Quick answer. Decide the question before the match, write a one-line definition for every event you count, and count only a handful of things, every match, with the same person doing it. Entries into the 25 and the circle by zone, plus penalty corners won and conceded with their outcome, are a good start. Write down how you thought the match went before you open the numbers, and when a number surprises you, watch the clips before you conclude anything. The Dutch women’s team logs about ten thousand events a match. You don’t need to.
That’s the short version. Here’s where each part of it comes from, drawn from the analysts, coaches and research behind The Hockey Site’s four-part series on match data.
Start with a question, not with the tagging
At the World Cup in August, the Dutch women’s team had a three-person analysis team keying roughly ten thousand events per match, according to Joost Bitterling of the KNHB, speaking to NOS. That’s one end of the scale. A professional analyst we asked for the series works the other way round: only collect what changes the chance of a goal, and settle what you’re testing before anyone starts clicking.
With one pair of hands, the second approach is the one that lasts a season. So write the question down before the whistle. “Do we get into the circle from wide or through the middle?” gives you something to count. “Tag everything and see” tends to give you three matches of half-finished data and then nothing at all.
Write a one-line definition for every event, and borrow one
Ask three people whether a ball that crosses the circle line and comes straight back out is a circle entry, and you won’t get one answer. Change the definition and the number changes, even with the same person tagging. The research world hasn’t been much tidier: Hughes, Cooper and Nevill found in 2002 that 70% of 72 notational analysis papers reported no reliability check at all.
You don’t have to invent your own definitions. Secret Analyst, a company that codes matches for international teams, shared the circle entry definition it uses on every match, and gave permission to reproduce it:
Distinct pass, carry or interception that immediately results in (1) controlled possession in the circle, (2) a penalty corner (or stroke), or (3) a deflection (or rebound) opportunity, including “near miss” far post dives.
It judges an entry by its outcome rather than by where the ball went. In and straight back out: not an entry. A deflection off a defender: only if it wins a corner or a stroke. It’s a choice rather than the one correct definition, and a team that counts every ball over the line will record more entries from the same game. What matters is that whoever tags uses the same written line every week. Put circle entry, turnover, penalty corner won and outlet completed on one sheet, one line each, and keep it next to whoever is tagging.
Count a few things, every match
The research on what separates winners from losers points at territory rather than possession. At the 2018 men’s World Cup, circle entries split winning and losing teams more strongly than any other stat measured, with possession at roughly half that weight (Bagchi and colleagues, 2023). Possession is also hard to tag reliably from the side of a pitch. Entries are not.
So build a sheet a parent can fill in: entries into the 25 and into the circle, left, centre or right, for and against. Add penalty corners won and conceded, each with what happened. After ten matches you have your own corner conversion and your opponents’. For each event, note where it happened and in which minute, and leave the player’s name off. That keeps the tagging quick, and the minute column is what lets you line the sheet up with video or running data later.
Then give each event type one owner for the whole season. If the tagger changes in November and the definitions were never written down, the trend line breaks without anyone noticing.
Tag live for Monday, count afterwards for the trend
Sebastien Roland, who coaches in Belgium’s men’s third tier, keeps his live tagging small: penalty corners, the fouls that win or concede them, and five “specials” (outlets, build-up, attacking build-up, defensive press and corners), so the clips he wants to show are marked before the final whistle. Emily Calderon, the Belgian federation’s senior performance analyst, codes circle entries after the match instead, because a count with mistakes in it is no use to anyone. Split the work the same way. Live: the moments you want on screen in training. Afterwards: the counts you want to compare from one match to the next.
Write down how it felt before you look at the numbers
Before you look at the sheet, write one line, something like “we were under pressure all second half”. Then check it against the counts. Emily Calderon suggests exactly this for a coach with no analyst and very little time. Her example is the coach who swears the team was pinned back, then finds the circle entries and shots were level with the opposition. Where the feeling and the numbers disagree is where your evening should go.
Roland calls those disagreements anomalies. His example was his back four’s outlets: “we succeed three, we missed 20, so we had 13% of success. So that’s an anomaly.” The number told him something was wrong, not why, so the answer came from the clips. Number first, then footage, then a conclusion.
Take three clips to Tuesday, not thirty
The analysis only counts if something changes in training. When an Australian Football League club’s coaches went through video with players one to one, the players recalled 6% of the feedback a week later (Mason, Farrow and Hattie 2021). Roland never runs a video briefing longer than 15 minutes. So mark three moments during the match with the minute, show them on Tuesday, and ask the players what they see before you say what you saw.
Two questions this usually raises
Can AI tag my hockey matches for me?
Not yet. When the series checked the vendors’ own documentation in autumn 2026, no company sold automatic tactical tagging for field hockey, and services sold as automated breakdowns were done by trained people. That’s fine, as long as you know it’s what you’re paying for. What works today is the camera that follows play. Our free report on AI for field hockey coaches going into 2027 covers the rest.
Do I need a camera and software to analyse a match?
No. A printed sheet, a pencil and one person on the side of the pitch will give you entries by zone and corners with outcomes, which is most of what the World Cup research points at. Video earns its place when a number surprises you, because that’s when you need the clips. If you do buy a camera, buy it for filming and sharing, and keep the tagging by hand.
Two free sheets to print. The game data sheet has entries into the 25 and the circle by zone and quarter, for and against, plus penalty corners, with Secret Analyst’s circle entry definition printed at the top. The Match to Tuesday sheet takes you from three moments flagged during the match to a fifteen-minute session and one-to-one clips. Both are PDFs, no sign-up needed.
Everything above comes from The Hockey Site’s four-part series on what a coach can do with match data. The first half of parts one and two is free to read. The second half of each, where the working detail sits, is in the Coach subscription, alongside the full masterclasses the coaches’ quotes come from.
Go more in depth here
Read next: Free to read in full, part four of the series on getting the findings to players, and Robert Noall on analysing an opponent from video: You found it. Will they remember it?, How to do a basic analysis from your opponent from video.
More coaching questions: What stats matter in field hockey?, We get into the circle plenty but hardly score. What should I be coaching?, What formation should I play in field hockey, and how do I pick the right system?, Why do my passes keep getting intercepted under pressure?. All of them are in Field Hockey Coaching Questions.
The experts mentioned in this article
Join us for more on this
Free coaching ideas from the world’s best, straight to your inbox.