Sports fans might be familiar with the favourite-longshot bias in sports betting. The odds for the favourite or underdog team are too high or low to represent the actual probability of winning. Joseph Buchdahl (2003) noticed that in the English soccer leagues the longshot odds (the one for the underdog that is less likely to win) are waaaaay too low, even after removing the bookies' margin. So even with higher odds you will never win enough money in the long run to cover the costs. Since I don't care about soccer, I did the same analysis for e-sports.
#tba: more explanations about the math following