The South’s Success

Rugby union evolved from soccer when, in 1823, a schoolboy at Rugby School named William Webb Ellis picked up the football and ran with it. The newly invented sport quickly became popular. The first international match took place in 1871 between England and Scotland (Scotland won!). The sport continued to gain popularity over the next century, but did not transition to a professional sport until the early 1990s.

Marking the start of the professional era, the first Rugby World Cup was inaugurated in 1987. Taking place every four years, it is an important direct measurement of success in the rugby world.

Timeline: Rugby World Cup Champion

A complete history since the World Cup began in 1987.

The South has eight World Cup victories, shared between New Zealand and South Africa (both 3 time Champions), and Australia (2 time champions), whereas the North has only England (1 time champions). Success in a single World Cup may be subject to many factors out of the control of a nation - key players being injured, unlucky pool draws, and random unfortunate moments in matches. However, when viewed in aggregate over nine World Cups, the South’s dominance cannot be ignored.

In order to more continuously measure success between World Cups, the governing body, World Rugby, introduced the World Rugby Rankings in 2003. Based on inferential statistics, the rankings award or deduct points following a match based not simply on whether the team won or lost, but based on how they performed against the expectation of winning, given the points standing of both teams going into the fixture. Updated each month, the rankings allow comparison between top teams who may rarely play each other but who play common intermediaries.

Timeline: Number One Ranked Nation

A complete history since rankings began in 2003.

In the 206 months that have passed since the rankings began in October 2003, New Zealand has held the number one rank for 170 months, far ahead of South Africa (26 months), England (8 months), Ireland (1 month), and Wales (1 month). For 15 years, between June 2004 and July 2019 Southern Hemisphere nations continually held the umber one rank.

Both of these important metrics establish the South’s dominance. It is already well known by fans of the sport, and the reasons for this imbalance of power between the hemispheres is regularly touched upon in the press, by coaches, or by fans through media like Twitter or Reddit. Most discussion is around a handful of hypotheses - this article examines six popular ones.

In this article we will be concentrating on the men’s professional game in the seven most successful nations, but it should be noted that the sport is officially played in 120 countries around the world. Women’s rugby is growing, and semi-professional leagues have recently been established in both the North and the South. There are a large number of amateur clubs and tournaments. The sport is vibrant in its growing diversity and inclusivity, and the focus in this article is not intended to go against that, but to examine a specific and well-known imbalance in the men’s game.

Hypothesis One: the South has a greater number of players.

The total number of players measures how popular the sport is, as well as the size of the pool of talent that national sides have to choose from. Common logic suggests that the more players you have, then the greater the number of ‘exceptional’ players there are to choose from at the apex of the sport in any given nation.

Number of Registered Players by Nation

⊕The area of each circle is proportional to the official total number of registered players in this visual, based on the latest data from World Rugby’s Year in Review 2019 report.
Data: World Rugby

Surprisingly, one of the sport’s most iconic nations, New Zealand, ranks only fifth by number of registered players. France and England have more players (65% and 127% more, respectively), but South Africa has by far the most players (79% more than the next nation, England).

In total, the three southern nations have 1.06 million players, 24% more than the four northern nations which have 0.85 million players.

However it cannot be ignored that there is over an order of magnitude in population between many of the nations. Wales is the smallest nation with 3.1 million inhabitants, followed by New Zealand with 4.8 million. France is the largest with 67 million inhabitants, followed by South Africa with 58 million. So when the number of registered players is divided by the population, we find the picture changes significantly.

Registered Players as a Proportion of Population by Nation

Registered players divided by population.

⊕
Data: World Rugby, United Nations

Wales and New Zealand lead by this metric with the sport being twice as popular than in the next highest ranked nation, Ireland. South Africa now lies fourth.

According to national sports authority sources, rugby union is the most popular sport in Wales and New Zealand. Soccer is the most popular sport in all of the other nations. Australia is the only exception, where Australian Rules Football is the most popular sport.

Although in absolute terms, the South has more players, it is not such a simple matter. By number and ranks of the two statistics we have looked at, there is no clear pattern.

Hypothesis One decision: rejected. Since evidence is mixed, the hypothesis that the number of players is a reason for the South’s dominance is rejected.

Hypothesis Two: in the South players are paid more.

Domestic professional competitions directly feed the national sides, and the amount of money in these leagues can be an indicator of strength. By this statistic, France’s Top 14 league leads with an average salary of €328,571 per player for the 2018/19 season. English and Irish clubs both pay more than the New Zealanders. South African and Australian clubs pay the least out of all of the top nations.

