In the build up to the England – Italy game on Sunday night Gianluigi Buffon came over a little bit Spiderman (something he retained to claw away Glen Johnson’s shot in the first half). He talked about how the game was likely to be close, and how the fact that Italy’s games are always close is both their gift and their curse – they’re rarely played out of games, but by the same token rarely dominate games and run away with them. I took to the web to find out whether he was suffering from sample size failure.
The England game was Italy’s 729th international game (with one abandonment). They have scored 1266 goals in those games (at an average of 1.74) and conceded a stingy 725 (at 0.99, surprisingly enough). All well and good. I will need to contextualise my findings with other teams, so I’ve decided to go with one who have been operating at a similar level to Italy, Germany (and West Germany) and one that would have been towards the forefront of Buffon’s mind at the time, England.
I’ve split the teams’ histories into 10 year blocks to make it manageable, but also to give us decent sample sizes to work with, except for England’s pre-1900 games, of which there were 67, and I’ve counted as one block.
The first graph for your delectation is an overall Goal Difference; naturally, the higher up the graph you go, the more dominant you’ve been. Italy’s line, you’ll note stays well within the boundaries of Germany’s and England’s, meaning they’re generally less dominant than Germany – not surprising when you consider that Germany have scored 10 goals or more on four occasions (16-0 v Russia – 1912, 13-0 v. Finland – 1940, 12-0 v. Cyprus – 1969, 13-0 v. San Marino – 2006); Italy’s highest tally is 9, in a 9-4 victory over France in 1920, and a 9-0 v. USA in 1948 – at Griffin Park (part of the 1948 Olympic Games).
The fact Italy’s overall GD line has never dropped below 1 suggests they’re a team who always wins more than they lose, but it was only ever above 2 once, during the 1940s, when they played 21 games – obviously, some of those would have been the time of the Grande Torino, previously discussed on From Inside, Right, so it makes sense there was a level of dominance there. Certainly, Italy’s overall GD suggests their games are not as close as England’s are.
So let’s move on and investigate both sides of the claim in detail.
Starting off with games that Italy have won (including ET) over the years, 389 in total – 53.3% – we can see how dominant the Azzurri have been in games in which they have triumphed. Of course, to record a victory, one needs to finish a complete goal ahead of the opposition, so this graph really begins at 1 (Germany won 2 games before 1920 – both by one goal, both against Switzerland – so that is why they start at exactly one). This graph is interesting in that, historically, Buffon has a point. Italy’s wins, from 1970 to 2010 were, generally, by fewer goals than both England’s and Germany’s. Since 2010, however, England’s goal difference in victories has dipped incredibly, to 1.563 (in short, if England win now, they’re only slightly more likely to win by more than 1 goal than by 1 goal – something which is certainly not going to change a lot under Roy Hodgson)
Let’s move on to the flip side of that coin, to defeats, which all three teams suffer pretty infrequently. Judging by Gianluigi’s claims, Italy’s average goal difference shouldn’t be much above 1 here, but that, along with Germany’s, and England’s, is a bit higher than that. It is, however, part of a pretty variable chain of lines which leads us to a point where Germany are more likely to lose by one than Italy, averaging defeats by only 1.667 in the period from 2010 to today. I think I can confidently say that doesn’t stack up entirely. Also ought to point out that all these numbers are negative, though it doesn’t appear so in the graph.
I have a final graph for you, and this one is probably the most worthwhile of the lot. This shows the percentage of games that finished within a one goal difference – so a percentage of games that each team has won by one, drawn, or lost by one. There’s an obvious trend upwards on the graph, which goes along with the theory that there are no longer any easy games in international football, although the current Germany team are bucking it (particularly odd considering their results at Euro 2012 – 1-0, 2-1, 2-1 then 4-2), taking their figure down to 51%. Italy, meanwhile, led the three in the 80s, but since 1990 and beyond have run pretty much concurrently with England, though the Anglians have now pulled ahead.
Buffon, then, was certainly not entirely correct. Italy’s fortunes in terms of staying in games, or not killing them off, is mirrored by England. Saying what he did on the eve of a game with the Roy Hodgson’s team was inopportune. Had he made the same point now, before Italy face Germany, he would very much have a point in direct comparison. International football is getting tighter. Italy are victim to it, Germany not so much. What does that suggest to me? Might be worth a sly bet on Germany to beat Italy by exactly 1 goal. They can’t buck the trend forever. It also means that England have become more like Italy than Italy are.
What does this all mean? It means I’m really intrigued to see how Greece would fit on the same graphs.
Ed: In fact, I put them on the % of 1 goal game graphs. That’s a very good way of seeing how football has got tighter. Greece used to be something of a whipping boy (lost 11-1 v Hungary at one point) but have become harder and harder to beat until now.