A Workers' March Filed Under Football: The Real Arithmetic Buried Inside 6.8 Percent

I opened the file expecting to write a match report. The label said football. Inside there was no club, no player, no goal, no stadium. There was a march in Mexico City, starting outside the Centro Médico Nacional Siglo XXI at three in the afternoon, and one demand: a ten percent direct salary increase. The employer is the Instituto Mexicano del Seguro Social, IMSS. The union is the Sindicato Nacional de Trabajadores del Seguro Social, SNTSS. Active workers and retirees, walking together.

I checked the tape, and the tape told a different story.

I ran the piece through the nine pillars I use for football analysis: tactics and technical quality, club finance and transfers, results and public-opinion cycles, league landscape and positioning, rules and governance, management and dressing-room health, risk, media expectation, and industry transmission. Across twenty-eight information points, not one pillar returned a football finding. That is a failure, but it is not mine. It is a labelling failure. And the labelling failure is the most interesting fact in the file. A system that can file a labour march under football can make exactly the same error about football itself, and nobody will notice.

So where is the game here? The game is inside the label. And a mislabelled story is really a story about verification.

IMSS is the spine of Mexico's public social-security system. Healthcare, retirement, maternity provision for tens of millions of people run through it. An institution that large carries equally sensitive labour relations, because a wage increase here does not just change a payslip. It moves hospital staffing, medicine supply chains and the health budget at the same time. SNTSS is the union that represents those workers. Its internal arguments land on the national ledger.

The march itself is described with precision. Wednesday, 21 October, three in the afternoon, starting at the Siglo XXI medical complex, moving along Paseo de la Reforma toward the IMSS headquarters. The route may change. Metrobús Lines 1, 3, 4 and 7 are likely to be affected, and commuters are advised to take alternative routes.

Here is the first crack. This is service journalism, not advocacy. The tone is measured, the purpose is practical, the aim is to keep people moving. The strange part is the inversion of scale: the bigger the political weight of the march, the smaller the news value assigned to it.

Look at the demands. Ten percent direct salary. More staff. Guaranteed medicine supply. Rejection of the pension system and the Afores. And the most explosive item of all: the departure of SNTSS Secretary General Rafael Olivos Hernández.

Five demands read together tell you something. They are not scattered anger. They are a platform. Platforms are built by organised caucuses. Anger does not build platforms.

Now the arithmetic.

The union has already approved an increase of 6.8 percent. Reporters write the number flat, but it is built from two parts: 2.9 percent on tabular salary and 3.9 percent in benefits.

That split is the crack the headline never shows. Tabular salary is the base table figure, the number from which future pensions, future wage rises and future overtime rates are all calculated. Benefits usually do not compound. They are paid within a window and they stop.

So 6.8 percent sounds large, while only 2.9 percent enters the compounding base. Workers are asking for 10 percent, and the word they use is direct. Direct means base. Direct means changing the foundation of every future calculation.

Put it in football terms. I have spent years decomposing xG. If a team posts 2.1, you do not know whether that is 2.1 or 0.9 plus a penalty. The headline is one thing, the composition is another, and the composition tells you how well the team is actually playing. In 2026, when everyone called Pep Guardiola's inverted full-backs a luxury, I pulled the data and found City's full-backs averaged 8.3 progressive passes per 90 when inverting against 4.1 when hugging the touchline, and City's xG rose 0.47 per game in the 3-2-4-1 shape. Same players, same matches, different position. The headline never showed it.

Wage tables work the same way. 6.8 percent is a headline. 2.9 percent is a structure. Workers are fighting over the structure.

The gap between the two sides is 3.2 percentage points. By labour-dispute standards that is moderate, not extreme. Gaps of that size normally close at a table. This one went to the street, which means the number is not the only cause.

The second cause is that the deal the union signed did not hold with its own members.

That is the real twist. SNTSS approved 6.8 percent. The membership calls it insufficient. The dispute runs first between employer and worker, then between leadership and base. The second dispute is the more dangerous one, because it raises a question of legitimacy rather than a question of price.

Nobody names a general secretary for removal on impulse. Naming an individual signals an organised internal faction with its own arithmetic and its own timetable. A march that starts at a specific hospital complex, follows a named avenue and begins at a fixed hour is not spontaneous. It is planned.

