How to Judge a Debate: Criteria, Scoring, and Fair Panels
To judge a debate, you assess three things: the substance of each side's arguments, the way speakers deliver and defend them, and how well the case is built across the debate. You score each speaker inside a fixed range on a rubric, weigh the clash between the teams, and reach a verdict. When several judges sit as a panel, each scores independently and the scores are combined into one result, with a chief judge harmonising differences and applying any shared rulings once.
That is the whole task in a paragraph. The rest of this guide unpacks each part: what you are actually scoring, how speaker points work, how rebuttal is marked, how a panel reaches one verdict, how time penalties are handled, and how to give feedback that helps. It applies to school, university, and parliamentary debating, whether the round is in person, online, or hybrid.
What are you actually scoring?
You are scoring three dimensions of each speech, commonly named matter, manner, and method.
- Matter is the content: the arguments, the evidence and reasoning behind them, and how well a speaker rebuts the other side. This is the heaviest weighted part in most rubrics.
- Manner is the delivery: clarity, persuasiveness, pace, voice, and presence. Style should serve the argument, not replace it.
- Method is the structure: how the speech is organised, how the team's case fits together, and how well speakers respond to the flow of the debate.
A good judging rubric turns these into named categories with a score for each, so two judges watching the same speech are looking for the same things. Detailed adjudication sheets break matter and manner into finer categories such as analysis, reasoning, organisation, and delivery, each with its own mark.
How do speaker scores work?
Each speaker receives a score inside a fixed range for their speech. The range is set before the debate so that scores mean the same thing in every room and can be compared fairly. A speech that is competent but unremarkable sits in the middle of the range; genuinely strong or weak speeches move up or down from there.
Two rules keep speaker scores fair:
- Stay inside the range. A score far outside the agreed band distorts the result and the speaker rankings, and it marks you as an outlier to the chief judge.
- Score the speech, not the speaker. Judge what happened in this round, not a debater's reputation or how they did last time.
Across the round, the speaker scores also decide the best speaker, the individual recognition given to the strongest performer regardless of which side won.
How is rebuttal judged?
Rebuttal is how directly and effectively a speaker engages the other side's case, and it is often scored on its own because it is where debates are actually won. Reward a speaker who identifies the other team's most important argument and dismantles it with reasoning, not one who simply lists disagreements or repeats their own points louder. A methodical response to the strongest opposing argument counts for more than a scattergun of small objections.
How does a panel of judges reach one verdict?
On a panel, each judge scores the same speakers against the same rubric independently, and the scores are then combined into a single result. Independence first, then reconciliation. This is the step most new judges underestimate, and it is where consistency matters most.
The chief judge, sometimes called the chief adjudicator, leads that process. Their job is to:
- brief the panel on the rubric and the score range before the debate,
- keep each judge scoring independently during the round,
- harmonise large differences afterward by discussing the reasons, not by forcing agreement,
- apply any shared ruling, such as a time penalty, once for the whole panel,
- confirm the verdict and oversee the feedback given to teams.
How are time and penalties handled?
Speeches have set time limits, with a short grace period. Going well over time is penalised, but the penalty is a panel-level decision applied once, not something each judge subtracts on their own sheet. Points of information, where allowed, are offered and taken within the body of a speech and are usually noted separately. The principle is simple: judges score the speech, the chief applies shared penalties, so nothing is counted twice.
How do you give useful feedback?
Good feedback tells a team why they won or lost in terms of the debate, not your personal opinion of the topic. Name the argument that decided it, say what the losing side could have done about it, and point to one concrete improvement per speaker. Keep your own views on the motion out of it. Debaters do not need to know what you believe; they need to know how they argued.
Judging in person, online, or hybrid
The criteria do not change when a debate moves online. You judge matter, manner, and method the same way over video as in a room. What changes is coordination: a remote panel cannot pass paper ballots or huddle in a corner, so judges score on a shared sheet or a scoring tool, and the chief harmonises and confirms the result in one place. Hybrid rounds, with some judges in the room and some remote, work the same way as long as everyone scores against the same rubric.
Common judging mistakes to avoid
- Scoring your own opinion of the topic instead of the arguments made.
- Drifting outside the agreed score range and skewing the speaker tab.
- Rewarding confidence and volume over reasoning.
- Each judge applying the same time penalty separately.
- Giving a verdict without being able to name the argument that decided it.
From paper to a scoring tool
Most panels still score on paper or a spreadsheet, then someone adds up the sheets by hand and reconciles the judges. That works, but it is slow and error prone, especially with a multi-judge panel or a remote round. A dedicated scoring tool gives every judge the same rubric on screen, collects their independent scores, and brings the panel to one verdict without the manual tally.
Tallyman is a free tool built for exactly this: multi-judge panel scoring on a shared rubric, with the chief applying time penalties once, and the result assembled automatically. It works in person, online, or hybrid, and it does not require a tournament to be set up first, so it fits a single class debate, a club round, or a one-off final as easily as a competition. If you run full tournaments with draws and standings, keep your tabulation software for that; Tallyman handles the judging of the round itself.
Frequently asked questions
What are the main criteria for judging a debate?
The content of the arguments, the way speakers deliver and defend them, and how the case is structured across the debate, usually grouped as matter, manner, and method, with a separate rebuttal mark and a best speaker across the round.
How do you score a debate with more than one judge?
Each judge scores the same speakers against the same rubric independently, then the panel combines the scores into one verdict. The chief judge resolves large differences and applies shared rulings, like time penalties, once.
How are speaker points awarded?
Each speaker gets a score inside a fixed range for their speech, based on the rubric categories. The range is set in advance so scores stay comparable across rooms, and the totals decide speaker rankings and the best speaker.
What does a chief judge do?
The chief judge briefs the panel, keeps judges scoring independently, harmonises differences, applies shared penalties once, confirms the verdict, and oversees feedback.
Can you judge a debate online?
Yes. Online and hybrid debates use the same criteria as in-person rounds. Judges score on a shared sheet or tool so the panel result comes together in one place.
Related reading: Debate scoring software vs tabulation, what each does. Tallyman is a free multi-judge debate scoring tool for in-person, online, and hybrid rounds.