Hey, remember this post from earlier today? Now that the draw’s done, let’s apply the “Soccernomics” multiple regression stat to each group and see who, statistically speaking, should win.
For those of you just joining us, these are the criteria:
- Home field advanatage is worth 2/3 of a goal.
- Having twice as much international experience as your opponent is worth about 1/2 a goal.
- Having twice the population of your opponent is worth about 1/10 of a goal.
- Having twice the GDP per capita of your opponent is also worth about 1/10 of a goal.
We’re probably not going to see teams winning by huge margins, of course…but this will give us an idea of what to look for.
GROUP A
South Africa, Mexico, Uruguay, France
Match #1
South Africa v. Mexico - If having twice the experience as your opponent is roughly half a goal, than Mexico (with 554 international matches from 1871 to 2001) is completely going to own South Africa (with 140 in the same time period). I’d anticipate Mexico winning this 2-1, including the South Africa homefield advantage. Mexico doubles them up on population, per capita, and almost quadruples their experience.
Uruguay v. France - Based on population and per capita income, France has a nice edge (although Uruguay has more experience). Again, expect 2-0 to France
Match #2
South Africa v. Uruguay - This one has a slight edge towards the home country based on homefield advantage and population; I’d tentatively call this a 1-1 draw or 2-1 South Africa win.
Mexico v. France – Too close to call on the factors available; I can see a scoreless draw or a 1-0 win for either side. For our purposes, since neither team technically has enough statistical advantage, I’m going with a scoreless draw.
Match #3
Mexico v. Uruguay - On population alone, Mexico’s good for about three goals; statistically these two are even in the other categories. Call it 3-0 to Mexico.
France v. South Africa - This SHOULD be a blowout (4-0 to France). France could choke, of course, but if they score less than two I’d call it underachieving.
End result: France wins the group, Mexico comes in second.
GROUP B
Argentina, South Korea, Nigeria, Greece
Match #1
Argentina v. Nigeria – Based on nothing other than the criteria we’re using, this should be a 2-0 win for Argentina.
South Korea v. Greece - This one was surprising: South Korea should win 1-0. They have four times the population, almost double the experience, and the 1/10 of a goal Greece gets for almost doubling their GDP per capita doesn’t cancel out the other stuff.
Match #2
Argentina v. South Korea – Scoreless draw. South Korea has surprisingly good stats for this metric.
Greece v. Nigeria - Greece should win 2-1 if the GDP figure means anything at all (although Nigeria’s huge population would’ve helped if they’d have drawn someone smaller).
Match #3
Greece v. Argentina – Argentina wins 2-1. They’ve got a bigger population and crush Greece on experience, but Greece pulls one back with the GDP.
Nigeria v. South Korea – I’d anticipate a 2-1 win to South Korea here, too, based on the criteria we’re using.
End result: Argentina wins on goal difference; South Korea comes in second.
GROUP C
England, United States, Algeria, Slovenia
Match #1
England v. United States – Okay, here’s the thing. This is a big game for most of our readers, so I’m going to explain why I came up with what I came up with first. First we look at population: the 300 million people in the United States crush the 51 million in England, so that right there is .6 goals for the U.S. GDP per capita is roughly even, so nothing there. England absolutely destroys the U.S. when it comes to experience, getting themselves a .5 goal lead. So, technically, the stats say that the U.S. should win this game .66 to .5; obviously, however, this is impossible. So I’m going to predict a 1-1 draw (rounding up), but this game is, based on this theory, too close to call. I’m just as shocked as you are.
Algeria v. Slovenia - Slovenia’s weird, because they have no real measurable experience. Algeria should win this 2-1.
Match #2
England v. Algeria – England 3 – 0 Algeria.
Slovenia v. United States - Slovenia 0 – 3 United States.
Match #3
United States v. Algeria – United States and Algeria have the same amount of experience, so it comes down to population and GDP (which the U.S. owns them on). 2-0, USA.
Slovenia v. England - England dominates Slovenia in every single category; 4-0 to England.
End result: England wins on goal difference over the U.S., who come in second.
GROUP D
Germany, Australia, Serbia, Ghana
Match #1
Germany v. Australia – Germany wins 1-0.
