The Weight of Countries

By Ian Sacs, 26 March 2025  

How much better is a youth team in Spain than in Finland?

Well, that depends... on a lot. But if all of their training, coaching, and play capabilities were the same - in other words, if they answered the YTSI survey questions exactly the same - would there still be other factors, such as a cultural style of play, that results in differences? I gambled with yes. But too what extent? Now, that's a deeper question. 

This line of thinking led to the refinement of country weighting in the YTSI model. The idea was simple: Youth soccer development is stronger or focused on different priorities in some countries than in others, meaning a team with the same characteristics in one nation might not perform at the same level as an identical team elsewhere. A team in Brazil, for instance, would likely face stronger competition and develop differently than a team in Canada, even if they had the same training regimen. To account for this, it was clear that a country factor was needed, and the first option was to default to the very precisely calculated FIFA rankings for 210 countries. The intent was to make it possible to compare youth teams from different countries as well as local teams. This is expecially important, for example, when trying to decide which international tournament to join.

The Original Method: A Harsh Reality Check

The first iteration of the country factor applied a simple multiplier based on FIFA rankings to the results of the team survey. A team’s survey score was multiplied by a factor derived from its country’s FIFA rank, which ranged from around 0.4 (weaker soccer nations like San Marino) to 1.0 (top-tier soccer nations, like Argentina).

At first glance, this seemed reasonable. If the overall level of play in a country is weaker, teams should have slightly lower adjusted YTSI Scores. However, after applying this model to real teams, a major flaw emerged—the adjustment was far too severe. Teams from lower-ranked nations were penalized excessively, resulting in YTSI Scores that did not align with real-world expectations, especially at youth levels.

Take, for example, a team in Finland that scored 73 based purely on survey responses. A similar team in the USA scored 66. Having watched these teams play personally, I believed this difference accurately reflected their relative strengths. However, after applying the original country factor (0.73 for Finland, 0.88 for the USA), the adjusted scores became 53 (Finland) and 58 (USA). Suddenly, the two teams reversed despite my intuition otherwise. Clearly, something was wrong.

Exploring Alternative Approaches

With the realization that the original country factor was too punitive, especially for lower-ranked countries, we set out to explore alternative methodologies that could achieve the intended effect without distorting scores unfairly. We considered several approaches:

• Weighted Country Factor Adjustment – Instead of multiplying the entire YTSI Score by the Country Factor, we only applied it to a portion of the score, reducing the impact.

• Logarithmic Transformation – A more gradual adjustment that scaled the Country Factor using a logarithmic function, ensuring smaller penalties for teams in lower-ranked countries.

• Stepped Thresholds – Assigning country-based score bands rather than applying a strict multiplier.

Testing & Zeroing In on the Best Model

To determine the best approach, we ran several tests, including applying each method to real teams and checking whether the results aligned with expectations, examining how adjustments affected teams across different YTSI Score brackets (100-90, 90-80, etc.), and comparing score changes to known head-to-head match results between teams from different countries.

The logarithmic approach quickly stood out. Unlike the original model, which scaled adjustments linearly, the logarithmic function ensured that:

• Teams from lower-ranked nations were still adjusted downward, but not excessively.

• Teams from top-tier countries retained their advantage without overinflation.

• The impact of country weighting was more gradual, rather than causing dramatic drops.

The Current Model & Next Steps

After extensive testing, we implemented the logarithmic transformation with a base value (minimum weighting) and a scaling factor to control the effect. To be totally transparent, there is still a "gut check" factor being applied here based on the shape of the curves throughout the brackets. That factor will come in handy if my future goals of improved calibration come to fruition. Still, this new model acceptably adjusts scores based on country strength, but in a way that maintains fairness and real-world accuracy

That said, YTSI is always evolving. While this model is a significant improvement, we plan to continue refining it by gathering more real-world match results to further validate adjustments, exploring youth-specific international rankings rather than relying on FIFA rankings, and analyzing how teams that move between countries perform to assess country factor impact.

The goal remains the same: to ensure that YTSI Scores truly reflect team strength, regardless of where they play. And with each iteration, we get closer to that perfect balance. Already, YTSI is the world's best tool for youth football coaches to assess the skill level of their teams and compare them to others.

So, does a team from Spain always have an advantage over a team from Finland? Well, no. But thanks to a more refined country factor methodology, we can now measure that difference far more accurately.

YTSI is free to use and available now to all youth soccer teams worldwide. To find out where your team stands, visit www.ytsi.net today.