TI #023: What Do Teams Do 10 Seconds Before a Goal?
Timeless Insights #023
Football analytics has become increasingly sophisticated at measuring shots. We can estimate chance quality, compare finishing skill, model goalkeeper positioning, and understand which teams consistently outperform their xG totals. All of that has genuine value.
But goals are rarely created at the instant a player strikes the ball.
They are usually built in the seconds beforehand: a forward dragging a centre-back out of position, a midfielder receiving between lines for one touch too long, a full-back arriving unnoticed on the blind side, or a defensive line shifting half a second later than it needed to.
The finish is visible. The construction is often hidden.
That is what makes this paper so interesting. Long before tracking data became a mainstream discussion point, the authors asked a question that still feels highly relevant today:
Can we identify the recurring ways teams create scoring chances, rather than only measuring the shot that ends the move?
Using a full season of player tracking data, they analysed the ten seconds before every shot, grouped similar attacking sequences together, and then combined those patterns with an expected-goals framework to measure which methods of scoring were most effective.
It was an early attempt to move football analysis away from isolated events and toward coordinated behaviours.
And even now, many clubs are still trying to solve that same problem. That’s why today I will give you 4 practical ways clubs can turn this idea into competitive advantage. Let’s first see what the idea is…
The central idea: goals leave tactical fingerprints
Most event data treats shots as separate moments.
A shot from 14 meters. Right foot. Assisted cross. Blocked.
That information matters, but it strips away context. It tells us what happened at the end of the move, not how the move became possible in the first place.
This study approached the problem differently. Rather than beginning with the shot location, it began with the sequence that produced the shot. The researchers extracted the ten seconds leading up to every attempt and included the movement of attackers, defenders, and the ball. They then clustered those sequences into recurring patterns.
That transforms the way we think about chance creation.
Instead of seeing thousands of isolated shots, we are able to see a library of repeatable attacking mechanisms.
Some chances come from quick transitions against an unbalanced back line. Others emerge from patient circulation followed by a cutback. Others come from second balls after set pieces, or wide overloads that force a late defensive rotation.
Those chances may end in similar shot locations and receive similar xG values, but tactically they are entirely different products.
For coaches and analysts, that distinction is crucial.
Because if you only measure outcomes, you may miss the process that sustains them.
Why combining patterns with xG was so clever
The paper’s smartest contribution was not simply clustering movement patterns. It was connecting those patterns to shot quality.
The authors used an xG model and separated chances into higher-value, medium-value, and lower-value opportunities depending on their likelihood of resulting in a goal.
That allowed them to ask a much more powerful question:
Which attacking patterns tend to create genuinely dangerous chances, and which patterns merely create volume?
That remains one of the most useful distinctions in modern football.
Two teams can produce the same xG total across a month of matches, but in very different ways.
One team may generate structured cutbacks, central box entries, and clear one-touch finishes. Another may accumulate xG through rebounds, speculative second phases, and low-frequency chaos.
Both totals appear equal in the dashboard.
But one profile is often more repeatable than the other.
This is where many public conversations around xG still fall short. Total chance quality matters, but the source of that chance quality matters as well. Some attacking processes are robust under pressure. Others are fragile and highly sensitive to variance.
This paper pointed toward that distinction years before it became fashionable to discuss process stability.
Team style becomes measurable, not descriptive
Once each shot had been assigned to a scoring-pattern cluster, the authors could describe every team as a distribution of how often they relied on each type of chance creation.
That may sound simple, but it changes what “style” means.
Too often, style is discussed through vague language: direct, patient, aggressive, transitional, possession-heavy.
Those labels can be useful shorthand, but they often collapse important differences.
A side described as direct may rely on diagonal switches and second balls. Another may use vertical combinations through the centre. Both are “direct,” yet they attack in completely different ways.
This framework instead asks:
Which scoring patterns does the team trust most often?
Which patterns generate the best chances?
Which methods disappear against stronger opponents?
Which mechanisms remain stable home and away?
That is a far richer tactical profile.
And for opposition analysis, it becomes extremely practical. Defending a team is not just about reducing possession share or blocking crosses. It is about disrupting the specific pathways they repeatedly use to create shots.
The deeper lesson: football is a coordination sport
The lasting value of this paper is philosophical as much as technical.
It reminds us that football outcomes are usually collective outcomes.
Shots are often credited to one player, assists to another, and errors to a third. But many decisive moments are produced by interactions between multiple players moving in synchrony.
A winger pins the full-back.
A striker occupies two defenders.
A midfielder arrives late.
A centre-back hesitates.
A lane opens.
No single action explains the chance, but the combination does.
Tracking data allows us to see those interactions more clearly than event data ever could. And even now, this remains one of the hardest challenges in football analytics: converting movement relationships into actionable insight.
Some players look elite because they operate inside elite structures.
Some players look limited because they operate inside poor ones.
Some defenders are blamed for situations created by collective spacing failures three seconds earlier.
This study understood early that many of football’s most important truths live between players, not inside one player’s stat line.
Why this matters even more today
Modern football has only increased the value of this type of thinking.
Pressing systems are more refined. Defensive spacing is better coached. Space appears and disappears faster. The average team is tactically sharper than it was a decade ago.
As a result, random chance creation becomes harder to sustain. Teams increasingly need rehearsed and repeatable ways to destabilise organised opponents.
That might mean:
overload-to-isolate patterns out wide
third-man combinations through central traffic
underlaps to create cutback angles
decoy runs that open zone 14
far-post occupation against shifting back fours
It is crucial to understand which collective sequences consistently manufacture xG in the first place.
So what should clubs actually do with this?
If you can identify the attacking patterns that lead to your best chances, you can begin to redesign training, recruitment, and match preparation around those mechanisms rather than around generic principles.
Fewer clubs than you imagine can truly operationalise chance creation patterns.
That means a competitive edge may belong to teams that build an internal operating system around how chances emerge repeatedly.
So these are the four practical ways clubs can turn this idea into competitive advantage:
