A University of Houston team has built a mathematical model that can tell whether a competitive system is healthy, stagnant or skewed. The work could help redesign everything from research funding to military training to better balance excellence and opportunity.
When is competition truly fair, and when is it quietly rigged toward a few winners or watered down for everyone? New research from the University of Houston suggests math can tell the difference.
Ioannis Pavlidis, a computer science professor at UH, and his collaborators have developed a mathematical model that can identify whether a competitive environment is healthy, stagnant or skewed. Their work, published in the journal npj Complexity, offers a general, testable framework for judging the quality and fairness of competition.
The model looks at the statistical pattern of repeated success in a given arena and then works backward to infer the kind of competitive system that produced it. Instead of relying on gut feelings or anecdotes about whether a system is fair, the approach uses data to classify how opportunity and achievement are actually distributed.
Pavlidis, the study senior author and Eckhard-Pfeiffer Distinguished Professor, noted the project began with a simple but bold idea.
“My hypothesis was that there is a universal pattern across human endeavors,” he said in a news release. “We tested that idea by analyzing competitive activities across a broad range of human achievement.”
The team examined three very different domains: Olympic athletes, scientists vying for federal research grants and World War II fighter pilots. In each case, they studied how success accumulated over time and how often new high performers emerged.
Across these fields, the researchers found that the healthiest competitive systems share a common structure. They are demanding enough to push people to improve and allow standout performers to rise, but not so extreme that success becomes nearly impossible for newcomers. At the same time, they avoid spreading rewards so thinly that no one can truly excel.
In other words, effective systems hit a “sweet spot” between cutthroat and complacent.
“In demanding but fair competitive systems, competition itself becomes a learning mechanism,” Pavlidis added. “Everyone is pushed to improve, but some improve more than others, and over time those accumulated gains can produce striking differences in success.”
Using mathematical modeling, the researchers grouped competitive environments into three broad types, each with a distinct “shape” of outcomes.
In the first type, which the team considers optimal, the system is tough but fair. A small group of high performers does emerge, but their positions are not permanently locked in. New competitors still have a realistic chance to rise over time. This dynamic keeps pressure high, encourages learning and innovation, and maintains a sense of possibility for those entering the field.
In the second type, the system becomes winner-take-all. One or two individuals dominate persistently, and their lead is rarely challenged. That pattern suggests a structural imbalance that can discourage participation, sap motivation and eventually lead to stagnation. When people believe the game is effectively over before they start, they are less likely to invest the effort needed to improve.
The third type spreads success broadly, with relatively small differences between participants. On the surface, this may appear more “fair,” but the model suggests it often reflects weak competitive pressure. If rewards are too evenly distributed, there is less incentive to push boundaries, and overall achievement can suffer.
By making these patterns visible, the UH framework gives organizations a new way to evaluate and redesign the systems that shape people’s lives and careers.
The potential applications are wide-ranging. Policymakers could use the model to assess research funding programs and adjust rules so that promising new investigators are not shut out by a handful of established labs. Universities and companies could examine promotion and tenure systems to see whether they encourage growth and recognize emerging talent, or whether they have drifted toward winner-take-all dynamics. Military leaders could apply the approach to training and advancement structures to ensure they reward real performance rather than entrenching early advantages.
Beyond institutions, the work speaks to broader debates about inequality and merit. In many areas — from sports and science to business and the arts — people argue over whether extreme concentration of success reflects pure talent and effort, or deeper flaws in the system. A mathematical tool that can distinguish between healthy and unhealthy patterns of competition could help ground those conversations in evidence.
The study also underscores that fairness is not the same as uniformity. Systems that feel gentle and egalitarian on the surface may, in fact, be holding back excellence. Conversely, environments that produce stars are not automatically fair if the same few people dominate indefinitely and newcomers have little chance to break through.
Pavlidis and his colleagues see their current work as a starting point. So far, the model has focused on competitions among individuals. In future studies, they plan to extend the methodology to team-based contests across different domains and time periods, testing whether similar universal laws apply when groups, rather than solo competitors, are in the arena.
“The sky’s the limit,” added Pavlidis. “This research offers a new way of thinking about competition, which is a fundamental part of civilization.”
If that promise holds, a better mathematical understanding of how success accumulates could help societies design systems that are not only more productive, but also more genuinely fair.
Source: University of Houston