The Fifth Substitution and the Final 20 Minutes: When the Referee Has to Officiate a Different Match

The 78th minute of a La Liga match I was watching from a studio in Barcelona, the score 1-1. Inside two minutes, the away team used all three of their remaining substitutions at once. Four players left the pitch at three different moments, three replacements entered in three different positions. A holding midfielder sat down to retie his boot beside the touchline, and the referee had to wait. The referee stood at the centre circle, right arm raised, eyes sweeping four directions. From that moment to the final whistle I counted six free kicks, three yellow cards, one VAR check on a goal, and nine minutes of added time. The second half ran nearly twenty minutes longer than the first.

I logged those numbers into my personal tracking sheet, following the same procedure I have kept for five years. Not to indict anyone. But to answer a question few people ask: when the law permits five substitutions, who really has to officiate the final twenty minutes — the coach, or the referee?

A temporary change that became permanent

In May 2026, the International Football Association Board (IFAB) allowed competitions to temporarily use five substitutions, to cope with a congested calendar as football returned after disruption. Two years later, IFAB wrote the provision permanently into Law 3. The number of substitution windows stayed at three, plus half-time.

What is rarely discussed is the structural consequence. A team can change almost half its line-up in the final 45 minutes. That turns the bench from a contingency plan into a primary tactical weapon, and turns the final twenty minutes from the closing act of a match into a different match — different in tempo, different in intensity of contact, different in the number of situations the referee must adjudicate.

I still remember my first lesson about preparing in advance. At the 2026 World Cup, in the opening Group B match between Iran and Morocco, I mispronounced the striker Sardar Azmoun /sæɾdɒɾ æzmun/ as Azmon three times in the first half, even with six pages of notes right in front of me. Viewers complained in the chat, and the editor had to message me. From then on I built a two-step procedure: check the international phonetic alphabet, then listen to a clip of the player or a local reporter saying the name. A mistake on live broadcast is like a mistake on the pitch: look it straight in the eye, learn the lesson, blow the whistle for the next match.

That is also why every piece I write about Spanish referees carries phonetics. Jesús Gil Manzano is /xesus xil manano/. César Soto Grado is /e.saɾ so.to ɣɾa.ðo/. Getting a person's name right is the first step to reading a match right.

How the final twenty minutes swelled

The table below is what I compiled from the most recent 40 matches of each season, sampled in La Liga, without distinguishing big matches from small ones.

| Season | Average added time | Share of yellow cards after minute 75 | VAR checks per match | |---|---|---|---| | 2026-20 | 5 min 10 | 28% | 1.4 | | 2026-22 | 6 min 20 | 31% | 1.9 | | 2026-23 | 7 min 05 | 36% | 2.6 | | 2026-24 | 8 min 15 | 39% | 2.8 | | 2026-25 | 8 min 40 | 41% | 2.9 |

Read that table the conventional way and you conclude referees are becoming stricter. I do not think so. The share of yellow cards after minute 75 rose from 28% to 41%, but the actual duration of everything after minute 75 also nearly doubled. Spread evenly, the number of cards per minute of live ball in the closing phase is almost unchanged.

Event density is the real variable. A match with eleven minutes of added time generates roughly 30% more contested situations than a match with five. Referees have not become more sensitive. They simply have more work, while their legs and their concentration do not grow to match.

I built an index to measure that, which I call the Referee Sensitivity Index. It has four variables: cards per match, share of fouls given after minute 75, VAR checks per match, and the share of decisions overturned on review. The scale runs from 5 to 10.

| Referee | Cards per match | Fouls after minute 75 | VAR checks per match | RSI | |---|---|---|---|---| | Jesús Gil Manzano | 5.4 | 38% | 3.1 | 8.2 | | Alejandro Hernández Hernández | 5.1 | 40% | 3.4 | 8.0 | | Ricardo de Burgos Bengoetxea | 5.6 | 36% | 2.4 | 7.8 | | César Soto Grado | 4.9 | 35% | 2.7 | 7.4 |

The numbers say something the naked eye misses: the gap between these four referees is not in how strict they are, but in how they handle added time. A referee with a high RSI tends to accept long stoppage so as not to miss an incident, and his matches lose rhythm in exchange. A referee with a lower RSI tends to cut stoppage short, preserving rhythm, but is easily accused of ignoring fouls in the 90th minute.

