• hours
  • minutes
  • seconds
  • Share

Ghana-U.S. game rates as even

June 26, 2010
By Nate Silver
Special to

We've been rather happy with the performance of the Soccer Power Index thus far; it predicted, for instance, the ascendency of the South American teams that we saw in the group stages, as well as the fact that both France and Italy were vulnerable.

Now that we're in the knockout stages, the football should speak for itself, and I don't think anybody should be particularly concerned about what a computer program says. With that said, here is how SPI views the probability of each team advancing into the quarterfinals.

Uruguay (66 percent) versus South Korea (34 percent). What a weird team Uruguay is, with a strong back four and some talented strikers like Diego Forlan. It is perhaps a bit weaker in midfield and often seems disinterested in controlling possession, instead preferring to counter attack. It will make for an interesting contrast against fit, fast, offensively minded South Korea.

But whatever Uruguay has been doing, it's been working; Group A was quite tough, especially since Uruguay played France when France was still trying and made it look easy. It's reflected well, frankly, on SPI, which recognized how difficult it was for any team to advance out of CONMEBOL qualifying and saw the Uruguayans as dark horses going into the tournament. With a rather favorable draw in this match as well as potentially in the quarterfinal, Uruguay has a clear path into the semifinals. With that said, South Korea is not bad and has about a 1-in-3 chance of pulling off the upset.

Ghana (51 percent) versus United States (49 percent). I've written about this matchup in more detail at my blog, but for those of you who are wondering why SPI sees this game as a toss-up even though the USA ranks considerably higher in SPI overall, there are two reasons.

First, we're still giving Ghana a bonus for playing on its home continent. If you removed that bonus, the United States would be about a 60-40 favorite. Since the African teams have been underwhelming in this tournament, I can empathize with the point of view that this bonus is undeserved. That said, there was pretty clear evidence of the home-continent effect in past tournaments -- certainly in both 2002 and 2006 -- and if it were going to have an impact, you'd think it might be here, with Ghana being the lone remaining representative of Africa.

Secondly, SPI's match predictions are based not strictly on a team's overall score, but instead on the two component rankings that we generate, for offense/attack and defense. This is a very skilled and entertaining United States team, but also a very attack-oriented one that sometimes can be too liberal about providing opponents with scoring opportunities, especially early in the game. While the USA rates eighth in the world in goal scoring, according to SPI, it is just 38th in goal prevention. The reason this matters is that we found in developing SPI that the defense rating tends to be somewhat more accentuated in games against other strong teams, especially under knockout conditions where the game could go to penalties. It's not a huge deal by any means -- and I certainly do not think the United States should change its approach to the game -- but there are definitely scenarios in which the U.S. plays great football but loses 2-1 by giving Ghana (whose attack has been somewhat disoriented) a chance or two it doesn't deserve.

In relative terms, this is obviously a pretty favorable draw for the United States -- avoiding the major world powers, particularly Spain, Brazil, Argentina and Holland, until at least the semifinals. But that doesn't mean, in absolute terms, that it's going to be easy; nothing is easy in the knockout stages of the World Cup. If the back four play as well as they did against Algeria, the U.S. should be in good shape, but if the team takes the match for granted, it could be in trouble.

Netherlands (89 percent) versus Slovakia (11 percent). This is the one real mismatch in the round of 16. With no disrespect intended to Slovakia, which played valiantly to advance, qualifying out of dysfunctional Group F is a bit like making the playoffs out of the NHL's old Norris Division. Conversely, there's been very little to critique for the Dutch thus far. While they could be caught in a complacent mood looking forward to their match against Brazil, or otherwise fall victim to a bad bounce or two, they're the overwhelming favorites here.

Brazil (66 percent) versus Chile (34 percent). Another SPI dark horse, Chile looked solid in the group stages, but Brazil surely was pleased to draw it rather than Spain. That does not mean, however, that Brazil will not have to work hard for its victory; had Chile converted on one or two more of the innumerable chances it generated against Switzerland and Honduras, it would have won its group. Meanwhile, we've seen only one vintage performance by Brazil -- against Côte d'Ivoire -- whereas it played a poor first half against North Korea and looked indifferent toward the outcome of the game against Portugal. The counterargument against this game being competitive is that Brazil won very comfortably in the two matches the nations played in CONMEBOL qualifying (outscoring Chile 7-2).

Argentina (67 percent) versus Mexico (33 percent). A lot of games in the round of 16 are shaping up about like this, with the established soccer power about a 2-1 favorite against an up-and-coming team. Statistically speaking, Mexico is tough to get a handle on since it plays tons of matches but not always against the best competition, nor with its best lineups. But Mexico certainly played plucky football to advance out of Group A, and should give Argentina a very competitive and entertaining game.

England (55 percent) versus Germany (45 percent). Germany has the right to curse its luck a little, having played fairly solid soccer in the group stages but drawing England and then potentially Argentina in the quarterfinal. Nevertheless, it's England that SPI sees as the very slight favorite to advance. If there's a concern for England, it shouldn't be about the goal Robert Green let in against the United States (especially since he's been replaced by David James) but rather that player fitness has been a bit questionable, which could be problematic if the game goes to extra time (never exactly an English specialty). This game should live up to its billing, and someone will go home very unhappy.

Paraguay (64 percent) versus Japan (36 percent). Paraguay was looking like a somewhat clear favorite, but after a somewhat listless performance against New Zealand -- and conversely a very strong one by Japan against Denmark -- the odds have evened a bit. In contrast to up-tempo Chile and the passive-aggressive Uruguayans, Paraguay is a bit boring to watch and the tournament might not suffer were Japan to pull off the mild upset. Of course, it obviously would have been folly at this point to bet against the South American team.

Spain (55 percent) versus Portugal (45 percent). Does Portugal deserve to be considered among the world's elite teams? SPI seems to think so, but it's been a weird tournament for the Portuguese. There was no shame in piling on a bit against North Korea when goal differential is a criterion used to differentiate teams, but they looked a bit flat against Ivory Coast and Brazil. In spite of the flashy 7-0 scoreline against North Korea, SPI actually thinks goal prevention is the relative strength of the Portuguese; they rank third in the world in their defense rating but sixth in attack. There are fewer questions about Spain. Quantitatively, it should not have lost to marginal Switzerland, but qualitatively, it played at least as well in that game as in victories over Chile and Honduras.

Nate Silver is a renowned statistical analyst who was named one of "The World's 100 Most Influential People" by Time Magazine in 2009. He gained acclaim for outperforming the polls in the 2008 U.S. presidential elections and created baseball's popular predictive system, PECOTA.