The risk assessment of options for margining purposes is determined on exchanges using sensitivity coefficients or fixed scenarios of changes in risk parameters. Such methods cannot accurately estimate risk because they do not consider the dependence of option risk on the time to exercise. This dependence should be taken into account in modeling due to the variability of sensitivity coefficients over time to maturity and the time-structure of risk factors. This paper evaluates the effect of time to maturity and time-dependent risk parameters on option risk: implied volatility, implied volatility structure, and volatility risk premium. It was proved that there is a significant trend for the increase of risk assessment as the option approaches the exercise da­te. Moreover, not only the average risk estimate increases, but also its variance. For options with a strike different from the value of the underlying asset, the trend becomes less explicit, and the accuracy of the estimate decreases with distance from the central strike. But when the strike and the value of the underlying asset are equal, the trend describes the risk dynamics almost completely. It was found that there is a dependence of option risk on the structure of implied volatility: relative volatility bias significantly reduces the level of risk at the central strike, while the distance of bias increases the level of risk. It is important to note that implied volatility, although describing the volatility of option value, does not affect the level of option risk. The volatility risk premium is a relevant factor in describing option risk, but only for the paired regression cases.