Average Top-Tier Professional Player Salary by Nation

Top-Tier Leagues: Super Rugby, Pro 14, Top 14, English Premiership

⊕
The area of each circle is proportional to salary.
Data: Premiership Rugby, Ligue Nationale de Rugby, Super Rugby, The Rugby Paper, Ruck.co.uk, The Guardian

Hypothesis Two decision: rejected. The evidence is clear that player salaries are not a reason for the South’s dominance. On the contrary, player salaries are generally far higher in the North.

Hypothesis Three: in the South the climate is better for rugby.

Cold and wet weather tends to be harder to play in. You are more likely to stay indoors than go out and practice rugby, from fewer ‘backyard throw-arounds’ at the grassroots level, to shorter training sessions at amateur club and professional level, and more injuries all round mean that cold and wet weather could conceivably affect an outdoor sport like rugby union. So let’s look at how the two hemispheres compare.

The two visuals below show the annual average temperature for the past ten years. Climate can vary across a country, so for simplicity we are examining the climate of the city of the highest placed clubs from each nation in the 2019/20 season. In the Southern Hemisphere, the seasons are six months offset from the North. July in the South is climatically equivalent to January in the North. Therefore the diagrams are rotated so that the seasons are aligned.

The top professional league season is marked with thicker lineweights, with off-season being thinner.

Climate: Temperature by Nation

2011-2020 average from the city of each nation’s most successful club in 2019/20 season.

Units: °C.

⊕Mouseover, zoom-in, and filter the climates to discover whether there is a difference between hemispheres. Double-click in the main visual to reset. Ideas: try filtering above so that you only compare: AUS and FRA. Next: IRE and NZL. What similarities or differences do you observe?

The Southern Hemisphere has higher temperatures overall. Durban is far hotter than any other location year-round. Toulouse and Melbourne both have very warm summers, but during winter Toulouse is far colder than Melbourne. Exeter, Cardiff and Dublin all have similar temperatures to Christchurch.

In the Southern Hemisphere, the season runs from late summer to late winter (7 months), and avoids the hottest parts of the year. In the Northern Hemisphere, the season also avoids the hottest parts of the year, running from late summer to early summer (10.5 months). It is noteworthy that despite having a colder climate, the North avoids playing during the hottest months of the year. Traditionally rugby is seen as a ‘winter’ sport, but in the professional era the playing season has clearly outgrown winter, with only a six week break in summer. The sport in the North would benefit from continuing to play in the warmer weather, and instead taking their short break during the colder months instead.

Rainfall is also an important aspect of climate.

Climate: Rainfall

2011-2020 average from the city of each nation’s most successful club in 2019/20 season.

Units: mm

⊕Mouseover, zoom-in, and filter the climates to discover whether there is a difference between hemispheres. Double-click in the main visual to reset. Ideas: try filtering above so that you only compare: RSA and WAL; AUS and FRA; NZL and RSA; IRE and WAL. What similarities or differences do you observe?

Rainfall is a more complex picture, with no clear pattern within either hemisphere, nor a correlation with temperature. For example, the hottest location, Durban, is also the wettest. One of the coldest locations, Cardiff, has a similar total rainfall to Durban. Christchurch, with a similar temperature climate to Cardiff, has the least precipitation.

In the South, the playing season avoids the wettest part of the year for all three nations. In the North, it is the opposite, and the playing season is during the wettest part of the year for all four nations.

Hypothesis Three decision: accepted (just about!). The South has marginally higher and less seasonal temperatures. Rainfall is a mixed picture, although we have noted that the South avoids playing during the wettest periods of the year.

Hypothesis Four: the South plays a more positive, running game, focusing on scoring tries.

In rugby union there are four methods of scoring:

• A try (5 points). A player grounds the ball in the opposition’s try area (the area behind the posts).
• A try conversion following a try (2 further points). Following a try, a player from the team that scored takes a conversion kick at goal.
• A penalty goal (3 points). In the event of an infringement during general play, the non-offending team is awarded a penalty kick at goal.
• A drop goal (3 points). A player may dynamically during general play attempt a kick at goal. This is relatively rare due to the difficulty and subsequent high failure rate.

Since try conversions only follow a try, and since drop goals are rare, this leaves tries and penalty goals as the main methods of scoring in a match of rugby union.

Focusing on scoring tries requires a more creative attacking mindset, whereas reliance on penalties leads to a negative and more passive mindset, playing ‘safer’ tactics, pressuring the opposition to concede a penalty.

