A key driver for the current over-dosing of MEG into flowlines is the choice to set the steady state MEG flowrate to a level required to inhibit short-lived transient cold spots. In gas condensate systems, these cold spots can be tens of degrees below the freezing point, yet once the system is running normally, it will warm above the freezing point. In these cases, design decisions can double the required MEG injection rate. Worldwide research efforts have consistently found that hydrate nucleation is in fact a stochastic process; this means that, as liquid transits cold points during a transient operation, hydrate may not even have time to start forming. This project aims to quantify hydrate nucleation probability distributions using large data sets (10,000 + data points) to deliver a fundamental probabilistic model for subsea tiebacks. The application of such a model represents a powerful design tool which will help evaluate the true requirements for MEG injection for temporary entry into the hydrate region.