How the World Rugby Rankings work

A rating only ever changes when a team plays, and three things decide how much changes hands:

• The result. A win, a draw, or a loss sets the direction of the exchange.
• The gap between the two teams going in. Beating a side rated well above you is worth far more than beating one rated below you.
• The margin of victory. Winning by more than 15 points adds a multiplier on top.

Ratings almost always sit between 0 and 100, and the best team in the world will usually be above 90. A single match rarely moves anyone by more than a point or two, which is why the table shifts slowly even in a busy season.

Which matches count

Only full internationals between two World Rugby full member unions count. Anything else — club and provincial rugby, age-grade matches, games against invitational or non-member sides — has no effect on anyone’s rating at all.

The formula

Every exchange starts from the same three ingredients: the two teams’ current ratings, which of them is at home, and the result. Home advantage is applied first. For the calculation only, the home side is treated as three rating points stronger than they actually are. That handicap means they earn less for winning at home and give away more for losing there.

Subtract the two adjusted ratings to get the Rating Gap, then cap it at 10. The Core Rating Change is one point, adjusted by that gap:

The favourite is whoever has the higher adjusted rating. A draw always moves points from favourite to underdog, because sharing the spoils is an underperformance by the side that was expected to win. One side’s gain is always the other’s loss.

Two multipliers can then scale the Core Rating Change, and they stack. A win by more than 15 points is multiplied by 1.5. A Rugby World Cup match is multiplied by 2. Because the gap is known before kick-off, so is every possible exchange.

A worked example: Wales vs Scotland

Here is that formula applied to the fixture World Rugby uses: Wales against Scotland in Cardiff, Six Nations, 14 February 2004. Both sides were in the world’s top 10 and separated by barely half a rating point.

1. Start with the pre-match ratings

   These are the ratings the two teams carried into the match. Wales were marginally ahead — a gap of 0.56, which is as close to level as the system gets.

   Wales76.92

   Scotland76.36

2. Hand the home side a handicap

   Wales are at home, so apply the +3 from the formula. Nothing is added to their real rating — this is for the calculation only.

   Wales (home) → 76.92 + 3 = 79.92
   Scotland (away) → 76.36, unchanged

3. Subtract to get the Rating Gap

   The gap is 3.56, well under the cap of 10, so it goes into the formula as-is. Wales are the favourite.

   79.92 − 76.36 = 3.56 in Wales’ favour

4. Plug the gap into the formula

   Substitute 3.56 into each branch of the Core Rating Change:

   Wales win → 1 − (3.56 / 10) = 0.64
   Draw → 3.56 / 10 = 0.36
   Scotland win → 1 + (3.56 / 10) = 1.36

   Wales win

   Wales+0.64

   Scotland−0.64

   The expected result, so it pays little

   Draw

   Wales−0.36

   Scotland+0.36

   Wales were favourites, so sharing costs them

   Scotland win

   Scotland+1.36

   Wales−1.36

   The upset is worth more than double

   That asymmetry is the heart of the system. Wales, stronger on paper and stronger again once home advantage is applied, had little to gain and plenty to lose. Scotland had the opposite deal. Even a draw would have nudged Wales down, because on a gap of 3.56 a draw counts as mild underperformance.

5. Check the multipliers

   Neither of the formula’s multipliers applies here: the eventual winning margin was 13, under the 15-point threshold, and this is the Six Nations, not a Rugby World Cup. Had Wales won 30–10 the exchange would have been 0.64 × 1.5 = 0.96. The same 20-point win at a World Cup would have been 0.64 × 1.5 × 2 = 1.92.

Why you can know the stakes before kick-off

Because the formula depends only on the Rating Gap and the result, every fixture has just five outcomes that can affect it: either side winning by more than 15 points, either side winning by up to 15 points, or a draw. World Rugby can therefore publish exactly what each team stands to gain or lose before the match, which is where the “win and they go top” previews come from.

Where teams enter the rankings

Since 1 December 2012, a union joins the men’s rankings with a rating of 30 as soon as it becomes a World Rugby full member. Before that date new unions started on 40, with the rating treated as provisional until they had played 10 matches. Either way the system reacts quickly enough that a team can travel from the bottom of the table to the top, or the reverse, in fewer than 20 matches.

Two other situations have their own rules. A union that goes several years without a match can have its rating declared dormant and be dropped from the published list; the rating waits for them, and they resume from exactly where they left off. And when countries merge, the new country inherits the highest rating of the countries that formed it, while a country that splits leaves each new team on a rating a fixed amount below the original.

How the women’s rankings set a starting rating

The women’s rankings run on the same Points Exchange engine. What differs is the starting rating, and there is a good reason for it. The first women’s full international was played in 1987, the women’s game has far fewer internationals on record than the men’s, and several teams play only occasionally. When a team has a short match history, one result can swing its position much further than its actual strength justifies.

The fix is a common anchor. Every team is treated as having started on 80 rating points in 1987 — an arbitrary figure chosen purely as a baseline, carrying no judgement about anyone’s strength. Two points are then deducted for each year between 1987 and the year that union played its first full international. The deduction stops at 2007, so any team debuting after 2007 enters on 40.

Common questions

Can you work out in advance what a result will do to the rankings?

Yes, and World Rugby often publishes it before kick-off. The Core Rating Change depends only on the Rating Gap and the result, so there are just five outcomes that can move the needle: either side winning by more than 15 points, either side winning by up to 15 points, or a draw.

Do teams get credit for losing narrowly to a stronger opponent?

No. You cannot gain points for a loss, and you cannot lose points for a win. Rewarding brave defeats sounds appealing until you notice it means docking points from the team that actually won. If a close match really does signal a shift in strength, later results will show it.

How much does margin of victory matter?

Less than you would think. World Rugby found that results predict future performance better than margins do: whether you beat far weaker opposition by 60 or by 100 says very little about how good you are. Winning with something to spare still counts, though, so a 1.5x weighting applies to any win by more than 15 points.

Are all matches treated as equally important?

Almost. Rugby World Cup matches are the only exception, and their exchanges are doubled. Every other full international between member unions counts the same, which keeps the system fair to teams whose calendars mix friendlies and competitive fixtures very differently.

Will teams tend to climb at certain times of year, such as after the Six Nations?

No. Every exchange is zero-sum, so playing more matches carries no built-in advantage. A rating sits exactly where it is until that team next plays, and a predictable result can move it barely at all.

How much do past achievements count?

The rankings measure current strength, not history. Older results steadily fade in influence as newer ones replace them.

How does the system handle freak results?

There is a cap on how far a rating can move up or down on the back of a single match, so one wildly out-of-character scoreline cannot send the table into a spin.

How is home advantage handled?

For the calculation only, the home side is treated as if their rating were three points higher. That handicap means they gain less for winning at home and give away more for losing there, which cancels out the edge of playing in front of their own crowd.

Is the system actually reliable?

Several years of research went into building it, drawing on a database of international results stretching back to 1871, and every weighting in the system was derived from analysing those results rather than chosen by feel.

Its performance is then checked in several objective ways, the most telling being Predictive Accuracy. If, over a long run of matches, the rankings are consistently good at identifying which side will win, that is solid evidence they are reading current strength correctly and reacting sensibly to changes in form.