đ Can Scott Robertson shore up the All Blacksâ defence this November?
My Week in Sport(s): Autumn rugby, the Fall Classic and Marine Johannès
Welcome to My Week in Sport(s) â a regular newsletter from Plot the Ball.
In this edition:
đ The All Blacks and the impact of selection consistency
âžď¸ Japanâs latest World Series champions
đ An assist from one of the most creative passers in basketball
đ Can Scott Robertson shore up the All Blacksâ defence this November?
Iâve worked out quite a lot of things about my relationship to sport over the period Iâve written this newsletter.
To take one of the examples thatâs most prominent in my mind: itâs teamsâ and athletesâ expression of skill that truly interests me as an observer, rather than winning.
And, for a long time, no sports team interested me more than the All Blacks.
They were renowned for their winning record in the 2010s, but it was the quality of their play that held my attention. I ended up writing pretty exclusively about rugby for a number of years â and, within rugby, focusing on the sport in New Zealand specifically.
As I highlighted a few weeks ago, though, results havenât been quite as good recently.
Theyâre still among the top teams in the world â but no longer indisputably the best.
We can tell that from their recent winning record1. But what does looking more closely at that record reveal about how their underlying skills have changed?
Early in the tenure of the teamâs new head coach is a good time to check in.
And, while Scott Robertson was expected to revive their fortunes, things havenât gone quite as well as his proponents might have expected.
The All Blacksâ attack remains in decent shape.
They scored 31.1 points on average over their last 30 games against âTier 1â opponents â a mark thatâs slightly below the teamâs average for the professional era, but far from out of their ordinary range2.
The defence, on the other hand, is historically leaky: theyâve conceded around 21 points on average in their last 30 Tier 1 tests.
This is the highest mark theyâve recorded in the professional era, and itâs been steadily climbing for a while.
The tenures of Robertson and Ian Foster look particularly stark juxtaposed with those of their immediate predecessors: Steve Hansen and Graham Henry, each of whom led New Zealand to World Cup wins.
The team ended the Hansen era conceding under 18 points per game on average, and finished over 20 with Foster.
Robertson hasnât been able to turn things around yet either.
His only major top-level job before the All Blacks was at the Crusaders, a Super Rugby franchise in his native New Zealand.
Like the NRLâs Penrith Panthers â who we talked about earlier in the year â the Crusaders are a darling of Gain Line Analytics, the Australian consultancy who consistently talk about the strong link between cohesion and defensive performance across sports.
And Gain Lineâs Ben Darwin makes the same point in rugby union as he does in rugby league:
Simply put, the All Blacks just donât have the same level of collective understanding in their side that they used to â and itâs likely to be significantly impacting their performance without the ball.
Their two consecutive World Cup wins â and the success they had against all comers before those major tournaments â came during a period in which they steadily built up cohesion within their side through consistency of selection.
After the second of those wins in 2015 â and the retirement of a number of key players3 â that level of cohesion understandably dropped considerably.
And itâs only recently that itâs started to build back up again from a relatively low base.
How can we measure this?
Iâve looked at the starting line-ups the All Blacks have named in all of their tests against Tier 1 opponents since 2004, and calculated a âsimilarity scoreâ for each one.
That score is based on how similar a given line-up is to those theyâve named in the teamâs 30 previous games.
A score of 100% would indicate that every one of the 15 starters had started all of the teamâs previous 30 games in the same position; a score of 0% would indicate that none of this matchâs starters had played even a single game in their assigned position over that period.
This isnât a perfect measure â it doesnât account for cohesion developed between players at domestic level, for instance, and certainly doesnât correlate perfectly with defensive performance.
But itâs a good gauge of whether a coach is doing whatâs within their immediate control to build what Darwin calls âinterpersonal understanding, system understanding [and] role understandingâ.
What it shows in this case is pretty clear: the line-ups named by Foster became steadily less consistent until the team hit a low point4 at the end of 2022.
Things did improve somewhat from there, and the team put in a creditable performance at the last World Cup â losing the final to the Springboks by a single point.
And Robertson has kept this metric developing in the right direction: up to a âsimilarity scoreâ of 38% in their most recent Tier 1 game against Australia from 35% in that final.
Heâs taken some of Fosterâs key combinations and run with them: his favoured front row of Ethan de Groot, Codie Taylor and Tyrel Lomax, for example, and his midfield of Jordie Barrett and Rieko Ioane.
But, in other areas, the team is still in flux.
Robertsonâs line-up for todayâs match against England includes six players who have started at least half of the teamâs last 30 Tier 1 matches in the same shirt; it also includes six who have started fewer than five of those 30.
This is the first big game of their customary end-of-year tour of the Northern Hemisphere; they also play Ireland and France, who are particularly formidable opponents on home soil.