Measuring the number of tries and penalties scored by teams at top-tier professional club and international matches in the seven nations is the most direct way to test this hypothesis. Using data on 1,719 Southern Hemisphere matches and 5,198 Northern Hemisphere matches between the 2012/13 and 2019/20 seasons, the two visuals below show the distribution of match results.

A match where one side scored 3 tries and 1 penalty would contribute towards the cell 3 across and 1 up. The team they played against would also contribute independently. This is done across all matches, then divided by the total number of scores to get a percentage probability of occurrence for each try/penalty combination.

Southern Hemisphere: Tries vs Penalties Scored by Team

3,438 team scores from 1,719 matches, 2012/13-2019/20 seasons.

⊕The cells are shaded with more intense colours the higher the probability. You can also mouse-over to see the raw percentage probability.

In the Southern Hemisphere, the most likely result is 2 tries and 2 penalties (5.4% probability), closely followed by 3 tries and 1 penalty (5.3%), and 2 tries and 1 penalty (5.2%). This is followed by 4 tries and 1 penalty (4.8%), 3 tries and 2 penalties (4.5%) and 2 tries and 3 penalties (4.4%). Probabilities drop away steeply after that from 3.9% down to 0.1%.

The overall gradient is along a line from top left to bottom right. Teams that score a lot of tries tend not to score so many penalties, and vice-versa.

Northern Hemisphere: Distribution of Tries vs Penalties Scored by Team

10,396 team scores from 5,198 matches, 2012/13-2019/20 seasons.

In the Northern Hemisphere, the most likely result is the same as the Southern Hemisphere: 2 tries and 2 penalties (5.6% probability). But the next highest results are 1 try and 2 penalties (5.3%), and 1 try and 3 penalties (5.1%). This is followed by 2 tries and 3 penalties (4.8%), and 2 tries and 1 penalty (4.6%). The overall top-left to bottom right gradient feature is also present.

The two distributions both share 2 tries and 2 penalties as the most likely result. However upon closer inspection the difference is clear - the next four most likely results tend towards scoring tries in the Southern Hemisphere, and towards scoring penalties in the Northern Hemisphere.

In order to better visualise the difference between the hemispheres, let’s take the Southern Hemisphere probability and subtract the Northern Hemisphere probability from it. Yellows and greens are negative values which show areas of Northern Hemisphere tendency, and purples are positive values which show areas of Southern Hemisphere tendency.

Difference (SH minus NH): Tries vs Penalties Scored by Team

⊕Small values of less than 0.5% points difference are coloured white as they could be considered statistical noise, but this data is still available to explore via mouseover.

Statistically, the difference is striking. There is a 22.9% difference in probability distribution of scoring between the two hemispheres. The total probability of the yellow and green regions is 10.9%. The total probability of the purple regions is 12.0%.

The areas are generally contiguous, that is to say that they are one continuous block, with only 2 blocks out of 12 in the Southern Hemisphere not being so (and they are not sporadically placed, which would be an indication of statistical noise).

Hypothesis Four decision: accepted. Given the large sample size over several seasons, there is strong evidence of a trend towards try scoring in the South and penalty scoring in the North.

Hypothesis Five: Southern talent moves between hemispheres more

We are going to look at the movements of ‘hemisphere hoppers’ - players and coaches who move between hemispheres. The first heatmap below shows the movements of the 51 players and 7 coaches at Southern Hemispheres clubs who have spent at least one season in the Northern Hemisphere in their career over the past 10 years. The second heatmap below shows the same thing, but for the 231 players and 9 coaches at Northern Hemispheres clubs who have spent at least one season in the Southern Hemispheres over the past 10 years.

Interact with the heatmaps: mouseover to see individual talent histories, and draw a window to zoom in, pinch to pan.

Players with NH Club Experience at SH Clubs

Players with SH Club Experience at NH Clubs

There is a clear difference in numbers of players in each set shown, although this is most likely because there are more professional clubs in the Northern Hemisphere. The first visual shows 9.4% of total talent in the South, and the second 10.9% of total talent in the North.

Hemisphere hoppers currently at Southern clubs are much more likely to have originated in the South, spent some seasons in the North, and then returned home again. Hoppers currently at Northern clubs are much more likely to have originated in the South, and finish their careers in the North.

• At Southern clubs, 33 out of 57 (58%) hemisphere hoppers have originated in the South.

• At Northern clubs, 222 out of 242 (92%) hemisphere hoppers have originated in the South.

• Globally, 255 out of 299 (85%) hemipshere hoppers have originated in the South.

