Using video replay to improve match performance

A match video can reveal details that disappear in the noise of a competition day. The recording shows timing, field position, driver decisions, alliance coordination and robot behaviour in a way that memory rarely can. Used well, replay turns a disappointing result into practical evidence for the next training session.

For Australian teams, this approach is valuable when travel creates long gaps between events. A squad from Brisbane may spend months preparing for a tournament in Sydney, while a regional Victorian team may have limited access to specialist mentors between school terms. Video gives students a shared reference point, even when everyone cannot meet in the workshop.

The aim is not to criticise drivers or replay every match indefinitely. It is to identify a small number of repeatable actions that improve scoring, defence, reliability and teamwork. A clear review process also gives students ownership of their engineering decisions and communication habits.

Recordings from your own robot, official event streams and alliance partners can all contribute useful evidence. Keep the focus on learning, protect student privacy and connect every observation to a change that can be tested.

Replay focus Evidence to collect Practical response
Autonomous period Starting position, route, missed actions, delays Adjust code, sensors or setup procedure
Driving and scoring Cycle time, alignment, failed attempts Practise paths, controls and approach angles
Defence Blocked routes, lost possessions, recovery time Change spacing, communication or strategy
Robot reliability Brownouts, jams, loose parts, resets Inspect hardware and improve maintenance
Alliance play Role clarity, traffic, hand-offs Agree roles and callouts before the next match

Choose footage that answers a question

Before opening a recording, write one question for the review. It might be, “Why did our scoring cycle slow down after the first attempt?” or, “Where did we lose possession during defence?” A focused question prevents the team from becoming distracted by dramatic moments that had little effect on the result.

Use a consistent camera position where possible. A high, wide view is usually more useful than a close-up because it shows robot movement, field traffic and alliance spacing. When filming at a school hall or community venue, place the camera safely outside the competition area and check that it does not obstruct officials, spectators or team operations.

File names should include the event, match number and alliance colour. Store a backup in a shared team folder, with access limited to appropriate students, coaches and mentors. In Australia, teams should also follow venue requirements and obtain suitable permission before sharing footage that clearly identifies children or other teams.

Review the match in useful layers

Start with a full-speed viewing to understand the flow of the match. Do not pause immediately after an error. Watch for the relationship between autonomous results, driver choices, alliance movement and the final score. This first pass can reveal whether a problem was isolated or part of a pattern.

Then watch selected moments at half speed. Mark timestamps for successful cycles, missed shots, collisions, stalled mechanisms and periods when a driver appears unsure. Separate facts from interpretation: “robot stopped for four seconds near the loading zone” is more useful than “the driver panicked.”

A replay can also support a constructive discussion format before the team draws conclusions. Students can first record individual observations, then compare them as a group. This gives quieter members a voice and helps prevent the loudest opinion from becoming the official explanation.

Turn observations into measurable changes

Every important observation should lead to a testable adjustment. If the robot takes too long to align, measure the current cycle time, change one element and record the new result. The adjustment could involve driver controls, a camera angle, an intake guide, field positioning or a revised autonomous path.

Use a simple review log with five fields: timestamp, observation, likely cause, proposed change and test result. The “likely cause” should remain provisional until testing confirms it. A robot that misses a target may have a mechanical issue, inconsistent game-piece placement, poor visibility or a control input problem.

Prioritise issues by impact and frequency. A rare minor collision may matter less than a three-second delay repeated ten times. Reliability problems deserve special attention because a robot that cannot finish a match removes options for the entire alliance, regardless of its theoretical scoring capacity.

Compare driving, code and machine behaviour

Video replay works best when it is connected to other records. Compare the match footage with driver-station notes, battery data, scouting sheets, autonomous logs and pit observations. If the robot behaves differently in matches than in practice, look for changes in battery charge, field setup, match pressure or communication.

Driver performance should be reviewed respectfully and specifically. Examine control inputs, route choices, awareness of alliance partners and recovery after an error. Ask whether the controls make the intended action easy. A student may appear hesitant because the mechanism responds slowly or because the camera view is obstructed.

Programming teams can use timestamps to compare planned autonomous actions with what actually occurred. If a routine fails at the same point across several matches, inspect sensor readings and environmental assumptions. Bright venue lighting, reflective surfaces and variations between official fields can expose weaknesses that were invisible in a school workshop.

Use replay to strengthen alliance strategy

A strong robot can still underperform when alliance roles are unclear. Watch how partners enter scoring areas, collect game pieces, defend space and recover after a missed action. Look for traffic patterns rather than assigning blame. Two robots competing for the same route may need a simple priority rule or a change in starting position.

Before an event, create a short strategy brief based on previous footage. Include the robot’s dependable actions, preferred locations, cycle time, limitations and emergency options. This helps students explain their capabilities to potential alliance partners quickly, especially during a busy competition day in Sydney, Perth or Adelaide.

Replay is also useful after a qualification match when an alliance has only a short time to prepare. Select one strength to preserve and one change to make. A concise conversation is more effective than attempting a complete tactical redesign between matches, particularly when students are tired or venue announcements are difficult to hear.

Build a repeatable review routine

Assign clear roles during analysis. One student can control playback, another can record timestamps, another can track the robot and a fourth can monitor alliance interactions. Rotate these jobs so that video analysis develops broad skills rather than becoming the responsibility of one technically confident member.

A practical routine might include a ten-minute full viewing, fifteen minutes of timestamped analysis and a final five-minute action plan. End with one or two changes for the next practice session. The team should then record whether those changes improved performance instead of relying on impressions.

Useful recommendations for a sustainable replay system include:

Carry lessons from the screen to the workshop

Video analysis has value only when it changes preparation. After identifying a problem, create a drill that reproduces it. If the robot struggles to align under pressure, practise repeated approaches from different angles. If alliance communication breaks down, run short simulations using the exact callouts expected at an event.

Australian teams should account for local conditions when transferring lessons. A Queensland event may involve heat and humidity that affect students, batteries and equipment, while a Melbourne tournament can bring cool mornings and rapidly changing travel conditions. Allow time for pit setup, battery checks and school timetable constraints rather than treating the match recording as the entire performance picture.

Review the results at the next team meeting and keep the evidence visible. A chart showing cycle times, successful autonomous runs or completed defensive stops can motivate students more effectively than a general statement that the robot “felt better.” Celebrate disciplined experimentation, including tests that disprove an assumption.

Teams seeking mentors, volunteers or broader event support can explore get involved through Maryland FIRST Robotics. The same habits apply across FIRST programs: observe carefully, collaborate respectfully, test ideas and help students explain what they learned.

Make video replay part of the normal engineering cycle rather than an activity reserved for major failures. Record the match, identify the evidence, choose a focused change and test it at the next opportunity. With that routine in place, every competition becomes a source of practical knowledge that can sharpen driving, design, programming and alliance performance.