Accurately predicting the genotype on primary hybrids is straight forward. Phenotype is another thing altogether as we are all familiar with some species exhibiting more dominance than another. When dealing with previously unknown primary hybrids there is always room for a surprise as you never know for certain how the genes will express themselves in the F1 generation. It is fairly safe to say though that in F1 primary hybrids the range of phenotype differences in the HYBRID population is minimal. Only when you get to F2 and beyond or start mixing different hybrids do you see a much broader range of phenotypes showing up.
As for the sex determination who knows... Perhaps it is multiple genes? If for example it were 2 genes A and B and in order to be female for example 2 B genes needed to be present. All plants with AA or AB were male. Then the ratio would naturally be about 75% male. On the other hand who can say for certain that the ratio really isn't 50% each? Maybe environmental factors cause more female plants to die in the wild so the ratio is skewed? I could probably come up with another dozen hypothesies too