This paper presents methodologies to predict thermodynamic conditions that initiate flash boiling by spontaneous nucleation of liquids consisting of hundreds of miscible liquids and their lower order surrogate mixtures. The methods are illustrated with a kerosene-based fuel and a seven-component surrogate for it. The predictions are compared to measurements of nucleation temperatures obtained from a pulse-heating technique that rapidly heats a microscale platinum film immersed in a pool of the test fluid. Nucleation temperatures are predicted using a generalized corresponding states principle (GCSP), and a modification of classical nucleation theory that considers the mixture as a pseudo single component fluid (PSCF). The intent is to offer a simple means to predict the initiation of flash boiling that can have important consequences for fuel efficiency in combustion engines. We show that contact angle has a strong effect such that predicted and measured spontaneous nucleation temperatures agree for a given heating rate only if contact angle is accounted for in the theory. For low heating rates (less than 2 x 10(7) K/s), predicted nucleation temperatures, assuming a spherical bubble, are 13% higher than measurements. The gap is closed to about 6% as the heating rate reaches 3 x 10(8) K/s (the highest that could be reached in the experiments) and to less than 1% when the measured nucleation temperature is extrapolated to an asymptotic (zero contact angle) limit. The PSCF method (using mixture properties as mole fraction averages of mixture component properties evaluated at the same reduced temperatures as the mixtures) and the GSCP predict virtually identical nucleation temperatures, while the GCSP method does not require any mixture property values.