How to Run a Robotics Competition [2026 Guide + Field Cycle Math]

A robotics competition is not one tournament. It is usually three of them running in the same hall, on the same clock, under one banner.

That is the thing that catches organizers out, and it is why the scheduling maths from every other sport quietly stops working at the door.

Here is the number most first-time organizers get wrong.

In a 2v2 alliance format, a qualification match consumes four teams at once. So the matches you need are teams x matches-per-team, divided by four.

Thirty-six teams playing six qualification matches each is 54 matches, not 216.

On one field at a five-minute cycle that is 4.5 hours. On two fields it is 2 hours 15 minutes.

I have run this at scale.

In July 2025 we scored the IIUM Robotic Competition — 406 school teams, 755 matches, 8 fields, several categories, all on one public link.

Key Takeaways

• In a 2v2 alliance format, qualification matches = teams x matches-per-team / 4. Each match burns four teams, not two.
• Field cycle time is the planning unit, not match length. A 2-minute match with a 3-minute reset is a 5-minute cycle, and the cycle is what you multiply.
• In an attempt-based category, runs = teams x attempts. 24 teams with 3 runs each is 72 runs — 3 hours on two tables, 6 hours on one.
• FIRST LEGO League ranks on your best match of three, not your total. That is a different scoring model from every bracket sport, and it needs software that understands attempts.
• A team's matches must be spaced. With 36 teams and four teams per match, a team appears roughly once every 9 matches — about 45 minutes of pit time between runs.

Why a Robotics Competition Is Three Tournaments at Once

Walk into any decent STEM event and you will see three completely different scoring models on the floor at the same time.

Robot sumo is head to head. Two robots, one ring, somebody gets pushed out. That is an ordinary single elimination bracket and it behaves itself.

Robot soccer is a group stage. Goals per match, points table, goal difference as the tie-break.

Time rush and line follow have no opponent at all. A robot runs the course, the clock stops, and the team is ranked against everyone else's best run.

Those three do not share a scoring model, a schedule shape, or a definition of "winning". They only share a venue.

Which is why most organizers end up with one spreadsheet per category and a manual merge at the end of the day.

The merge is where the mistakes happen. It gets done fastest, latest, and by the most tired person in the building.

Running each category as its own competition, published under one event, is the fix. The multiple divisions guide works through the court-sharing half of that problem.

How Many Matches Does an Alliance Format Actually Need?

This is the arithmetic that has no equivalent in ordinary sport, so it is worth doing slowly.

In VEX V5RC and FIRST Robotics Competition, a qualification match is 2v2. Four teams on the field, randomly assigned, ranked individually afterwards.

So each match delivers four team-appearances, not two.

• 24 teams x 6 quals = 144 appearances = 36 matches
• 36 teams x 6 quals = 216 appearances = 54 matches
• 48 teams x 6 quals = 288 appearances = 72 matches

Now turn matches into hours. A V5RC match is two minutes of play. Field reset, scoring and getting the next four teams on takes about three more.

Five minutes a cycle is the honest planning number. At 54 matches that is 270 minutes on one field, or 135 minutes on two.

The second constraint is spacing, and it is the one that gets forgotten.

A team playing 6 of 54 matches appears in one match in nine.

That is about 45 minutes between runs — enough time to fix a robot, which is the entire point.

Halve your field count and you halve that gap. Teams then have twenty minutes to repair a drivetrain, and the pits become the bottleneck instead of the field.

How Long Does an Attempt-Based Category Take?

Attempt-based categories are easier to schedule and much harder to score.

Easier, because there is no dependency. Nobody is waiting on anyone else's result, so the queue is a pure division: runs / tables.

• 24 teams x 3 runs = 72 runs. Two tables at a 5-minute cycle: 3 hours.
• 36 teams x 3 runs = 108 runs. Two tables: 4.5 hours. Three tables: 3 hours.

Compare that with a bracket, where a round cannot start until the round feeding it has finished. The court allocation maths explains why that distinction matters so much.

Harder, because of how the ranking works.

FIRST LEGO League Challenge gives every team three 2.5-minute matches and counts only the highest score. Your two worse runs are thrown away.

That is the same scoring model a school sports day uses for field events, and it is genuinely rare in tournament software.

Most bracket tools cannot express it at all. They want a winner per match, and an attempt has no opponent to beat.

