TL;DR
- The question: when one thing changes in a club’s travel week, can one screen show everything it affects, in time to decide?
- What ran: a demo on fictional data, and a replay of Crystal Palace’s real 2025/26 season against a simulated travel plan.
- Result: 31 real fixture changes touched 28 of 60 matches. In the simulation, a change reached 8.4 linked items on average, but only 4.0 needed action.
- My take: the links are the easy part. The rules that decide what matters are the hard part. Next is a test with a club’s operations team.
When a team plays away, how many people are involved in getting it there?
The players are the visible part. Behind them are coaches, medical staff, analysts, kit staff, security and media. There are flights, hotels, coaches from the airport, meal plans, recovery equipment and a lot of boxes. Someone has to book all of that, check it and change it when the schedule moves.
And the schedule moves. A kick-off time changes for television. A cup replay lands in the middle of a planned week. A flight is delayed and the hotel is suddenly too far from the training pitch.
I built a demo to test one question: when one thing changes, can a single screen show everything it affects, early enough for someone to decide? The demo above runs on sample data for a fictional club. Click the three story buttons at the top in order, and mark each flagged item right or wrong. The frame shows the phone view; open the full dashboard for the desktop version.
#Where it stands
#Why it is getting harder
Most clubs handle this well, mostly because experienced people carry a lot of it in their heads. The kit manager knows which cases go on which flight. The team operations lead knows which hotel handled a late arrival badly last season. That knowledge works until the calendar gets busy enough that nobody has time to use it properly.
The calendar has grown in two directions at once: more matches and more distance.
Since 2024, the Champions League league phase has given each club eight matches against eight different opponents, four of them away. That means more unfamiliar cities, airports and hotels than the old group format. The expanded Club World Cup in the United States in 2025 added a summer tournament on another continent. Pre-season tours still go to Asia and North America because that is where much of the commercial audience is.
The volume of equipment is easy to underestimate. Add kit, medical and recovery equipment, analysis hardware and nutrition supplies, and a single trip becomes a small freight operation.
The sporting cost matters more than the financial one. Elite teams now plan around recovery windows measured in hours. A late arrival, a poor night in a noisy hotel or a long transfer from the airport can take something away from the next match, and possibly the one after.
#What one change sets off
Consider a Champions League away match on a Wednesday. The plan is to fly out on Tuesday morning, train at the opponent’s stadium in the evening, play, and fly home straight after the match so the squad can recover before a league game on Saturday.
On the Friday before, the league match is moved from Saturday afternoon to Sunday for broadcast reasons.
That single change can affect a long list of things. The return flight may no longer need to leave at 1am. The recovery schedule changes. The Saturday training session at home moves. The hotel for the following away trip may need a different check-in date. Catering, security and bus bookings all shift with it. Some of these changes carry fees, and some cannot be made at all if nobody notices them in time.
None of this is difficult on its own. The difficulty is that the information sits in different places: emails with the charter company, a hotel booking system, the medical team’s recovery plan, the kit list and someone’s spreadsheet. The person who notices the problem is often the person who happens to remember that two bookings depend on each other.
#What the demo does
Each trip becomes a set of linked items rather than a collection of separate bookings. The flight is linked to the hotel, the hotel to the training slot, the bus and the kit, and all of them to the match and the recovery plan around it. When something changes, the dashboard shows what depends on it, who decides, the deadline and what each option costs.
The demo has six parts: an overview for the morning, a 30-day travel calendar with recovery windows between matches, trip details with each booking’s supplier, terms and source document, the kit sets and how they rotate between trips, a map of shipments in transit, and a single queue of incidents and decisions.
The system suggests. People book. Nothing in the demo makes a booking or a payment.
I built it with Claude Code (Claude Opus 5.5) in October 2026. It is one web page, with no server and no AI calls when it runs. The links between items, the rules (such as a 64-hour minimum between matches) and the consequences of each change are written into the sample data by hand. What the demo tests is whether this way of showing a change is useful, not whether software can work out the consequences on its own.
#What happened when we tested it
In the demo itself, the only test so far is my own run through the story; the next section replays a real season. All the demo’s data is fictional: Harbourside FC, its fixtures, suppliers, terms, costs and shipment positions. It has not been tested with a club or with team operations staff.
Here is what the demo does in each beat, as it behaves today.
Morning, Monday 23 November, 08:00. The overview shows the next trip, Istanbul for a Champions League match, leaving in 1 day 2 hours. Six of six bookings are confirmed and Kit Set B is in transit. One item is amber: the party list is not final because two players have fitness tests. It is the only open decision, with a deadline of 18:00 when the charter manifest closes.
Kit held at customs, 09:12. A carrier update holds Kit Set B for inspection in Istanbul. Its ETA moves to Wednesday 09:00, after the first training session on Tuesday evening. The shipment turns red and an incident is raised automatically. It flags four items that need a change: the first training session, the kit manager’s plan, the Newcastle kit truck on Friday (Set B can no longer be back in time) and the return of Set B by freight. It also lists two linked items it checked and left alone: the match itself, which Set B still reaches, and the Lisbon trip, which uses another set. The options are to send Kit Set C by air freight (€7,700 including Set B’s return), to move the first training to Wednesday morning (the head coach decides, and the medical staff keep it short), or to wait for customs until 14:00. Picking an option updates the kit rotation and the calendar, and the log records who decided, why, and how long it took.
League match moved to Sunday, 11:40. The league moves the home match after the Lisbon trip from Saturday 15:00 to Sunday 16:30. The dashboard flags the pre-match hotel, the Lisbon return charter and hotel, the recovery session and the last training session. The calendar turns the gap before the next away match red: 49 hours, below the 64-hour minimum. The two options are to stay over in Lisbon (€7,600) or to fly home after the match as planned. Neither fixes the 49 hours, and the dashboard says so.