Serbia v. Ghana - All things being equal, I’m seeing a 2-1 win for Ghana. This is mostly based on Ghana’s experience and population (Serbia’s GDP helps them out a bit). Also, if we’re giving all African teams a home field advantage (I didn’t with Nigeria, but whatever), Ghana could win 3-1.
Match #2
Germany v. Serbia – Germany should win like 5-0. It shouldn’t even sort of be close.
Ghana v. Australia – This one, however, is interesting. Ghana matches up great against Serbia because of experience (it’s why they’re probably not going to win a match all group by this method). Australia is as experienced as Ghana is, plus they’ve got roughly the same population and like forty five times the GDP (Ghanaian’s make about $900 a year). So this one goes to Australia 1-0.
Match #3
Australia v. Serbia – Australia wins 2-0 based on dominating Serbia in every category.
Ghana v. Germany – Unfortunately for Essien & Company, this is a 2-0 win for Germany.
End result: Germany wins the group, Australia is the runner up. I’m not confident of this one because the Slovenian stats are a bit off, and because they’re a new country (kind of a wild card).
GROUP E
Netherlands, Japan, Cameroon, Denmark
Match #1
Netherlands v. Denmark - Scoreless draw. Seriously. Why? Well, Denmark makes more money per capita than the Netherlands and has about sixty more games; there are more Dutch citizens than Danish citizens, but Denmark actually edges the Netherlands in two of the three categories.
Japan v. Cameroon – Japan wins 2-0 based on their population and GDP; experience is about even.
Match #2
Netherlands v. Japan - This should be a 1-1 draw; again, Netherlands was unfortunate to draw against other affluent countries as far as this measurement is concerned.
Cameroon v. Denmark - I’m going to call this 3-1 to Denmark. Denmark’s got most of the edge, but Cameroon has more people…and, honestly, I can’t see Cameroon not scoring against Denmark.
Match #3
Denmark v. Japan – As far as this group is concerned, whoever wins here is through. Unfortunately, this one is too close to call; Japan’s vast population advantage (25 Japanese to every 1 Dane) is outweighed a little by the Danish experience, but the stats are putting this down as a 2-1 win for Japan.
Cameroon v. Netherlands – The Clockwork Orange should take this 3-0.
End result: Japan wins the group, the Netherlands are runners up. Not sure I see it happening, but if it does you heard it here first.
GROUP F
Italy, Paraguay, New Zealand, Slovakia
Match #1
Italy v. Paraguay – Italy should roll Paraguay 3-0; they’ve got more people, more money, and more experience.
New Zealand v. Slovakia - There’s an ever-s0-slight edge to the Kiwis, believe it or not. This is again partially because Slovakia’s not been around that long (and because they make about half the money that New Zealand does); however, the rules are the rules. The 1/10 goal that New Zealand scores isn’t enough for me to bother rounding up; call this a scoreless draw.
Match #2
Italy v. New Zealand – Italy’s “B” team could probably beat New Zealand; call this 3-0 again (pre-Confederations Cup friendlies notwithstanding, of course).
Slovakia v. Paraguay - Paraguay wins 1-0 based solely on experience.
Match #3
Paraguay v. New Zealand - Paraguay wins 2-1, again on the basis of experience.
Slovakia v. Italy - Italy wins this one, too. The score’s irrelevant; I’m gonna stick with 3-0, however, for continuity’s sake.
End result: Italy wins the group in a landslide, with Paraguay coming in second.
GROUP G
Brazil, North Korea, Côte d’Ivoire, Portugal
Match #1
Brazil v. North Korea – The stats say that Brazil could win this 5-0; Brazil got a little lucky, because their one weakness (a low GDP per capita) isn’t really exposed in this group.
Côte d’Ivoire v. Portugal - This one’s really, really close…so close that I’m calling it a scoreless draw. If anyone were to score it’d be Portugal.
Match #2
Brazil v. Côte d’Ivoire - Didier Drogba said he wanted to avoid Brazil, and looking at the projections it’s clear why: Brazil should win this 3-0.
Portugal v. North Korea - Portugal wins 2-0.
Match #3
North Korea v. Côte d’Ivoire - This is actually too close to call as well; population and GDP are about even, but Côte d’Ivoire has played about three times as many games, giving them a 1-0 win.
Portugal v. Brazil – Brazil wins 3-1.
End result: Brazil wins the group; Portugal edges Côte d’Ivoire on goal difference to come in second.
