GOAT Index ranks the boxers of every era - world-title holders and ranked contenders across the four major bodies, WBA, WBC, IBF and WBO - on one and the same scale, as an independent site. Reaching across generations and organisations, a Glicko-2 rating makes visible who is really the greatest.
Independent and neutral
Tied to no promoter and no sanctioning body, every fighter is judged by the same yardstick, with no favours.
Updated continuously
Results are taken in every week. From prospects to all-time greats, the rankings never stop moving.
Every era comparable
Peak re-scoring lets the old masters stand next to today's champions, without any era adjustment.
How the rankings work
The GOAT Index rankings are produced with Glicko-2, scored bout by bout. The point is that a vast body of results across generations is connected as one single large network. By joining up who beat whom, and how, across eras and organisations, boxers who never shared a ring can still be compared.
The rating weighs more than a win-loss count: it takes in the strength of the opposition and its uncertainty (rating deviation), the manner of the win (a stoppage counts for a little more than a decision), activity, and how firmly the record is established. A fighter who has been away a long time is rated with less confidence, one who keeps performing with more - the aim is a ranking that shows where each fighter stands.
There is no era adjustment. In the All-Time (current rating) board a rating is frozen after retirement, so a longer career carries its late losses. The Peak board re-scores each fighter at the highest rating they ever reached, so old masters such as Ali, Robinson and Roy Jones Jr. climb back into the upper ranks. The rating itself is plain Glicko-2 - no hidden coefficients.
Men and women are ranked separately
Men and women have separate rating pools that never meet, because they never fight each other. Two ratings from different pools are not comparable, so the site keeps the boards - and the virtual-match win probability - within a single sex. On the top page you can switch between the men's and women's rankings, and the fantasy-match engine declines to score a cross-sex pairing rather than print a number that would mean nothing.
About the operation
GOAT Index is run independently by one person who loves the sport. It exists to promote no particular fighter and no particular organisation, and so that the data and the computed results stay in the foreground, nothing is published about the operator personally. The value of the site lies not in who built it but in how accurate and how verifiable it is.
Neutrality and advertising
To cover running costs the site uses advertising and affiliate programmes. However, rankings, ratings and the order of listing are never adjusted for an advertiser or for a fee. Content that carries advertising is marked PR in plain sight. Rankings and the rest of the data are computed entirely apart from advertising.
Rankings = decided by the rating aloneMethod = plain Glicko-2, no era adjustmentSponsored content = marked PRRank adjusted for a fee = never
The conversation continues on Discord.
Objections to the rankings, the prospects worth watching, dream fights across the eras - a community where fans can talk it over.
Every week. The latest results are taken in, and title changes or the rise of a prospect are folded into the rankings as they happen.
Q. What is the difference between All-Time and Peak?
All-Time (current rating) freezes a fighter's rating at retirement or the present, so a long career carries its late losses. Peak re-scores each fighter at the highest rating they ever reached, which is why old masters climb back near the top there. There is no era adjustment in either.
Q. How recent is Active P4P? Is two years not too long?
A boxer who has fought within the past two years is treated as active. That may look generous, but at the top of the sport a gap of more than a year is not unusual. When a fighter goes a long time without a bout the confidence in the rating falls, and the rank drifts down on its own.
Q. Why are men and women on separate boards?
Because they never fight each other, their ratings come from separate pools and are not on a comparable scale. Ranking within each sex is the only honest way to do it, so the boards - and the virtual-match win probability - stay within one sex.
Q. Who is included?
The pool is built around fighters who have contested a world title in the four major bodies (WBA, WBC, IBF, WBO) and those in the major rankings. World Series of Boxing, amateur and exhibition bouts are excluded. The scope widens as the data comes together.
Q. A fighter is missing, or a record looks wrong.
Reports are welcome through Discord or the contact page. A source, such as a link to an official record, makes a correction certain, and every report goes towards better accuracy.
Contact
Reports of data errors and listing issues, and enquiries about coverage or partnership, are welcome through Discord or the contact page. Every report received goes towards better accuracy.
Disclaimer:The rankings and ratings on this site are analysis and estimation by an in-house algorithm, and are not a guarantee of any result. Care is taken over the accuracy of the published data, but neither completeness nor currency is guaranteed. This site neither recommends nor solicits any form of gambling. This site uses advertising and affiliate programmes, and a purchase made through a link may earn the operator a fee (rankings and ratings are not adjusted for any fee).
Organisation names, brand names and fighter names belong to their respective owners.