If a population is small and you are only looking at a few generations, of course random events can have bigger effects than selective pressure. However the odd

If a population is small and you are only looking at a few generations, of course random events can have bigger effects than selective pressure. However the odds of two options being perfectly equal in terms of selective pressure is low - and even a very tiny advantage matters for large numbers over large time spans. It is an unfortunate reality of genetics and many other fields that we have only a few decades of good data and a scarce few older data points from which we must extrapolate complex trends that occur over many thousands and even millions of years. Certainly in the case of the blue tigers, where there is a clear mechanism by which the variation would affect fitness (the red photoreceptors that make tigers stand out to us as a bright orange are quite rare among mammals, to most prey animals tigers appear as grey or a dull yellow which blends in quite well with their surroundings, while blue photoreceptors are much more common and blue tigers would thus stick out clearly to many prey animals), and where the data is so sparse (we don't even have a preserved specimen) it seems extremely inappropriate to conclude the absence of evidence for selection pressure is in fact evidence of absence of selective pressure.