The difficulty of succeeding as an international manager

In many ways, managing a national side seems like an ideal job. International managers have very few day-to-day responsibilities besides, well, watching a lot of soccer. Of course these managers every so often have to choose a squad for a friendly or a qualifying match for a major tournament (e.g., the World Cup, Euros) and then coach the squad in these matches. But on the whole, international managers have far fewer daily responsibilities than club managers. Sounds like a pretty cushy job, right?

In fact, it's far from ideal. The much-maligned Fabio Capello, who recently stepped down as England manager, in fact had the best record of any England manager in recent history.1 Despite that England won two-thirds of their games under Capello, FA councillor Peter Coates recently declared that Capello should not have stayed on as manager after the World Cup: "We had a terrible World Cup, he was responsible for that and he should have gone then." Coates is hardly in the minority; many other English fans and journalists have expressed a similar sentiment. International managers are not judged on their overall record. Instead, they are judged based on their team's performance in knockout style tournaments that don't occur all that often.2 Most of the better national teams (that regularly qualify for the big tournaments) compete in a regional tournament every four years and the World Cup every four years. So, these teams compete in a major tournament every other year. Their performance in these tournaments is essentially the sole criterion used to evaluate the success of an international manager.

Using simple probability theory, it quickly becomes clear why it's so difficult to succeed in international management, especially under circumstances in which supporters expect their team to win trophies. Capello had the best record of an England manager in recent history winning 66.7 percent of games, losing 19.0 percent of games, and drawing 14.3 percent of games. This is certainly a remarkable record in international football. Let's assume, for the sake of simplicity, that Capello's record with England represents England's true (constant) probability of winning, losing, and drawing games in a major tournament. Such an assumption is, in fact, almost certainly a huge overstatement since these tournaments are by their nature "the finals." The level of competition is, on average, much higher than general competition since teams in these tournaments had to prevail in an often arduous qualification process in order to participate in the tournament.

Since games in the knockout phase of the tournament cannot end in a draw, I further assume that England wins 67.7 percent of the games that are tied at the end of regulation. This assumption equates to a 79.4 percent chance of winning games in the knockout phase of the tournament, since these games cannot end in a draw.3 In this informative albeit terribly simplistic model, England have an 87 percent chance of advancing past the group stage,4 a 69 percent chance of advancing past the first knockout round (the round of 16 in the World Cup and the quarterfinals of Euros), a 55 percent chance of advancing past the second knockout round (the quarterfinals of the World Cup or the semifinals of Euros), a 38 percent chance of advancing past the third knockout round (the semifinals of the Cup or winning Euros), and a 29 percent chance of winning the fourth knockout round or the final match of the World Cup (see the figure below entitled, "The Capello Assumption..."). In other words, even imposing incredibly unrealistic assumptions--e.g., that England have nearly an 80 percent chance of winning games in the knockout rounds of the World Cup or Euros against the likes of Spain, Brazil (not Euros obviously), Germany, France, Portugal, Argentina (again, not Euros obviously), etc.--England would have less than a one-in-three chance of winning the World Cup (and less than a two-in-five chance of winning Euros). Yet, England fans expect trophies! That is not to say that England haven't underperformed, as they most likely have, but an expectation that a manager should bring home a trophy from a knockout tournament is unrealistic.

(Click on figures to enlarge)

Now, let's analyzes a team's chances of advancing through a 32-team tournament with, perhaps, more reasonable probabilities of winning (no need to pick on England any further). The figures above display these probabilities for a team with a 60/20/20 percent chance of winning, losing, and drawing, a 50/25/25 percent chance of winning, losing, and drawing, and a 40/30/30 percent chance of winning, losing, and drawing.5 As these figures indicate, the likelihood that the 60/20/20 team, which has a 72 percent chance of winning games in the knockout round, wins the tournament is 21 percent. In other words, we would expect this very strong team to win a World Cup about every 20 years.6 The 50/25/25 team, which is also a quite strong team with a 62.5 percent chance of winning games in the knockout round, has a 10.5 percent chance of winning the tournament (which equates to winning a World Cup about every 40 years). And finally, the 40/30/30 team, which is still a very decent team with a 52.5 percent chance of winning games in the knockout phase, has less than a 5 percent chance of winning the tournament (which equates to winning a World Cup about every 96 years).


1. To be clear, Capello resigned after the English FA intervened and took away the captain's armband from alleged racial-abuser John Terry. Even so, some in the Italian media have asserted that Capello was "escaping" England.
2. The World Cup and most of the regional tournaments (e.g., the UEFA European Championship) are structured as a group stage followed by a knockout stage. Four teams are assigned to each group; all of the teams within a group play each other; and the teams with most points (3 points per win and 1 point per draw) advance to the single-elimination, knockout phase of the tournament.
3. I arrive at 79.4 percent by assuming that England have a 66.7 percent chance of winning a game that is a draw at the end of regulation: 0.667+0.190*0.667=0.794 or 79.4 percent.
4. I assume that teams with 5 or more points always advance out of the group stage, teams with 4 points have a 50 percent chance of advancing out of the group stage, and teams with 3 or fewer point points never advance past the group stage. In the past 4 World Cups, 11 out of 23 teams that ended the group stage with 4 points advanced to the knockout phase.
5. I again use the same methods as in note 2 to assume that the 60/20/20 team has a 72 percent chance of winning games in the knockout rounds (0.60+0.20*0.60=0.720), the 50/25/25 team has a 62.5 percent chance of winning games in the knockout rounds (0.50+0.25*0.50=0.625), and the 40/30/30 team has a 52.5 percent chance of winning games in the knockout rounds (0.40+0.30*0.40=0.525).
6. For the estimates of how often a team is expected to win the World Cup, I round to the nearest multiple of 4, since the World Cup is held every 4 years.