The July tests against Fiji allowed the All Blacks to find rhythm with their set-piece attacking play as the visiting Fijians made committed tackle attempts with bruising force.
The downside for the Fijians was their commitment to making tackles often came a touch too late, allowing the All Blacks to find space and construct tries from set-piece plays in early phase counts as defenders took themselves out of play.
All of the tries in Dunedin bar one came from set-piece launches, and with Richie Mo’unga back in the line-up in Hamilton, it was more of the same.
As the All Blacks returned to Bledisloe Cup fixtures against the Wallabies, the return of counter-attacking ‘click’ plays became more prominent.
A tougher defence meant scoring from set-piece would be more challenging, while in recent years the All Blacks have a track record of beating the Wallabies using their own ball.
No side dines on Wallabies errors quite like the All Blacks, and this year was no different.
All Blacks tries from transition plays, off errors, kicks, turnovers, made up 44 percent of all the tries scored against the Wallabies over their three tests. These strikes came quickly within one or two phases of the genesis moment of winning back possession.
Phase play tries also increased to 33 percent, as they had to work high phase counts to break the Australians down outside of those transition moments. This was where having Aaron Smith was invaluable in two tests at Eden Park.
If the set-piece launch pattern didn’t yield a result immediately, Smith’s control of the ruck and distribution would break down the defence eventually, something that was missing when the side travelled across to Australia.
Just 22 percent of the tries came from set-piece where they were able to score within three phases of the scrum or lineout, an indication of the difference between Fiji’s defence and Australia’s.
Against Los Pumas, the All Blacks again found more set-piece success as the undercooked Argentinians struggled to withstand them.
With little possession to give away on attack, New Zealand’s transition tries dried up again with just 20 percent scored in that fashion.
Both matches never troubled them, but the All Blacks weren’t completely on top of their game with ball-in-hand leading to many missed try-scoring opportunities and patchy periods of play.
When it came time to play the Springboks, the transition plays were unlikely to come due to South Africa’s style of play.
The high number of kicks meant low phase counts to turn the ball over, and the Springbok kicks went up not deep, making it hard to open up play on kick returns.
Receiving a contested kick is hard to turn into a productive counter-attack as most of your own players are retreating and the kick-chase will likely swallow up the catcher quickly in the traffic around the landing area.
Springboks handling errors or breakdown turnovers were another low probability event. They generally carry the ball well and rely on low-risk, one-pass phases often, even from set-piece where Damian de Allende usually takes the pill, before kicking not long after.
How the All Blacks defeated the Wallabies over three tests was never going to be an option against the Springboks.
It would come down to how potent the set-piece attack would be and how much game speed they could muster over long phase counts, which, over the two tests against Argentina, was not that convincing.
If they wanted broken field opportunities against a defence in transition, it would have to come from their own kicking game. The side kicking the ball, with all the players moving forward, has a much better chance of creating something should they regain possession.
The two tests against South Africa put the spotlight on more tactical, planned aspects of the game. In that respect, what the All Blacks tried to do in the first test against the Springboks made Dave Rennie’s Wallabies look a lot smarter by comparison.
One of the standard plays the All Blacks have been using is direct ball to the 12 who has options wrapping around the back or the flat option to the outside centre. The play is often used inside the opposition’s 22 in a prime attacking zone.
During the early stages of the first test against the Springboks, the All Blacks showed intent to use the maul before using their base play around David Havili when they were in this area of the field.
The All Blacks played into South Africa’s hands multiple times by taking the ‘free ball’ at the front of the lineout, which the Springboks had left deliberately for them.
The only problem is trying to maul on the Springboks from the front of the lineout is a nightmare.
The Wallabies found that they would try to force the maul over the sideline every single time, which the Australians then used against them by running peels with Michael Hooper (7) detaching almost immediately.
The All Blacks weren’t as smart and persisted with the maul, which was sacked time and time again.
The key to the sack was Eben Etzebeth (4) targeting the closest lifter to him, Scott Barrett (5). By pulling Barrett’s leg up, Etzebeth removed one of the foundation pillars of the maul, making it ripe for collapsing.
With his foot off the ground, Barrett is helpless to stop the drive of the Springboks spinning the maul to its eventual collapse.
The Springboks were penalised for entering from the side, but not for Etzebeth’s leg lift. The Springbok lock confirmed this with a conversation the referee, which meant he would definitely try it again.
The All Blacks got the same defensive set-up from the Springboks and took the front option again with a throw to Blackadder, while Etzebeth targets Barrett once again.
Once Barrett hit the deck, the maul couldn’t go anywhere. After the maul was stopped twice, the All Blacks finally have to play it out to the backs with their base red zone play around Havili.
