Orange 2015 Africa Cup of Nations Betting: Statistically hard to separate teams

Although one of the better teams travelling to the tournament, Algeria are unlikely to be celebrating a tournament victory

Although the market may have its favourites, this year's Orange African Cup of Nations is incredibly closely-matched, writes Jack Houghton, making short-priced Ivory Coast and Algeria teams to avoid...

Over the last few years, I've shown how - even if, like me, you know relatively little and do not follow the game - you can make a lot of money by using widely-available statistical information to better inform your football punting.

The best opportunities to do this occur in markets which are most likely to trigger irrational behaviour from other punters, distorting the accuracy of the odds available.

For example, prior to the 2013 African Cup of Nations, I argued that the market was being set askew because, of the 15 Premier League players travelling to the tournament, six would be representing Ivory Coast. Odds of 2.767/4 suggested Ivory Coast had a one-in-three chance of winning, whereas statistically-based ratings put their chance at more like one-in-six. The squad with a preponderance of Premier League talent lost in the quarter-finals.

The same distortion is not as obvious in the 2015 tournament; however, it might explain why Algeria and Ivory Coast, with four Premier League players each, are half the odds of Ghana, who only field two: Kwesi Appiah, who is currently on loan to Cambridge; and Christian Atsu, who has only played 70 minutes for Everton this year. Double the players, half the odds?

Statistically speaking, though, there is something far more eye-catching in this year's competition than Premier League representation.

Using an Elo-based rating system (I keep my own, but very good versions can be accessed online for free, such as those at World Football Elo Ratings) to assess the relative chances of teams, the most striking feature in 2015 is how closely matched the tournament is.

The eight top-ranked teams are separated by less than 100 points, and the top 14 by less than 200 points. Only the lowly rated Congo and Equatorial Guinea bring any statistical diversity, taking the spread to 400 points if included.

To demonstrate how closely-matched this makes this year's tournament, consider that, at the last World Cup, a spread of 100 points only covered the top three teams, with a spread of 200 covering only the top five. Or that, at the last African Cup of Nations, the top eight teams had 300 points between them, with 600 points required to cover all competing teams.

To further set this in context, according to the ratings, the best team playing in this year's tournament only has the same advantage over the eighth-best team that Germany had over Argentina going into the World Cup final.

The best team playing in Equatorial Guinea over the next few weeks is Ivory Coast, who are the 5.04/1 tournament favourites. The eighth-best team is Guinea, and yet they are available at 44.043/1. Although no doubt those who closely follow African football will be able to explain that disparity, it is hard to do so statistically.

Given the closeness in ability of the teams competing, I'm fully expecting results to be considered "surprises" by pundits, and will certainly be avoiding Ivory Cost 5.04/1 and Algeria 5.85/1 in the outright market, whose odds are far too short.

Teams undervalued in the outright market are Ghana (10.09/1), Senegal (15.5n/a), South Africa (21.020/1) and Guinea (44.043/1), but perhaps a more straightforward approach to profiting in this year's African Cup of Nations is simply to lay short-priced favourites in any individual match, because as long as Congo or Equatorial Guinea are not playing, the match is likely little more than a coin toss.