During 2015, 35,092 people died in motor vehicle crashes on the U.S. roadways, an increase from 32,744 in 2014. The 7.2% increase is the largest percentage increase in nearly 50 years. To reduce reckless driving and the resulting accidents, law enforcement agencies deploy speed traps. However, limited resources prevent full coverage at all times, which leaves many roads uncovered. Law enforcement agencies cannot rely on deterministic coverage as it allows drivers to observe and anticipate covered areas. Therefore, randomized speed trap deployment is vital for active road security. This paper provides random and optimal speed traps deployment based on our innovative STOP framework. STOP utilizes game theory to model drivers’ and law enforcers’ behaviors. In particular, we provide distinct weights to different actions based on the accidents probability, derive the Nash Equilibrium and Stackelberg Security Equilibrium, and determine the best strategies to deploy. The optimal game solution maximizes law enforcer utility, consequently minimizing the cost paid by the society in terms of reducing vehicle accidents.