What is performance analysis in sport?

Performance analysis in sport is the systematic observation and measurement of athlete or team performance, used to give objective feedback that improves the next performance. In practice it means recording what happened — through video, notation or sensor data — turning it into evidence a coach can act on, and closing the loop by checking whether the intervention worked.

It exists because watching is not remembering.

The reason it exists, with the number most people get wrong

The founding study in this field is Franks and Miller, 1986. It is quoted constantly, and it is quoted wrong.

What that study found was that observers recalled around 42% of the critical events in one half of an international football match. The figure has been repeated on coaching blogs for thirty years as "coaches only remember 42% of a match" — but the people tested were novice physical education students with no soccer experience. They were not coaches.

The number for actual coaches came later. Laird and Waters (2008) ran the same design on qualified, experienced coaches and found recall of critical events at 59.2% (SD ±15.3). Familiarity with the sport helps, by about seventeen percentage points.

Two honest conclusions follow, and the second one matters more:

• Being a good coach genuinely improves what you retain. Expertise is not a myth.
• You are still, on the best available evidence, losing around four events in ten, and you do not know which four.

That gap is the entire argument for performance analysis. Not that coaches are unobservant — that no one watching twenty-two people for ninety minutes can hold it all, and the parts that fall out are invisible by definition.

The four types of performance analysis

The term covers four quite different activities. Most arguments about analysis are really people using the same word for different jobs.

Notational analysis counts and codes discrete events — passes, tackles, serves, turnovers — to produce frequencies, sequences and patterns. It needs a defined event list and somewhere to tally. A pen and paper is enough to start. Full guide with templates.

Biomechanical analysis measures how the body moves: joint angles, ground contact time, technique breakdown. Frame-by-frame video, force plates or motion capture. This is where Kinovea and Dartfish live.

Time-motion analysis quantifies movement demand: distance covered, sprint counts, work-to-rest ratios. GPS vests or tracking video.

Tactical and pattern analysis looks at how units behave in relation to each other — shape, transitions, set-piece structure. Usually video, usually the whole pitch in frame.

Where Sidetalk sits, and where it does not

Worth saying plainly, because the honest answer is unusual: Sidetalk does none of these four.

It does not count events, measure joints, track distance or map shape. If you need notational data, you need a notation tool. If you need biomechanics, you need frame-by-frame software.

What Sidetalk does is the step after all four — the one described below as stage five, which is the stage every setup skips. It records what you said while you coached, works out which player each remark was about, and turns it into individual feedback you can review and send. The only overlap with the four types is structural: you can mark kick-off, half time and goals, which is enough to place your words on the video's clock. It is not an event-coding system and does not pretend to be.

The performance analysis cycle

Six stages, and it is a loop rather than a funnel.

1. Define the question. "Are we losing possession in our own third?" — not "let's analyse the match." Analysis without a question produces data nobody uses.
2. Capture. Fixed camera, phone on a fence post, or an automated system.
3. Code and tag. Mark the events that answer your question, and only those.
4. Interpret. Compare against your own baseline, not against elite benchmarks. How to read the numbers without fooling yourself.
5. Feed back. Short, specific, to an individual or a unit. Clips beat spreadsheets, and a sentence aimed at one player beats both.
6. Re-measure. Did the behaviour change?

Stage six is where almost every amateur setup stops being analysis. If you capture, interpret and present, but never check whether the thing you asked for actually happened, you are doing reporting. Reporting feels productive and changes nothing. The re-measure is the only stage that tells you your intervention worked, and it is the cheapest of the six — you tag the same category again next match and compare two numbers.

Stage five is where the other failure lives. A great deal of analysis effort produces a team meeting: eighteen players watching clips, of which perhaps three are about them. Why individual feedback is the part that moves a player.

Performance analysis vs sports analytics

These get used interchangeably and should not be.

They overlap at the edges and the tooling barely does. If you are a grassroots or youth coach, performance analysis is your discipline; analytics is interesting reading.

What a setup actually costs

Three honest tiers.

Free. A phone on a tripod, a notebook, a spreadsheet. This genuinely works, and a coach doing it properly beats a coach with expensive software and no question. What it costs is your hours — realistically 45 to 90 minutes per match once you include the tallying.

Middle. Analysis software layered over footage you already record. You keep your existing camera and add tagging, clipping or feedback tooling on top. Sidetalk sits here, and is free during the pilot — no card, and the price gets set with the coaches using it now.

Full. Automated capture — Veo, Trace, Spiideo, Pixellot — plus analysis software above it. The thing worth knowing before you spend: capture hardware and analysis software are two separate purchases. Buying the camera does not get you the analysis, which surprises a lot of clubs in year two. What to do once the camera is doing its part.

Five mistakes that show up in every club

• Collecting data with no question behind it. The tally sheet fills up and nothing is decided.
• Coding too many event types. Twelve categories is sustainable for three weeks. Five to eight survives a season.
• Comparing your U14s to professional benchmarks. Elite averages are the wrong baseline for developing players and produce despair rather than information.
• Feedback sessions longer than attention spans. Three clips per player, not twenty per team.
• Never re-measuring. See stage six. This is the common one.

Frequently asked questions

What is performance analysis in sport?

The systematic observation and measurement of performance, used to give objective feedback that improves the next performance. It runs in a loop: define a question, capture, code, interpret, feed back, re-measure.

Why is performance analysis important in sport?

Because unaided recall is incomplete in ways you cannot detect. Qualified coaches recall roughly 59% of critical match events (Laird and Waters, 2008), and nobody knows which 41% they lost. Analysis replaces argument about what happened with a record of it, and makes feedback specific enough to act on.

What's the difference between performance analysis and video analysis?

Video analysis is one method within performance analysis. Performance analysis is the whole discipline, including notation with no video at all, GPS data and the feedback loop.

What qualifications do you need to be a performance analyst?

At professional level, typically a sports science degree with a performance analysis specialism, plus practical experience with the standard software. In the UK, BASES accreditation is the usual professional route. At club and youth level there is no qualification requirement at all — most analysis is done by coaches, and competence comes from doing it consistently rather than from certification.

Can you do performance analysis without expensive software?

Yes, and the free tier above is how. One question, five event definitions written down, a tally sheet, and a re-measure next match. The equipment is not the constraint; the question and the follow-through are.

Where to go next

If you want the practical workflow rather than the concepts, start with how to analyse sports video step by step. If you are choosing tools, the buyer's guide covers the three layers coaches usually confuse.

And if the bottleneck in your setup is stage five — you can see what to fix, you just cannot get it to eighteen players individually — that is the specific thing Sidetalk is built for.

---

Sources: Franks, I. M. and Miller, G. (1986). Eyewitness testimony in sport. Journal of Sport Behavior, 9(1), 38–45. · Laird, P. and Waters, L. (2008). Eyewitness recollection of sport coaches. International Journal of Performance Analysis in Sport, 8(1), 76–84.