Understanding the FIFA Ranking System
The FIFA ranking system uses a modified Elo formula that factors in match result, importance of the match, strength of opponent, and continental strength coefficient. It sounds straightforward but produces results that sometimes look absurd, like a team ranked outside the top 25 when they clearly dominate their region. I've spent years watching this system produce headaches for analysts, journalists, and fans who don't understand the mechanics behind the numbers. The formula itself changes occasionally. The latest version, adopted after 2018, adjusted how friendlies and continental competitions are weighted, and it still gets criticized every time a top footballing nation drops unexpectedly. Brazil has found itself outside the top 25 in FIFA rankings on multiple occasions, most notably around 2021 and again in early 2023. The immediate reaction is usually confusion or outrage. The reality is that the ranking model doesn't care about history or reputation. It cares only about points gained and lost in matches over the preceding four years, with half-points from five years ago still lingering. When Brazil goes through a period of poor results in competitive matches, those old points start dropping off and nothing replaces them. I remember working through a spreadsheet analysis of the 2021 ranking drop. Brazil had lost to Peru in a World Cup qualifier, then dropped points from Copa America matches played the previous year. Their points declined by roughly 60 in a single month. That kind of erosion happens because the system deducts based on the expected result at the time, and when favorites lose, the penalty scales with how unlikely the loss was. A defeat to a team rated 40 spots below you costs significantly more than a loss to a peer. This is the part most people miss.
Friendlies also carry minimal weight in this system. If a team plays a string of exhibition matches against lower-ranked opponents, the points available from those games are capped and the strength-of-opponent factor suppresses any meaningful gain. I once tried to predict where Brazil would land after a tour of four friendlies against teams ranked between 80th and 110th. The model projected a gain of approximately 4 points total. Four points across four matches. The system essentially tells you that non-competitive fixtures are mostly noise, which explains why friendlies don't move the needle the way casual observers expect. Another factor that trips people up is the continental coefficient. Concacaf and AFC matches are weighted differently than UEFA or CONMEBOL fixtures because the average team strength in those confederations differs. When Brazil plays South American opponents, the points swing is larger than when they play teams from weaker confederations. This means that a bad run within CONMEBOL qualifiers hurts more than a bad run against Caribbean or Central Asian teams in friendlies. I learned this the hard way when I initially assumed all losses would penalize equally, which led to a forecast that was off by about 30 points in one projection exercise. The system also doesn't account for squad quality, historical prestige, or fan expectation. It is purely mathematical and retrospective. A nation can field a weakened squad, lose consistently, and watch its ranking fall without the model making any allowance for context. This is by design. FIFA wanted an objective system. Objectivity doesn't mean fairness in a traditional sense. It means consistency, and consistency sometimes produces outcomes that feel wrong.
If you want to track where a team stands or predict movement, you can use the publicly available FIFA ranking calculator or the raw point tables published monthly. The data is transparent. You take each match in the rolling window, apply the Elo-based formula with the current coefficients, subtract the old result, and add the new one. It takes about ten minutes per ranking update if you have a clean dataset. I built a simple Python script that pulls the CSV from FIFA's website and computes projected changes before the official announcement. It gets you within two points of the final published ranking in most cases. The main source of error is when FIFA makes unannounced adjustments to coefficient values between releases. The broader lesson here is that FIFA rankings measure recent competitive performance, not overall footballing strength. Brazil is not objectively a worse team because it sits outside the top 25 at any given moment. The ranking reflects a cluster of poor results compounded by the decay of older strong results. When those old high-point matches age out of the window, the drop accelerates. It is a lagging indicator, and it will always look backward while the team moves forward. There are alternatives if you want a different perspective. Models like SPI from FiveThirtyEight (now archived) or various academic Elo adaptations incorporate strength of schedule more heavily and adjust for home advantage differently. Some analysts also look at expected goals data and youth ranking projections to gauge where a team is heading rather than where it has been. None of these are perfect. They each have their own blind spots. But they can complement the official FIFA table when you need a fuller picture.
Get the Full Details
The rankings will continue to produce moments that look like gaffes. That is not a bug. It is the natural result of reducing thousands of variables in world football to a single integer update once a month. You learn to read past the number and look at the match data underneath it. That is where the actual story lives.
