Characterizing and modelling persistence in the number of lottery winners

Lottery is the most famous branch among all the games of chance. By analysing data from Mega-Sena, the major lottery in Brazil, we investigated the presence of persistent behaviour in the time series of the number of winners. We found that the demand for tickets grew collectively as an exponential driven by the size of the accumulated jackpot. Finally, we identified that a stochastic model grounded on the rolling-over feature of lotteries can generate correlations qualitatively similar to those observed empirically. The model is consistent with the idea that the growth in the number of bets, motivated by the size of the expected jackpot, is a mechanism generator of correlations in an apparently random scenario.