Running a smaller 12 in David Havili up into the middle of the Springbok midfield might not seem that wise on the surface, but that’s what they tried to do anyway.
On this occasion, it wasn’t Havili’s size that was the main problem – it was his decision-making and change of angle that only served to unravel the play.
Rieko Ioane (13) had to run a tight support line to bait the defence inward, but Havili tried to bounce to the outside and ran into his path, nearly causing a collision with his teammate. He did not stay square and direct as he should have.
Running into Ioane’s path meant the outside centre was unable to clean out properly. Once in Handre Pollard and De Allende’s clutches, he was dragged sideways, complicating matters for the arriving cleaners.
The counter-ruck by Siya Kolisi blew the All Blacks off their own ball as the cleaners enter from the side, and the Springboks captain forced a turnover on TJ Perenara after the first phase from the maul.
The better option for Havili would have been to play Beauden Barrett (10) wrapping around the back, much like against Fiji when the release out the back led to a try to Jordie Barrett.
They did realise this and adjusted the next time, with Havili making a good read the second time around which led to a yellow card to Sbu Nkosi for slapping down the pass. Quicker hands from Beauden Barrett would have likely seen the All Blacks score.
Unfortunately, this would be the only time the All Blacks tried to attack wide directly off set-piece in the entire match in this zone, despite it providing the most evidence that this was the weakest part of the Springboks defence.
This evidence had also been collected by the Wallabies over the previous two weeks, who scored all of their tries via breaks down the edges.
With Nkosi in the bin for the first five minutes of the second half, the All Blacks seemingly did not have the brains to find a way to exploit that missing man on the right edge.
With five lineouts inside South Africa’s 22 in that period to start the second half, they tried to maul every single time, before carrying off 9 and getting narrow and losing possession if they hadn’t been sacked or pushed into touch from the maul.
The decision-making in that period – not the decision to turn down three to kick to the corner, but the types of plays called, all mauls, and the inability to identify where the weakest part of the defence was and attack it – left a lot to be desired.
It was after that where the All Blacks finally made adjustments to their lineout.
In midfield zones, the All Blacks persisted with their launch plans around Havili through the match, which resulted in four turnovers on the first phase.
The first turnover came from a play that uses the blind winger Will Jordan (14) at first receiver and Havili as a supporting runner.
As a big, speedy carrying option, Jordan can either crash the ball up himself or play his bail out option Havili (12) short, again into the 10-12 Springbok channel.
Havili ran a great line on this play and sliced through the first-up tackles of Pollard and De Allende, but Jordan and Blackadder (7) both stumbled in their approach to the ruck and Duane Vermeulen (8) pounced on the isolated midfielder.
The first two launch plays used first phase hit-ups by Havili into the teeth of South Africa’s midfield where their back row were waiting, who then forced turnovers.
It really wasn’t that smart, particularly with a rookie and developing openside tasked with protecting New Zealand’s possession against a vastly more experienced Springboks back row.
But, to understand why the All Blacks kept running plays into the Springbok midfield and in and around the strong back row, we have to rewind.
The simple answer to this question is because they have always run these plays under head coach Ian Foster.
These plays have been a part of the All Blacks playbook for years, we saw them earlier this year in the Fiji series, last year with Jack Goodhue at 12, and before that with the likes of Sonny Bill Williams and Ryan Crotty.
There are a handful of plays that the All Blacks rotate through and different opponents can end up getting the same plays.
With an option-based play like this around the 12, the All Blacks must believe that one option will present as the right one, and any player inserted into the role will be able to come up with the right decision.
However, the reality is what Havili sees versus what Goodhue or Williams sees is likely to all be different, and with all different skill sets, you will get a range of different outcomes. Particularly when the read requires a run, what Havili will produce is not going to be the same as a Laumape or Williams.
When you add the variables in the defences they come up against, there are bound to be times it falls apart. And that was the case in the first test and at times in the second, where Havili was up against a very strong Springboks midfield.
There are particular launches that the All Blacks like to run against South Africa, like this tight formation on first phase using big bodies in small space around the 10-12 channel. This starter play was brought back for the second test on the Gold Coast.
Used here in 2018, the first-five, Beauden Barrett, had a hard running 12 off his hip, in this case Anton Lienert-Brown (13), and his big winger slightly in behind in Ioane (11).
It is a very direct approach that uses a lot of running momentum to achieve gain line and a ruck not too far from the lineout, enabling support to get there easier.
Crashing the ball into this part of the defensive line on first phase allows for optionality across the field with the rest of the launch pattern.
There is lots of space going the same way for forward carries around the corner before setting up ’21’ or ’31’ patterns switching back to the edge.
