When is a multiple bet better than a single ?

Main Article Content

Nicos Zafiris


The paper addresses an apparent paradox observed in betting on football scores, drawing on 20-year data from the English 2nd tier division (Championship). While accumulator bets have a lower Net Expected Value than single bets, ‘cross double’ bets on the scores, placed over successive playing rounds, produce distinctly better results and indeed a positive return overall. It is argued that this effect rests on the essential stability of the score frequencies across playing seasons and on the bookmakers’ failure, in setting the odds, to allow for occasional and temporary deviations from long run average frequencies. A betting strategy based on overdue scores occurring with compensating frequencies, and possibly clustered together, can then produce positive returns. Neglect of overdue scores can be expressed formally as a bias augmenting the probability of these and turning the odds in the bettor’s favour. It is shown that, while normally the bettor’s disadvantage is compounded in multiple bets, a compounded advantage results once the odds become better than fair. The paper also discusses certain quasi binomial characteristics of the betting involved and explores possible ways of hedging such bets ‘in running’. 

Article Details



Bernstein Peter (1996) Against the Gods, Wiley

Dixon Mark J and Coles Stuart G (1997) “Modelling Association Football Scores and Inefficiencies in the Football Betting Market” Applied Statistics, 46, No 2, pp 265-280.

Forrest D and Simmons R (2000) “Forecasting Sport: The Behaviour and Performance of Football Tipsters” International Journal of Forecasting, 16, pp 317-331

Jackson David A (1993) “Independent Trials are a Model for Disaster” Applied Statistics, 42, No 1, pp 211-220

Moore Peter G (1994) “A national lottery in the UK” Journal of Applied Statistics, Vol. 21 Issue 6, pp 607-23

Mukhtar M Ali (1998) “Probability models on horse-race outcomes” Journal of Applied Statistics, Vol. 25, No. 2, 221± 229

Ross G S S and Preece DA (1985) “The Negative Binomial Distribution” The Statistician, 34, pp 323-336

Skinner G K and Freedman G H (2009) “Soccer matches as experiments: how often does the 'best' team win?” Journal of Applied Statistics, Vol. 36 Issue 10, pp 1087-1095.

Stefani Raymond T (1983) “Observed Betting Tendencies and Suggested Betting Strategies for European Football Pools” The Statistician, 32, pp 319-329

Zafiris N (2002) ‘The Distribution of Football Scores: Towards a Betting Strategy’ (W R Bayldon). Published Proceedings of Annual Conference of British Academy of Management, London, 2002, CD, English Partnerships