And note who the demands are aimed at. The source is explicit: the authorities and the union leadership, both. This is a dual-track pressure strategy. Pressure on the employer, pressure on its own leadership. Labour history shows it rarely, and when it appears, leadership usually changes.

The article notes a series of demonstrations across October. That detail is small and heavy at once. One protest is grievance. Repeated protests within a month are organisational capacity: the ability to turn people out, to find time, to hold a message.

Retirees march alongside active workers. That is the third crack. Retirees do not take to the street over wages. They take to the street over pensions. Two generations in one march means the complaint is structural, not cyclical.

Now the least discussed part.

The demands reject the Afores, Mexico's private pension-fund administrators, created by the 2026 pension privatisation. Once a wage dispute starts naming the Afores, it stops being a wage dispute. It becomes a grievance against a framework that is nearly three decades old. Structural grievances are measured in years, not months.

In May 2026 I wrote about empty stadiums. People told me it was temporary, that everything would normalise once the pandemic passed. I looked at the first ten rounds of the Bundesliga restart: home win percentage fell from 43.3 to 33.3, home goals per game from 1.7 to 1.2, and away teams took 1.8 more shots per game. The numbers said the crowd was never background noise. The crowd was the tactic.

Same logic applies. The Afore demand tells you this march is not the last stage of a wage cycle. It is the opening of a pension argument. Wage bargaining resolves in weeks. Pension questions go to the political table.

And this is exactly where the football label does damage. A football label makes a story a one-day story. Match over, piece filed, attention moves. This story does not move.

Now the uncomfortable professional part.

How does a labour report enter a football feed? Three possibilities. An automated classifier misread surface vocabulary. The feed's bucket is too wide and swallows any large public gathering. Or somebody tagged it without checking and nobody downstream read it.

The third is the most dangerous. The first two are technology problems, fixable. The third is a habit problem, and habits are hard to repair.

My own rule for years has been tape first: no piece goes out without at least one clip and one number. A hot take without evidence is not a hot take, it is a rumour wearing a headline.

There is no clip here, because this is not football. There are numbers: 6.8, 2.9, 3.9, 10, 3.2. Those numbers tell you the label is wrong and the underlying information is solid.

Good facts under a bad label get ruined along with the label. That is the whole cost of misclassification in one line.

Now let me doubt myself.

I could be wrong three ways.

First, I am assuming the label is an error. Maybe it is not. Maybe in this newsroom, football is a wide bucket that catches any large public gathering and nobody bothers to rename it. If so, the fault is not one bad tag. The fault is the dictionary. And blaming a dictionary means rewriting the dictionary.

Second, I may be over-reading a commuter advisory. The piece mainly says which streets are closed, which lines are slow, which routes to take. I am digging for union legitimacy and pension reform. That could be my own professional deformation. Years of watching matches trains you to hunt for structure behind every event. Sometimes a march is just a march.

Third, and worst, the 6.8 percent figure has no attributed source. Union announcement? IMSS release? Something else? If the number is wrong, my entire 3.2-point calculation collapses.

There is a further problem any data journalist dreads. The date reads Wednesday, 21 October, with no year. 21 October fell on a Wednesday in 2026 and will again in 2026. So this report is either six years old or six years early. Either way, no policy inference built on it is stable.

The route carries its own uncertainty. The article itself warns the path may change, which makes every street name potential rather than confirmed. Useful for a commuter. A hole for an analyst.

Accepting those three weaknesses, here is what I can say.

Test one: if SNTSS does not move the tabular component above 2.9 percent and only adds benefits, there will be more marches in November. If tabular crosses four percent, the streets empty.

Test two: if the demand for Rafael Olivos Hernández's departure does not reach a formal process within two months, the internal faction is loud but small.

Test three: if the Afore rejection stays on placards and no legislator or pension regulator picks it up, the demand is symbolic rather than strategic.

And one test for my own trade. If the football feed swallows another mislabelled story after this, the problem is not an incident. It is a system. A system's error can be forgiven once. Not repeatedly.

A march entered the football file and nobody noticed. If the same machinery one day turns a transfer rumour into confirmed fact, or misreads a fee by a decimal, who measures that damage?

What looked like chaos was a system we had not named yet.

It is time to name it.