When you have a big, tall winger like Ioane and a blockbusting midfielder like Williams, it is a handful for the opposing flyhalf to cover and you stand a good chance of getting front foot ball to play from, with plenty of width going the same way to use on latter phases.
It is a little bit of different situation when your left wing is replacement back Damian McKenzie, who doesn’t play wing and is one of the smaller players running around in test rugby.
On Richie Mo’unga’s (22) first play against the Springboks coming off the bench, the All Blacks tried to run the starter play that they had employed throughout the match.
McKenzie (23) was a long way off the pace from being a valid option for Mo’unga, who decided to pull it back and take on the defence himself.
He was eaten alive by Elton Jantjies and Malcolm Marx, driven a long way backward in the two-man tackle, leaving the launch devoid of any first phase momentum.
Perenara had to take evasion action to avoid colliding with the back-peddling tackle, while Savea and Lienert-Brown can only watch on until Mo’unga hits the ground. The launch pattern is over at this point before really getting going.
Running this play with Mo’unga, McKenzie and Lienert-Brown in 2021 yielded a vastly different result than with Mo’unga, Ioane and Williams two years ago in 2019.
But that’s what the All Blacks do at times, fit square pegs into round holes by re-using the same game plans, rather than working out new plays entirely around the pieces they have, or the opposition they are playing.
They ran a lot of the same plays against the Springboks as they did against Fiji with their options-based attack plays.
Theoretically it can work against both, but, against a different opponent with different strengths and weaknesses, it can just as likely come unstuck, and that couldn’t be clearer after two games against the Springboks.
Little consideration seemingly was given to what is the best way to break down the specific defence in front of them, which the Wallabies gave the blueprint for.
The Wallabies stayed away from the maul while also baiting the South African pack to commit there. They attacked the seam at the lineout tail relentlessly, making huge gain line wins on first phase to set-up their multi-phase plans.
On the two examples below, the tail defender was drawn into contact, leaving Pollard’s inside shoulder unprotected for Marika Koroibete (11) and Samu Kerevi (12) to target.
After massive gain line runs, they then had front-foot ball to work with and broke down the South African defence going the same way before a strike play through the backs to target the edge.
There is no surprise, then, that the New Zealand’s best set-piece play of the two games targeted this area and exploited the same weak spot at the lineout seam.
Using a more complex lineout call that including a slip and fake jump, the All Blacks got clean ball off the top at the back through Hoskins Sotutu, allowing Brad Weber to get some space to run and draw Mbonambi.
The All Blacks have called the same first phase play with the 10-11-12 running hard in short space, but this time the seam opens up for Barrett.
As highlighted by Nick Bishop, Mbonambi tackled Weber (9) and took himself out of play, leaving Pollard’s inside unprotected, as had been happening against the Wallabies.
Instead of passing to the two support options Barrett (10) took advantage, stepping inside Pollard and breaking through to the other side. With Savea (7) winning the space in front of Kolisi, there is no South African cover to tackle the All Blacks first-five.
The supporting players found a way to link up with Barrett and after a brilliant run by Ioane, Savea scored one of the best All Blacks tries of the season.
It may have only been Barrett’s vision and instinct that exploited the weakness, rather than the design of the generic starter play. He had options by design, and one of them was to run, but that is as much as you can correlate to the outcome.
The try highlighted the possibilities for the All Blacks had they schemed deliberately for what they were facing, instead of rolling out Foster’s same old menu and trying to take on the Springboks in areas of strength like the maul.
Based on their two games against the Springboks, Rennie and his staff looked far superior to the All Blacks brains trust when it came to game planning for the world champions.
They tailored their game to the opposition to attack the weaknesses, whereas the All Blacks naively went brazenly forward trying to attack South Africa’s strengths and it backfired.
This should be a key lesson for Foster’s All Blacks moving forward. In the organised parts of the game, they cannot change the pieces performing the roles, run the same plays against everyone and expect them to work.
There needs to be some flexibility there and more thought into the process.
The fact that they were running the same starter play with McKenzie on the left wing showed a certain level of naivety that, to no surprise, blew up.
In tight test matches where counter-attacking opportunities dry up, the All Blacks will need to have a clinical set-piece attacking plan which, at times, needs to be adjusted for the opposition.
They might have the most attacking talent in the world, but they still need decent plans to be devised to help them get the match-ups and space they need to let their skills shine.
You cannot roll up and play the same against Fiji as you do against the Springboks and expect to win handsomely.
It could work, or it could just as easily blow up like it did twice in the Rugby Championship.
Join RugbyPass+ now to continue reading this article.
Access our new premium content area bringing you the highest quality rugby content from award-winning journalists, opinionated pundits, leading coaches and the biggest stars in the game.