Note: for the following explanations we used a different data set of an M4+ boat not rowing at race pace but for which we had the corresponding synchronized video as a side view available for better and consistent explanation.

Annotated screenshot explaining different rowing events and phases in the boat's acceleration and speed data

Graph 1: Continuous boat acceleration (top, red line) and speed (bottom, blue line) for five strokes at 28.5 strokes per minute. The red shaded areas mark the check phase. Usually, at higher rates, the boat reaches peak speed at the start of the check and minimum speed at the end of the check. Therefore, the check phase is the "crucial" phase for optimizing speed loss, and, thus, rowing efficiency. The catch happens around the minimum of the acceleration around the middle of the check phase and marks the start of the drive phase. The drive phase ends at the finish, when the blades leave the water. The rest of the stroke is the recovery phase until the next catch happens. Note that the check phase covers part of the recovery and drive phases.

This graph shows the continuous boat acceleration throughout the whole race. The horizontal black line represents zero acceleration. When the acceleration curve is above this zero line, it indicates that the boat is accelerating, in other words picking up speed. And so, when the red line dips below the zero line, the boat is decelerating or losing speed.

Each complete cycle in the graph corresponds to one full rowing stroke. In rowing, the "check" phase refers to a brief but noticeable deceleration of the boat that occurs just before the catch. This check happens as the rower's momentum shifts. Minimizing check is crucial, as excessive deceleration wastes energy and slows the boat.

This graph is a great tool to pinpoint strokes that stand out and how each phase of a stroke causes acceleration and deceleration.

The catch: The rower's blade enters the water, legs fully compressed and ready to drive. Just before this moment, the boat decelerates as it glides forward without applied force. Once the catch occurs and the legs push off, acceleration shifts from negative to positive, marking the start of the drive.
The drive: The legs extend powerfully, producing the steepest acceleration.
The finish: As the stroke finishes and the blade exits the water, acceleration tapers off briefly reaching negative acceleration.
The recovery, part 1: The goal is to maintain speed and reduce losses. At lower rates, the boat continues gliding for a split second while the athletes briefly hold still. The rower slides forward smoothly and is able to accelerate the boat again because of the shift of center of mass.
The recovery, part 2, and the check phase: The rower slides forward smoothly to avoid excessive deceleration followed by a fast and well-balanced knee and hip flexion to ensure a controlled speed loss before the next stroke begins.

Continuous Boat Acceleration from Basel Head

Graph showing selected strokes during the upstream part at Basel Head

Graph 2: Continuous boat acceleration from the Basel Head regatta.

This figure is a screenshot showing a few strokes from the upstream part of the Basel Head regatta rowing at race pace with a stroke rate of around 35 strokes per minute. The strokes are remarkably consistent: each stroke produces a similar increase in acceleration, and the intervals between strokes remain steady. This regularity suggests that the crew is able to maintain a controlled, efficient pacing. Although it's from a different boat class (M8+ instead of M4+ from before) the pattern is very similar to the example from before. Can you identify the different phases and describe which part of the pattern changed because of the increased stroke rate?

Acceleration Pattern

Graph 3: Average acceleration pattern comparing the upstream (blue) and downstream (yellow) parts. It is computed by cutting the continuous acceleration at each check start until the subsequent check stop (i.e., it shows one complete stroke plus an additional check phase) and taking the average acceleration for 11 strokes centered at the distances indicated in the legend. Here at 1300m (upstream) and 4300m (downstream). The shaded area around each thick line marks the standard deviation of the eleven strokes. The time is set at zero at the end of the check which allows to read the stroke duration (time at which the average curve ends, here just under 1.7 seconds) and check duration (the negative time to the very left of the plot when the lines go below zero, here at around -0.45 seconds).

This graph represents the average acceleration per stroke (please refer to the figure legend for an explanation how the curve was computed). I briefly addressed the stream's impact, or lack thereof, and this graph illustrates my point clearly. The relation between the stroke time (x-axis) and the acceleration (y-axis), for the downstream and upstream segment, created two curves that are virtually indistinguishable, lying directly atop one another. This demonstrates that the way the boat accelerates and decelerate during the rowing cycle is governed primarily by the rower's technique, not by the direction of the current. It shows the consistency in technique that the rowers have. They execute each stroke the same way, resulting in the same acceleration pattern.

The key takeaway from this graph is that undistinguishable acceleration curves indicate a highly consistent rowing technique. If the lines weren't laying on top of one another, that might indicate fatigue of technical issues that need attention, and you will need to investigate the rowing pattern (see next chapter below). A good rowing technique is represented in the shape of the curves: they should be smooth and look "elegant". What exactly this means is still unknown and subject to research.

Stroke Pattern

Graph 4: Stroke pattern and acceleration colorbar indicating the coloring of the acceleration. The two thin shaded lines in the background mark the strokes that were used to compute the average curves above. The blue line marks the upstream segment, the yellow line marks the downstream segment.

This graph is a visualization of rowing stroke acceleration patterns over the almost 700 strokes from the entire Basel Head regatta. The horizontal axis represents stroke time in seconds, starting just before the catch and extending through the drive and recovery phases of each stroke. The vertical axis shows stroke number, with earlier strokes at the bottom and later strokes at the top. Each row in the graph corresponds to a single stroke, and the color at each point within that row indicates the boat’s acceleration at that moment in the stroke, measured in meters per second squared.

The color scale on the right maps acceleration values: purple represents strong deceleration, green is near zero acceleration, and yellow indicates strong acceleration. This graph is a different way to visualize the continuous boat acceleration. The strokes are stacked one on top of the other to create a heatmap that reveals patterns and consistency in rowing technique.

Consistent vertical color bands indicate a stable and repeatable stroke pattern, while shifts or irregularities in the color distribution may signal fatigue, technical adjustments, or errors. For example, the prominent purple and yellow bands near the catch and finish show the moments of, respectively, greatest acceleration and deceleration in each stroke. This type of visualization helps rowers and coaches quickly identify trends and areas for technical improvement. The strange things happening just before stroke 400 mark the 180° turn in the middle of the race.

Thanks to this tool, it is remarkably easy to distinguish the upstream segment, the turn, and the downstream. In the acceleration pattern section, we discussed how stream direction does not impact acceleration. This graph illustrates that point from a different perspective: the colors remain consistently aligned.

Conclusion

By leveraging high-frequency acceleration data from Archinisis' sensor, you've gained a nuanced understanding of your rowing performance. Through the three key graphs - continuous boat acceleration, acceleration pattern, and stroke pattern - you've been able to break down each phase of the rowing stroke, pinpointing strengths, and identifying areas for improvement.

The continuous boat acceleration graph is a depiction of the continuous boat acceleration. This graph is a key tool to analyze and interpret specific and individual strokes throughout the race.

The acceleration pattern graph examines the pattern generated from the data of each individual stroke, highlighting the consistency across the two portions of the race. This graph is valuable for identifying how well the crew maintains a steady rhythm.

The stroke pattern graph is a different way to visualize the continuous boat acceleration. This graph makes it easy to immediately pinpoint which strokes where different, how different they were, and in which rowing phase weakness occurred.

Analyzing acceleration curves helps you understand not just how fast you’re moving, but why and when your speed changes, offering insights into the effectiveness of the athletes’ technique. In the end, using acceleration to set clear goals and fine-tune your coaching can lead to real, measurable improvements in both technique and overall race results.

Now you have a good grasp about how to read and interpret boat acceleration data. If you have not yet done, we recommend to read the article about boat speed data.

Blog Post: Understand Speed Data