The Group of Death Round of 16 of Death?

By: Kevin Chroust | March 27, 2018

In the unlikely event of World Cup group stage turmoil, prepare early for #BRAvGER 3.0 and/or #ARGvFRA 3.0. STATS win probability shows the likelihood of one or both happening earlier than expected, and neither scenario is out of the question. Whether it happens has as much to do with countries such as Mexico and Croatia as it does with the perennial powers.

 

Brazil vs. Germany. Argentina vs. France. They’re World Cup matchups that have both happened twice before. They’re matchups we want. They sound like realistic World Cup finals for Russia 2018, and the STATS predictive analytics team supports that with its tournament win probabilities.

But based on how each country is situated in group play, those matchups could theoretically happen at the start of the knockout phase, which has never happened before with the specific ties in question. In fact, only one combination of those four countries has ever happened in the Round of 16 – Argentina-Brazil in 1990.

A similar shakeup nearly happened four years ago, thanks in part to Mexico punching above their projected weight. It’s a bit early to put their Group F opponents Germany on alert, but El Tri could similarly have a hand in any 2018 chaos and disrupt the plot.

Recall June 17, 2014 – if you can. If not, you’re not alone. Here’s a refresher: On paper, it stands as one of the more inconspicuous match dates in Brazil’s comprehensive World Cup history, and one the host country’s opponents from that day are more likely to remember. It was a scoreless draw with Mexico, a match in which El Tri goalkeeper Guillermo Ochoa did things with his body to keep the ball out of his goal that no football-simulating model can adequately predict.

Based on the chances Brazil created, their expected goals came in at 1.53. Mexico’s? 0.44. Round up and down accordingly, and that’s right in line with the pre-match calculations of the model put to use by the STATS predictive analytics team, which projected a 2-0 Brazil victory with the following odds for each possible result:

(Graphics by Stephan van Niekerk)

But Ochoa kept a few glaring attempts from crossing the line, and the countries split the points. It cast doubt on whether Brazil would win Group A as was expected all along. After the match, Brazil and Mexico were tied atop in Group A on four points each with Brazil holding a one-goal advantage in goal difference. If that swung Mexico’s way on the group’s final match day, Brazil would have met eventual third-place opponents the Netherlands in the Round of 16. The way it shook out was both teams finished with seven points and Brazil going top with a plus-5 goal difference to Mexico’s +3.

Brazil went on to defeat Chile on penalties in the Round of 16, while Mexico bowed out to the Dutch after grabbing a 48th-minute lead and eventually losing on a very controversial stoppage-time penalty conceded to Arjen Robben, whose prolific left foot ranks a close second in the winger’s arsenal of football tools behind only his ability to fall on the ground.

Nearly four years later, the same predictive team is back at it, conducting more than 100,000 simulations to come to initial win probabilities for all 32 countries among their individual matches and groups, as well as odds to win the entire tournament. It’s all available to media partners to help create more dynamic fixture listings and help craft stories like – well – this one.

For 2018, it shouldn’t come as too much of a surprise that, in the first series of simulations following the draw, Brazil are back leading the tournament favourites followed closely by defending champions Germany. France are third, and Argentina aren’t far behind in fifth with 2010 champions Spain between.

But what’s maybe most interesting to consider here is just how important it is for each of the first four teams we’re focusing on here to follow through on STATS’ projection of them winning their groups. That’s because the way the Round of 16 is structured, the winners of one predetermined group play the second-place side from another. That means the winners of Group C – France, Australia, Peru and Denmark – will play the runners-up from Group D – Argentina, Iceland, Croatia and Nigeria – and the winners of D will play the runners-up of C. The same goes for Group E – Brazil, Switzerland, Costa Rica and Serbia – and Group F – Germany, Mexico, Sweden and South Korea.

So Mexico – equipped with what’s said to be one of their best teams in years after cruising through qualifying – will play a role in this once again. What’s the likelihood of El Tri or someone else in one of those groups having a hand in bringing those must-see matchups to us earlier than we might want?

There’s a 30.1 percent chance of Argentina and France meeting in the Round of 16. And there’s a 25.3 percent chance Brazil and Germany have their rematch of the infamous 2014 semifinal two rounds earlier. That puts the probability of both matches coming to pass in the Round of 16 and four of our top five ranked teams meeting much earlier in the tournament than we’d expect at 7.6 percent.

Mathematically, the most realistic way for the first to happen is for Croatia, the No. 15 team in FIFA’s world rankings and ninth-ranked team in STATS’ tournament projections, to win Group D ahead of Argentina. Croatia, led by Luka Modrić, Ivan Rakitić and Ivan Perisić, have a 23.6 percent chance to winning their group compared to 61.7 for Argentina. But those odds could change substantially in their second group stage match if Croatia are able to get at least a draw:

The Germany-Brazil rematch is less likely because both sides are heavy favourites to win their groups with Brazil leading all teams at 79.5 percent and Germany at 74.9, which is the third-highest likelihood after France (75.6).

The match most likely to shake up those Round of 16 matches between Group E and F? It doesn’t involve Brazil. The Seleção have at worst a 70.7 percent chance of beating each of their group stage opponents. Rather, it again has to do with Mexico, No. 17 in FIFA’s world rankings and 14th among STATS’ tournament projections:

Quick math tells us that’s a 32.9 percent chance of Mexico getting a result capable of rewriting the expectation of the group. It might not sound like much, but recall 2014’s 26.2 percent chance against Brazil.

Get Croatia in on the act, and what’d come next is a true Round of 16 of Death.