The present research examined the influence of improved knowledge of odds and mathematical expectation on the gambling behavior of university students. A group of 198 Introductory Statistics students received instruction on probability theory using examples from gambling. One comparison group of 134 students received generic instruction on probability and a second group of 138 non-Statistics students received no mathematical instruction. Six months after the intervention, students receiving the intervention demonstrated superior ability to calculate gambling odds as well as resistance to gambling fallacies. Unexpectedly, this improved knowledge and skill was not associated with any decreases in actual gambling behavior. The implication of this research is that enhanced mathematical knowledge on its own may be insufficient to change gambling behavior.