How Do You Rank Teams Who Never Play Each Other?

Three different answers, depending on the category, and you should publish which one you are using before anyone competes.

Best attempt. Rank on the single best run. Used by FLL and by most time-rush categories.

Accumulated points. Add every attempt together. Used where consistency is the skill being tested.

Alliance qualification points. VEX ranks individually on Win Points, then Autonomous Points, then Strength of Schedule.

That last one is the losing alliance score, so it rewards close matches you lost.

That third one exists for a specific reason. You were randomly partnered, so your record is partly luck, and ranking on wins alone would punish teams who drew weak partners.

Whichever you use, write down the tie-break before the day. Two robots finishing the course in 12.4 seconds is not a rare event, it is a Tuesday.

If that rule lives in a judge's head, the team that loses the decision will believe it was arbitrary. From where they are standing, it was.

The tournament rules template has the general clauses worth publishing; for robotics, add attempt limits, the scoring window, and the tie-break order.

What Actually Breaks at 400 Teams

Not the arena. Paper.

At the IIUM competition the bottleneck was never the fields. It was scoresheets walking from the floor to the judges' table, and a leaderboard permanently twenty minutes behind reality.

Three things mattered more than anything clever:

• Several scorers entering results at once, from different arenas, without overwriting each other.
• Standings recomputing as each run finished, so nothing waited for a merge.
• One public leaderboard the teams could check themselves, so they stopped queuing to ask.

That last one is not a nicety. A queue of coaches at the judges' table is thirty minutes of your day, every hour.

The same shape of problem applies at 40 teams. It just does not hurt yet.

Frequently Asked Questions

How many matches does a robotics competition need?

For an attempt-based category, matches equal teams x attempts — 24 teams with 3 runs each is 72 runs.

For a 2v2 alliance format, divide by four instead: teams x matches-per-team / 4, because each match uses four teams.

Thirty-six teams playing six quals each is 54 matches, not 216.

How long does a robotics competition take?

Multiply matches by the field cycle, not by match length. A 2-minute run with a 3-minute reset is a 5-minute cycle.

Fifty-four alliance matches on one field is 4.5 hours; on two fields, 2 hours 15 minutes. Add registration, inspection and judging, none of which compress.

How are teams ranked if there is no opponent?

On their own runs.

Either the best single attempt, which is what FIRST LEGO League uses across its three matches, or the total across all attempts where consistency is the point.

Publish which one applies, along with the tie-break, before the first run.

How does VEX ranking work?

Win Points first, then Autonomous Points, then Strength of Schedule. A qualification win is worth 2 WP and an Autonomous Win Point adds 1.

Strength of Schedule is the losing alliance's score, so a narrow loss ranks above a blowout. It exists because partners are assigned randomly and your record is not entirely yours.

Can one event run sumo, soccer and time trials together?

Yes, and it is the normal case. Run each category as its own competition with its own scoring model, then publish them under one event.

What you want to avoid is one spreadsheet per category and a manual merge at the end of the day. That merge is where results go wrong.

How many fields do I need?

Work backwards from the finish time you have booked, then check the spacing.

More fields shorten the day, but they shorten the gap between a team's matches too.

With 36 teams and four per match, two fields give roughly 22 minutes of repair time instead of 45.

Below about twenty minutes, the pits become your bottleneck.

Running a Robotics Competition: Recap

• Alliance formats need teams x matches-per-team / 4 qualification matches. Four teams per match, not two.
• Attempt categories need teams x attempts runs, and they queue freely because nothing depends on anything else.
• The field cycle is the planning unit. Match length alone will understate your day by more than half.
• Ranking is on your own runs — best attempt or accumulated — not on beating an opponent.
• Space each team's matches. Roughly 45 minutes between runs is repair time; twenty is a queue at the pits.
• Publish attempt limits, scoring windows and tie-break order before the event, not at the judges' table.

Tournamyx scores attempt-based categories natively — configurable attempts per team, ranking by points or time, and priority-ordered tie-breaks set before the day rather than argued about during it.

Head-to-head categories run as brackets in the same event, and everything publishes to one live leaderboard.

One honest limit: if two teams tie on their best run, the ranking falls through to your next metric rule, not to their second-best run.

Add a secondary metric such as fastest time and the tie resolves itself.

It is free for events up to 8 teams, and it ran 406. Set up your next competition and stop merging spreadsheets at midnight.