Every flagged item has right and wrong buttons, a box for anything it missed, and an export of the marks, missed items and decision times. That is how the next step will be measured.
The next step is a test with a club’s team operations lead. They write their own list of what each change affects, then run the same story and mark every flag. We will publish those results here.
#Results: one real season, replayed
I can’t test with a club yet, so I tested the method on a real calendar. I asked a second Claude instance (Claude Opus 5.5, with web search, November 2026) to research one Premier League club’s 2025/26 season. It also built a simulated travel plan for that club, separately from the demo. It picked Crystal Palace, the Premier League club with the most European travel that season: 60 matches in all competitions, 17 of them in the Conference League, which Palace won.
The real part: the calendar
The research found 31 fixture changes, affecting 28 of Palace’s 60 matches. Each one has a source. I checked six against the club’s own announcements, and all six matched.
- 13 changes were caused by a European match on the Thursday before and 13 by television (10 date moves, 3 kick-off times). Two were caused by cup congestion, one was a postponement, one was that match’s rearrangement, and one move has no confirmed reason.
- The median notice was 49 days, ranging from 15 to 72.
- With a 70-hour minimum between the end of one match and the next kick-off, two changes pushed a gap below it, from 89 to 67.5 hours. Both came from the same announcement on 31 October. A third shortened an already short gap from 52 to 46 hours. 16 changes made recovery better. In the final schedule, 16 of 59 gaps were under 70 hours.
The simulated part: the travel plan
The travel plan is invented: 305 bookings with made-up suppliers, terms and fees, based on how Premier League clubs usually travel. It is not Palace’s plan. Replaying each real change against it:
- A change reached 8.4 linked items on average, but only 4.0 needed a change or a decision. The most was 10.
- 5 changes needed nothing, mostly because the bookings didn’t exist yet.
- 14 needed at least one decision by a named person, such as the head coach or the medical staff.
- With the real notice, no change fees would have been due. If every change had been noticed only after its free-change deadline, the invented terms would have added up to £264,700. The closest call was Bournemouth away: the hotel’s free-change window closed about 9 hours after the announcement.
What this shows, and what it doesn’t
The calendar numbers are real. Fixture changes were common, touching almost half of Palace’s matches, and most came with weeks of notice. The risk is less about surprise and more about volume, and about the knock-on effects, like a hotel deadline only hours away.
The gap between 8.4 linked items and 4.0 needing action is the noise question from the demo. A dashboard that flags everything linked would flag about twice as much as matters. The rules that filter those flags matter as much as the links.
None of this shows that the dashboard is accurate, or faster, for a real operations team. That is still the next test.
#What still needs people
The head coach decides when the team travels and how much training it does on the road. The medical and performance staff decide what recovery the players need. Those decisions shape the trip. A planning tool should start from them, not overrule them.
There are also trade-offs that have no single right answer. Flying home straight after the match saves a night in a hotel but costs sleep. Staying over is better for recovery but loses a training day at home. One coach will prefer the first option, another the second, and both can be right for their squad.
Then there are the relationships. The team operations lead who knows the charter company’s duty manager by name can solve a problem in ten minutes that a system would turn into a ticket. That kind of knowledge is part of what a club pays for.
What a system can do is make sure those people see the consequences of a change early, with the facts in front of them.
#What it will not solve
It will not stop a volcano, a strike or a fog-bound airport. Disruption will still happen. The best a planning tool can do is help the club respond faster, with a clearer view of its options.
It also depends on the quality of what goes in. If bookings are made over the phone and never recorded, or supplier terms live only in someone’s inbox, the system will have gaps. It will show where those gaps are, but it cannot fill them.
There is a quieter risk too. A tidy dashboard can make a plan look more certain than it is. A transfer time that was fine in October may not be fine in December traffic. People need to be able to see where a recommendation comes from and to override it without a fight.
And it cannot fix a calendar that is simply too full. If the fixture list leaves no good option, the tool will show that clearly. Someone still has to decide which compromise to accept.
#What we are still testing
The main question is whether the dependency view is accurate. If the system misses a link between two bookings, people will stop trusting it after the first surprise. If it flags too many links, it becomes noise.
We are also testing how much of a club’s travel history is usable. Some clubs keep detailed post-trip notes. Others keep almost nothing in writing. We want to understand how much value the tool adds in each case, and how much setup work it needs before it helps.
We do not yet have evidence that this improves recovery or results. That would be a strong claim, and we are not making it. For now, the practical question is narrower: does it help the operations team see the effect of a change sooner and spend less time chasing information?
If you run team operations or travel at a club, I would like you to take part in that test: write down what you think each change affects, run the demo, and tell me where it was right and where it was wrong.
#My take
The people who move a football team around Europe are very good at what they do. Most of the time, the problem is not skill. It is that the information they need is spread across too many places and changes too often.
I find this area interesting precisely because it is unglamorous. Nobody announces a transfer for a better hotel booking. But a squad that arrives rested, with the right equipment, on time, is in a better position to play well. That is worth some attention.
I would start small: one competition, one season’s worth of trips, compared against the current process. Let the operations team challenge the output and keep track of what it gets wrong.
Beyond football
The same pattern shows up wherever a group of specialists has to arrive somewhere at a fixed time with the right equipment: touring orchestras, film crews, field engineering teams, medical teams moving between sites.
The details differ. The underlying question does not: when one thing changes, who finds out what else it affects, and how quickly?