What happens to the team that uses all five substitutions

I split the most recent 120 matches by how many substitutions each team had used by minute 75, then measured PPDA, the number of opposition passes allowed per defensive action, between minutes 76 and 90.

| Substitutions used | PPDA, minutes 76-90 | Counterattacks per match | Goals per match | |---|---|---|---| | 0 to 2 | 11.8 | 1.4 | 0.21 | | 3 to 4 | 10.2 | 1.9 | 0.26 | | 5 | 8.7 | 2.6 | 0.34 |

The lower the PPDA, the higher the pressing. Teams that use all five substitutions press markedly harder over the final fifteen minutes, launch more counterattacks, and score more goals. On the surface, that is evidence the fifth substitution benefits attacking football.

Based on my experience of watching matches, most of that advantage comes not from the player coming on but from withdrawing the right player at the right time. Pedri /pe.ðɾi/ is usually removed around minute 70 to 75, not because he is playing badly but because his high-intensity pressing minutes have hit the ceiling. Lamine Yamal /lamin jamal/ is the opposite case, usually kept until minute 85 to exploit the space opponents leave once they have burned all their substitutions. Antoine Griezmann /ɑ̃.twan ɡʁijɛzman/ is a third type: on at minute 70, not running much, but holding structure and passing rhythm. Three kinds of player, three kinds of response to the same law.

But read that last column another way. Per match, a team using all five substitutions scores 0.13 goals more than a team using 0 to 2. Thirteen percent of a goal. In exchange, they generate 1.2 extra counterattacks and about 2.5 minutes of dead ball per match. Every minute of dead ball is a minute the referee spends standing in the middle of the pitch, explaining, warning, and deciding.

The paradox of the team that substitutes most

This is where the counterintuitive part sits. People assume the team making many substitutions is the proactive team, the one controlling the game. My data shows the opposite in roughly half of cases. A team using all five substitutions is usually a team that is behind or being dominated, and the substitutions are a reaction rather than a plan. The team using fewer, especially the team leading, keeps its shape, keeps the distances between its lines, and makes fewer positional errors.

Data does not blow the whistle, but it illuminates the angles the naked eye misses. The most overlooked angle in every stoppage-time argument is the shape of the team that is winning. Nobody complains when a side shields the ball by the corner flag in the 88th minute. But when the referee adds enough time to cover those minutes, people call it a farce.

I use a three-step model to check any contested incident quickly, called 3F: Foul, meaning whether there was an offence; Field, meaning where on the pitch it happened; Frame, meaning which camera angle is clearest. Those three steps take about 40 seconds for an ordinary incident. But in the final twenty minutes of a match where both sides have five substitutions, I have to run that model on twice as many incidents, while the added time is itself determined by those very incidents.

Emotion demands one thing, the law gives another

Viewers want two contradictory things. They want the match to end on time, and they want every foul punished. In the final twenty minutes, those two desires collide head-on. Every VAR check stops the clock. Every player who goes down adds time. Referees do not create that paradox; they are merely the ones who must stand in the middle and choose.

And here is the point I want to state plainly: a decision that breaks no law can still be wrong in substance. Adding nine minutes is within the law. But if those nine minutes were created by the time-wasting of the team that is winning, then what people need is fairness, not merely accuracy.

I chose La Liga as my analytical sample because it has enough data to verify. Fairy tales in the lower divisions are consumed and then thrown away, and genuine reform of how resources are distributed never reaches them. A fourth-tier match in Andalusia also has a final twenty minutes, also has a referee, also has arguments. It is just that nobody measures it.

People remember the goals; I remember the whistles that protected them. Football is a game of moments, and the referee is the person who keeps time for each of those moments. When the law allowed five substitutions, we extended the number of moments by nearly a third without extending the capacity of the person keeping time.

There are situations with no absolutely correct answer, only a person with enough courage to take responsibility for the decision. But one thing could be done immediately: publish added time broken down by cause — how many seconds for substitutions, how many for VAR, how many for injuries, how many for time-wasting. The argument would then shift from how many minutes the referee added to how many minutes each team consumed. And people would have to look in the mirror instead of at the whistle.

If next season you watch a second half run to 55 minutes, try counting how many of those minutes belong to the referee, and how many belong to the people who chose to stretch it.