The four primary moments of most field sports are structured attack, structured defence, unstructured defence, and unstructured attack.
Rugby is no different and, if anything, it more closely embodies these concepts than most sports.
This isn’t new information, though; the idea of “counter-attacking” rugby is engrained in the sport. I went back to watch older rugby games – selected games from the ’70s, ’80s, ’90s and early 2000s and the one thing that stood out in the older games was how many transitions there were in the older, pre-professional version of the code.
There were far more set pieces back then – the 1987 World Cup had an average of 32 scrums and 45 lineouts per game – but there were far more periods of “unstructured” transition rugby that lead to those set pieces.
For the first three World Cups, the average tackle success rate was 75% on an average of just 52 tackles per team. By comparison, the average tackle success rate in the 2007 World Cup was 87% on an average of 84 tackles per team, per game.
In the 2019 World Cup, that tackle success rate was 84% but with an average number of 129 tackles per team per game.
This also tracks when you consider the average number of rucks per game across the recent history of the code. In the 1987 World Cup, each team averaged just 25 rucks per game. That is not a typo. It took until the 1999 World Cup for that number to double, approximately, with an average of 57 rucks per team per game. It then increased year on year.
By the 2015 World Cup, the average was 83 rucks per team per game. Now, the average ruck numbers have come down very slightly in the last two World Cups in 2019 and 2023 – only by an average of two or three rucks per game – but they are still almost treble the ruck numbers from the first few iterations of the game at a World Cup level.
And this is the last bunch of silly numbers I want you to look at, courtesy of Opta. The link between turnovers and possessions per game.
| Rugby World Cup | Matches | Possessions per game | Avg. Phases per Possession |
|---|---|---|---|
| 1987 | 32 | 134.2 | 1.4 |
| 1995 | 32 | 114.4 | 1.7 |
| 2003 | 48 | 106.9 | 2.3 |
| 2011 | 48 | 81.4 | 2.9 |
| 2019 | 45 | 84.8 | 3.0 |
| 2023 | 48 | 85.5 | 1.8 |
What does this mean? As you would expect, phases per possession have gone hand in hand with the growing ruck numbers. 2019 was the highest in the test game for phases per possession which, in an environment where overall possessions have been declining since the inaugural World Cup, told World Rugby that the game was getting too “static”, especially since Ball-In-Play time was ballooning.
What does this mean in reality? As Ball-In-Play time has increased, we’ve seen far more rucks, far fewer possessions, fewer set pieces and fewer turnovers.
In 2023, something unusual happened – average phases per possession dropped to a level last seen in the 90s but possessions per game only increased by under one possession per game.
So what happened?
Teams decided, pretty much en-masse and more or less at the same time during the knockouts, that they were better off without the ball. The Springboks went through the World Cup knockouts with an average kick to pass ratio of 1:2.5 across the three games.
In the semi-final against England, both sides had a kick to pass ratio lower than 2.5 – England’s was 1.9. Both sides had a pass per carry ratio approaching 1.0, which in basically means the following; both teams kicked the ball all day and, when they passed the ball, they barely moved it wider than one pass before taking contact. The rain in Paris during that knockout series played a part, of course, but World Rugby were unhappy with the “spectacle” of those games, something echoed by coaches and media.
Ironically, one of the biggest critics of that tournament was… Warren Gatland during the tournament itself when he was coaching Wales.
The 2023 World Cup featured more kicking than any World Cup since 1995. The biggest signifier in the game was the reversion of the ELVs of 2009 which made the average number of kicks per game at the 2011 World Cup drop by an average 14 kicks per game compared to the 2007 World Cup – widely panned for broadly the same reasons that the 2023 World Cup was. Too much kicking, too much defence, and the game is too slow.
By the 2015 World Cup, the average number of kicks per game was at an all-time low but shot up in at the next two World Cups, peaking in 2023.
This was entirely practical. As Gatland went onto say;
“The dominance of kicking in the game stems from the fact that there is currently a really fine balance between attack and defence in terms of risk and reward and sometimes the advantage is not to have the ball because defences are so well organised. […] And at the moment the game is defence focused. Teams are not prepared to play in their own territory, or take the risk of playing for two, three or four phases and then conceding a penalty, which is kicked into your 22 and you are potentially under pressure for four or five minutes.”
So why did phases per possession go down? Because teams were kicking at a rate we hadn’t seen in the 90s and, rather than trading possessions back and forth, both teams knew that the only practical response to that kicking volume is… more kicking of your own.
That is what the analysts tell you is the most pragmatic way to play the game. Their job isn’t to create a spectacle – it’s to win rugby matches. World Rugby have a different view. If anything, the 2023 World Cup copper-fastened what they had already come to believe after the 2021 Lions tour – that the game needed to change.
This isn’t the first time that the meta of the game as it had become required a tweak, so far as World Rugby saw it.
