Transient faults in VLSI circuits could lead to disastrous consequences. With technology scaling, circuits are becoming increasingly vulnerable to transient faults. This paper presents an accurate and efficient method to estimate fault-sensitivity of VLSI circuits. Using a binary counter and an RC5 encryption implementation as examples, this paper shows that by performing a limited amount of random simulations, fault sensitivity can be estimated accurately at a reasonably low computational cost. This method is then used to show that the combination of two circuit level techniques can make circuits more fault-tolerant than using these techniques individually.