By the end of the month, Iâll be interested to see whether any more consistent patterns emerge in the head coachâs approach to selection. But it probably wonât be for another 12 months or so that weâll have a clear answer â and that he might start to reap the rewards of such consistency.
There is always external pressure on teams and coaches to chop and change when results arenât up to scratch. Gain Lineâs analysis suggests youâre betting off staying the course in search of improving results, though, and that this is especially true on the defensive side of the ball.
Restoring the All Blacks to the top of the world again will take patience â just like it did the last time.
âžď¸ Run the Numbers
At the end of his first season as a Los Angeles Dodger, Shohei Ohtani ticked off another one of the goals he set for himself as a teenager and became a World Series champion.
His contributions in the final round of the MLB postseason were limited, though â with his performance at the plate declining after his shoulder popped out at the end of Game 2 against the Yankees5.
While he played a comparatively minor part in the final stretch of the Dodgersâ title run, his compatriot â starting pitcher Yoshinobu Yamamoto â was integral.
Ben Lindbergh of The Ringer noted that the Japanese star âdelivered their best start of the World Series and two of their four strongest of the postseasonâ.
And Michael Rosen highlighted one of the most interesting aspects of Yamamotoâs win in Game 2 in a great piece for FanGraphs: his judicious use of the slider6.
He only used the pitch 3% of the time during the 2024 regular season, but was able to lean on it to deal with the most dangerous bats in the Yankeesâ line-up:
âBy saving the pitch for when it was most needed, he maximized its effectiveness and limited its decay. With the fastball, splitter, and curve more than capable of retiring most hitters, Yamamoto was content on Saturday to save the ever-improving slider as an emergency hammer.â
As Rosen notes, itâs rare to see an effective pitch held back for âhigh-leverage spots to the scariest hittersâ â and Yamamotoâs pitch mix will be something Iâll be looking at closely when his second season in the Majors begins next spring.
đ Watch the Games
Earlier in the year, I recommended Rob Friedmanâs âPitching Ninjaâ account on X as a great resource for learning about the game of baseball. This week, I have a second social media recommendation â but this oneâs purely for fun.
French basketball player Marine Johannès has only featured infrequently in the WNBA for the New York Liberty over the last five years.
Sheâs played enough to gain something of a cult following, though, and one aspect of her game has always turned heads: her passing.
Iâm very thankful for the existence of the âOut of context Marine Johannèsâ account on X, which has been feeding me a steady diet of her highlight-reel plays since I discovered it at the start of the EuroLeague season.
Johannès is currently playing for Turkish team ĂBK Mersin â but, despite her obvious passing talent, sheâs not their primary ball-handler7.
Regularly playing off the ball does mean that she gets opportunities to pick passes to teammates with opposing defenders already on the move.
And that movement creates windows of opportunity for a wondrously creative player like Johannès. Take an example from earlier this week.
With around 12 seconds remaining on the shot clock, Johannès is hovering near the logo as two of her teammates consider initiating a pick and roll on the right hand side of the court.
When that action doesnât go as planned, she moves towards the ball and offers herself as a bail-out option. Catching a pass on the run, she quickly identifies that all five Bourges defenders have their eyes trained on the ball â and that Karlie Samuelson is free in the weak-side corner.
In one fluid motion, Johannès scoops the ball with her right hand and fires a pass to her teammate. ĂBK have three easy points, and âOut of context Marine Johannèsâ has yet another highlight to post on X.
You can watch a clip of this sequence here.
The next edition of My Week in Sport(s) will be published in two weeks, on Saturday November 16th.
Four starters in the spine of that side â openside flanker Richie McCaw, fly-half Dan Carter and midfielders Maâa Nonu and Conrad Smith â retired from international rugby immediately after the tournament.
I have this data going back to the start of the professional era in 1996, and the only lower point on record was immediately prior to Graham Henryâs appointment: a score of just over 30% at the 2003 World Cup.
It popped out as he tried (and failed) to steal second base â one of only two steals he attempted across 16 playoff games. He attempted 63 in 159 regular season games this year.
Stephen Sutton-Brown of Baseball Prospectus also followed the piece up with an interesting thread about the âdetectabilityâ of Yamamotoâs slider on X.
Sheâs listed as a shooting guard, and was one of three ĂBK Mersin players averaging 5.0 assists per game in EuroLeague play through Wednesday.
Can you share how you calculated the similarity score please?
I have all of the Ireland lineups from nearly the last 20 years and I'd love to see how Ireland compares
But then again, Erasmus and the Springboks seem to be getting rewarded for a policy of regular selection gambles and changes. I'd be curious to compare the recent Springbok consistency score to the NZ one.