[htag]

I'm glad you found something that worked with you. This type of experiment (consume less self administered psychoactives and access emotional changes) is something I would encourage everyone to do. However, it's a common mistake to take an experiment with a small sample size and assume it's results will generalize well over the population. For that we can look to the scientific literature. Inside we find it's well supported that increasing dose of alcohol is associated with increased depression. We do not find the same amount of support for caffeine/coffee, where it's more plausible there is a negative correlation between caffeine and depression.

Your experiment is great because now you know what works best for you. The way you talk about it seems to indicate that you use the experiment as a basis for knowing what is best for the population at large. A scientific study is a much stronger piece of evidence for shaping our views on how alcohol/caffeine effect the population.