The ELVs of 2008 (implemented officially in 2009 but trialled during the entire 2008/09 season) were broadly introduced to speed up the game in the aftermath of the 2007 World Cup and, to a certain extent, make the way that teams like Munster had exploited the meta of the game to essentially grind to victory through narrow ball retention, penalties usually forced from teams trying to win possession back – virtually impossible at the time – illegal. World Rugby – then the IRB – wanted to reduce stoppages, and encourage more attacking play by focusing on changing how rucks and mauls operated, and other aspects of the code that would keep the ball in play and reduce penalties. Sound familiar?
The ELVs weren’t fully implemented by the time they made it into law, but they did have the effect of sending kicking volume per game on a massive downward trajectory.
Even then, this needed to be adjusted in 2010 with a law tweak that demanded all tacklers must now release the ball carrier to be awarded a steal at the breakdown. That had been the Brian O’Driscoll defensive special for a time.
As Lyndon Bray, the Sanzar referee manager in 2010, told the Dominion Post as quoted in this New Zealand Herald piece: “We’ve agreed philosophically to change what the tackler can and can’t do. We’ve allowed … him to remain in contact with the ball and ball carrier. After he leaves his feet … he stays on the ball and jumps up and rips it away. It’s actually against the law … we’ve agreed the tackler must [now] release everything when he goes to ground and not hold on as he gets to his feet.”
Up to that, despite the ELVs, the threat of being “isolated” on a ball carry meant that most teams played narrow, tight rugby to ensure that rarely happened.
So what was it about the game as it became post 2015 encouraged kicking to return?
It’s quite simple; a fundamental tenet of the game is that playing the game away from 22 and closer to the opposition’s 22 is massively overpowered in a game where penalties are overpowered. If you are able to kick the ball into the opposition’s half of the field and defend them there, you are incrementally more likely to either win a turnover or a full penalty that is worth three, five or seven points. In a tight game, that’s the difference between winning and losing. In a world where the jackal turnover is legal, there is very little downside to kicking the ball away and a huge upside to defending a team in their half of the field.
When you combine that with scrummaging that can generate match altering penalties off any handling error, you can see why off ball rugby became and stayed a hugely overpowered element of the game.
***
So what are World Rugby telling us?
We have to look at the law trials and the law tweaks – they are two different things.
The spirit of the new law trials seems to be about speeding the game up – a common trope – through the use of shot clocks on penalties and conversions but one that stood out to me is the protection now afforded to the scrum-half during scrums, rucks, and mauls. What does this tell us? That World Rugby want to make it easier for scrum halves – or any player at the base of the ruck – to play the ball without immediate disruption.
When we combine this with the new “anti-escorting” law on kicks, I think the full picture becomes clearer.
You’re never going to get rid of kicking’s inherent advantage without also outlawing the jackal turnover, so why not make contestable kicking more than a sequence of teams trading kicks to screened receivers over and back again, something we saw in the data at the 2023 World Cup?
It’s leading to more kicking at the moment, especially from teams with scrums who can generate penalties – there’s a correlation between an increase in contestables and scrums – but in the longer term, I think it’s leading to a shift in focus from teams. Maybe we won’t see it this season, but in the next year I think it’ll become clear.
I contend that World Rugby wants to radically shift the sport’s focus from kicking and defence to a kicking-focused attack.
If changing the bonus for taking territory without the ball is impossible, and you don’t want to remove kicking as a beneficial tactic, you must gradually remove the defensive weapons available to teams.
I believe that after the 2027 World Cup, World Rugby will implement no tackling above the waist as a safety measure in light of the dangers of concussion – both morally and legally. While they’re at it, I think they will also look to remove the jackal turnover from the game at the same time. Players like Dupont getting taken out of the game for 9 months by the poaching action and “clean out” counter action is reason enough on its own, but there are other reasons too. The flow of the game chief amongst them.
I think in the next two seasons in the buildup to the 2027 World Cup, World Rugby will continue to nerf aspects of the kick pressure game as they relate to defence and in-depth defensive structure powered by longer kicking. I believe the change in the escorting law was intended to shorten kicking while also discouraging the box kick by removing pillar blockers. We’re already seeing a drop in that tactic season on season, albeit with England as a current outlier in that regard.
The Sharks have the best kick retention in the URC byt they only retain 20% of them. That means 80% are in active transition.
This is the game World Rugby want and they will legislate to get the game they have openly stated they desire – a game where possession flows between both sides, that reduces the physical impact on players and makes it so that overly sized teams have an incremental advantage.
Kick pressure and off-ball rugby is dead. At the moment, it doesn’t realise it’s dead, but the end is coming.
Kicking isn’t dead, and will never be unless we remove two players from the field on each team, so that means we’re entering the world of transition rugby. That means offloads, it means ball-in-hand and it means short-form kicking and expansivity.
Understand what World Rugby want and get ahead of the curve. The last thing you want – like Munster in 2008 – is to be an example of what World Rugby want to get rid of.