• Globally, 44 out of 299 (15%) hemisphere hoppers have originated in the North.

• Out of 1,396 players and coaches at professional clubs in the North, only 18 are hemisphere hoppers who have originated in the North.

There are many reasons why this could be the case, such as higher salaries in the North; England, Ireland and Wales having a strict policy of only picking home-based players for the national team - in the South, only New Zealand has this policy; and Southern players having a reputation as being naturally more talented than Northern players.

Whatever the reason, the South has a well established diaspora in the North. Whilst this surely does benefit the North to some extent, the South must gain more strategic rewards with so many ambassadors abroad. Ultimately, at the highest level, players and coaches will be loyal to their national sides and when it comes to international fixtures including World Cup competitions, it is clear that the Southern coaches will have a broader level of intelligence on operations in the North.

Hypothesis Five decision: accepted. 85% of all hemisphere hoppers presently active at top level professional clubs in the 2020/21 season have originated in the South.

Hypothesis Six: the North plays more fixtures.

Average Fixtures Per Season by Club

Aggregate over 2012/13 to 2019/20 seasons.

⊕Mouseover to view more detail in this visual.

Club sides in the North play between 22 to 28 regular league fixtures per season. This variability is due to the size of the league and how successful a team has been. The English premiership has 12 teams, whereas the French Top 14 and the European Pro 14, which Irish and Welsh sides play in, have 14 teams each. Each league has a play-off semi-final and final at the end of the season, adding 2 extra fixtures for the top four highest league finishers.

Club sides in the North take part in one of two top-tier European competitions, depending on how highly they finished in their regular league in the previous season. For higher finishers, there is the European Champions Cup, and for the other teams there the European Challenge Cup. These competitions accounted for 7 or 8 fixtures on average.

Finally, the English premiership squeezes in an additional Premiership Rugby Cup competition held between its 12 teams, which accounts for a further 3 fixtures on average.

Club sides in the South played between 15 and 17 regular league fixtures. There are 13 Super Rugby teams spread across South Africa, Australia and New Zealand. Being geographically distant, the league does not require all teams to play each other which leads to fewer fixtures. Like the Northern leagues, there is a play-off semi-final and final at the end of the season which adds 2 extra fixtures for the best four teams.

Given the sparser fixtures at the top level, there is a condensed second tier of domestic rugby that takes places in the final six weeks of the playing season, after the Super Rugby league has finished. In South Africa, the Currie Cup has 8 teams and accounts for 5 to 6 fixtures. In New Zealand, the Mitre 10 Cup has 8 teams and accounts for 7 to 10 fixtures. In Australia, the National Rugby Championship has 7 teams and accounts for 7 fixtures. Players from Super Rugby sides are selectively allowed to play in this second tier. Junior players are more likely to play all fixtures in this second tier, whereas senior players may only make appearances in key fixtures in order to save their energies for the higher tier Super Rugby competition.

Hypothesis Six decision: accepted. Whilst being a more nuanced fixture landscape than the North, nevertheless even the busiest Super Rugby player will play only 27 club fixtures in a season, which is still less than the least busy professional player in the North who might play 29 club fixtures in a season.

Conclusion: The North Can and Should Catch Up!

In order of benefit from North to South, we have found that:

• The North has much richer leagues (hypothesis 2).
• The South has marginally more players (hypothesis 1).
• The South has a marginally better climate (hypothesis 3).
• The South has a shorter regular season for its top-tier clubs and has pathways with its second-tier clubs (hypothesis 6).
• Southern talent is more cosmopolitan between hemispheres (hypothesis 5).
• The South scores more tries, implying a more positive style (hypothesis 4).

The South has five hypotheses in its favour, however of those two are marginal. Of the remaining three, they are elements which are within the North’s control and can be changed. According to the data, the North needs to adopt a more positive ‘running rugby’, try-oriented playing style, and needs to encourage and provide incentives for its players and coaches to take sabbaticals to the South. The North must also consider a shorter playing season for its top-tier clubs. Transfer of the short 6-week summer break to winter may be worthy of further investigation.

Northern players and coaches must get more creative. Northern fans must expect better from their clubs and national sides. Northern administrations have the power and resources and must strategically evolve the sport to catch up with and eventually even surpass the South.

An example of this forward thinking has been shown by the English Premiership club Harlequins, who in 2018 signed a cooperation agreement with New Zealand rugby to collaborate on ‘playing and coaching resources, training and commercial opportunities’. This remains the only such agreement of its kind in the rugby world, but surely others can and should follow in their footsteps.