This isn't a surprise, nor is it evidence of individual greed or stupidity. What's wrong with betting when there is no house edge? In which case, some people ar

This isn't a surprise, nor is it evidence of individual greed or stupidity. What's wrong with betting when there is no house edge? In which case, some people are inevitably going to wrongly pick heads or tails and thus lose their $20. Had the OP actually properly described how to build a useful Monte Carlo simulation, you could find your article, get the actual rules, run a simulation and check whether 33% was in line with expectations.