GROUP H
Spain, Switzerland, Honduras, Chile
Match #1
Spain v. Switzerland – If this whole regression theory thing is true, Switzerland should win this 1-0. They’ve got double the GDP and double the experience. “Soccernomics” discusses this weird dynamic, actually, in regards to England (and, to a lesser extent, Spain): the sport is insanely important, and because the national teams don’t win often they’re labeled as underachievers. In fact, the national teams are overachiving when you look at what their resources dictate they should be able to accomplish.
Honduras v. Chile - Speaking of overachieving, Honduras is called the most overachieving side in the world. They’re a tough team to play and there’s really no reason for it; however, according to the stats, Chile should win this 2-0.
Match #2
Spain v. Honduras – Spain should take this 4-0.
Chile v. Switzerland - Again, this is a 1-0 to Switzerland. Switzerland’s got a long, long soccer history that people forget because it isn’t a top-flight league; however, they’ve got one of the highest experience totals in the entire World Cup. Only six teams have more experience than them.
Match #3
Switzerland v. Honduras – Switzerland should win this 3-0, based on experience and their other asset: money.
Chile v. Spain – Spain’s only really got the stats to win this 1-0.
End result: Switzerland wins the group, Spain comes in second.
So there you have it: according to the guys who wrote “Soccernomics” (and an admittedly amateur application of their model with a few pieces of spotty data), those who should be the group winners and runners up. Do I buy it in every situation? Not at all; I think Group E and Group H undervalue the Netherlands and Spai, for example, and I think that the African nations are woefully left out of the mix (I personally see Côte d’Ivoire advancing past Portugal, for example). However, this is still a halfway decent jumping off point utilizing a criteria that’s been set forth. We’ll check back in, I don’t know, nine months and see how this turns out.
Very interesting. This is perhaps a better way to pick than just going on gut feeling, you’re not gonna get 100%, but I’d guess that you’re going to be more accurate than you normally would be because you can’t really factor in the personal bias one seems to always have when making picks.
I think I’m gonna have to buy this book tonight.
Except that soccer has nothing to do with gdp or population
so US will not pass, slovenia should and by no mean uruguay has more experience that france
Soccer has everything to do with those things on certain levels.
-Population directly influences the number of people available in your talent pool; it’s a big reason why San Marino and the Faroe Islands are non-entities in the international field. They aren’t inherently worse at soccer; there just aren’t enough of them to make a difference.
-Similarly, GDP per capita directly impacts how much money (and, theoretically, coaching resources) can be allocated towards a specific player’s coaching or development. It also relates to the overall health of the player, his life expectancy, and how viable he is physically.
-Uruguay not only has more experience than France, they’ve won as many World Cups. From May of 1901 to April of 2001 (when my data stopped tracking), they’d played 688 matches; France, in a very similar time frame (1904-2001) played 601. Since 2001, I have no idea, but I’d imagine that Uruguay (with a considerably more rigorous qualifying run in CONMEBOL) is still on top.
Now does that make the correlation automatic? I don’t know; I’d guess that the authors would say that I’m using information meant to explain long-term trends to predict the outcomes of specific events, which is probably not the intended use. However, my point was (and is) that some of these “minnow” teams are actually positioned to overcome some of the teams that we’ve long given weight to, and that they’re only going to get better with more experience.
I will say this, however: I think the idea of experience quality needs to be introduced in some manner.
Cool stuff, I’m a bit of a regression geek although still haven’t found anyone willing to pay me to play with it.
Although I should there think should be a way to allocate experience among states emerging from a country split. Not that it would help Slovenia much.
I thought so at first, too, but after thinking about it more I changed my mind.
Say the U.S. split into four different countries (North, South, West, East); players from the North would go north, South go south, etc. So maybe three guys go East; now they need to start a team from scratch to fill in those gaps. Those guys are likely to be ill-experienced, and so you’re basically starting over…meaning the low number is really the most accurate.
Really nice post, couple questions did you accept the way fifa gave the experience to new countries (Serbia is awarded the games play by Yugoslavia, Germany only has the West germany games count in their favor and Czech Republic not Slovakia get the one Czechoslovakia)or use another way? and i haven’t read the book but does recent performance has been weighted, or it wasn’t statistically significant?, Thanks alot.