LMB: 93 Games, 20 Teams, and the AI Gamble — Liga Mexicana de Beisbol Weighs a Schedule Expansion

Toros de Tijuana and Olmecas de Tabasco have just closed out the 2026 Serie del Rey. But the most notable story of the Liga Mexicana de Beisbol season is not a swing — it is a number mentioned very little: 93. Since 2026, every LMB team plays exactly 93 regular-season games. For comparison, MLB plays 162. The Mexican Baseball League, with 20 teams split into two zones — North and South — operates at roughly 57% of MLB's schedule size. And league leadership is considering adding four to five more games for the 2027 season. The signal comes from Horacio de la Vega, LMB president. Speaking as the Serie del Rey reached its decisive phase, he confirmed the league has hired an artificial intelligence analytics firm to assess the feasibility of expanding the schedule. Results are expected in two to three weeks. He said: 'That is an important number, but not much more than that.' The framing is notable. It is not a statement of ambition — it is a statement of process. Context for Vietnamese readers: the Liga Mexicana de Beisbol is Mexico's top professional baseball league, founded in 2026, currently with 20 teams divided into two geographic zones: Zona Norte and Zona Sur. Teams mostly play within their zone during the regular season, then the zone champions meet in the Serie del Rey — the national championship series. Toros de Tijuana represent the North, Olmecas de Tabasco the South in the 2026 season. The two-zone structure exists for geographic reasons: distances between Mexican cities are vast, and travel cost is a life-or-death variable for a league with a modest budget compared to MLB. The core issue is this: LMB is trying to solve an optimisation problem between revenue and operating cost, and it has chosen to solve it with data rather than instinct. De la Vega made clear the AI model will analyse 'administrative issues, flight logistics, and competition between teams.' Those three variables are not random. Adding four to five games per team, multiplied by 20 teams, creates 40 to 50 additional games league-wide. If those games are cross-zone North–South matchups, travel costs rise exponentially. If they are intra-zone, commercial value rises less because audiences are already familiar with those pairings. This is the point conventional analysis often misses. The question is not 'should LMB add games?' but 'which games should it add, where, at what point in the calendar, and who pays for them?' A cross-zone game between Toros de Tijuana and Olmecas de Tabasco has far higher television value than a tenth intra-zone repeat. But it is also more expensive to stage. The AI model LMB has hired is expected to answer that question: does the marginal revenue from a cross-zone game cover its marginal cost? There is a notable social signal. According to the source, 'some months ago, LMB fans had suggested the proposal.' This is a small but important detail. It means pressure for schedule expansion is not coming from league leadership, but from the audience. Fans want more cross-zone games between North and South. Leadership is responding to that signal, but in a controlled way: hire outside analysis, wait for results, then decide. This is a de-risking communication strategy. If the result says no, they have grounds to explain. If it says yes, they have evidence to persuade 20 club owners. And here is the biggest blind spot in the story: there is no voice from the clubs in the original report. De la Vega speaks, but no club president confirms support or opposition. In a league where any format change must be ratified by members, the silence of 20 teams is an unpriced variable. They may agree. They may be waiting to see who benefits and who loses. A team in Tijuana, near the US border, has a completely different travel-cost structure from a team in Tabasco, in the southeast. Adding cross-zone games could be a windfall for one and a burden for the other. Historically, LMB has changed its game count before. The 2026–2026 period saw adjustments that led to the current 93-game figure. That means precedent exists: this league can and does change its calendar. But one thing must be stated clearly: there is a data conflict in the source. One information point says games were 'originally planned to be 23,' while two other points confirm 93 games per team since 2026. These two numbers cannot both be correct. The 93 figure is corroborated by two independent points in the same document and is more structurally plausible. The 23 figure is most likely a transcription or summarisation error. But its existence is a reminder: even in official data, errors happen. Based on my experience tracking transfers and league structures, this is the kind of news analysts tend to undervalue. No goals, no referee controversies, no transfer stars. But it shows a league operating like a business: identify the problem, hire experts, set a deadline, wait for data. This process differs fundamentally from how many football leagues operate — where schedule decisions are often made based on media pressure or the interests of a small group. Look at the number four to five games. It is small. It equates to a 4.3% to 5.4% increase over the current 93-game schedule. This is not a revolution. It is a probing step. And the way it is framed — 'not much more than that' — shows LMB leadership understands the political feasibility of a large change is low. A small step can be accepted. A large one can divide. This 'incremental' strategy is the safest way to test stakeholder reaction before committing to a larger structural change. The question is: if the AI model recommends no change, what happens? De la Vega has already built the answer: 'we will see the results and decide.' This is a statement usable for both scenarios. If expansion happens, it is evidence the league listens to fans. If not, it is evidence the league is cautious with data. Both are positive media stories. This is how a well-managed organisation handles uncertainty. But there is another risk less discussed. Baseball is a sport with specific physical loads. A pitcher needs three to four days of rest between appearances. A hitter playing 93 games over about five months is a workload optimised over decades. Adding four to five games is not a large numerical change, but it can create a chain effect on rest days, night flights, and recovery time. The AI model is said to analyse this factor too, but the source provides no detail on how it is incorporated. A final point: LMB is using AI as a decision-making tool, but AI cannot replace the consensus of 20 teams. A model can say expansion is theoretically optimal. But if ten Southern teams feel they bear more cost than additional revenue, the deal will not pass. This is the point technical analysis never captures: the internal politics of a league. And that is a variable no AI model can simulate accurately. When the 2026 Serie del Rey ends and Toros de Tijuana or Olmecas de Tabasco lifts the trophy, another question will begin: how many games in 2027? The answer comes in two to three weeks. And it will be decided by an algorithm, a president, and twenty club owners — in that order.