> Again, all I am arguing is that if you use a sequential test, you should analyze it accordingly.<p>I hate to be picky and negative but this doesn't seem like a justification for the use of the word 'harmful' in the title. The phrase 'considered harmful' has been used quite a bit lately and I get that it attracts attention but a title like 'Bayesian A/B Testing Considered Harmful' sounds like a group of authoritative individuals has collectively decided that it is indeed, harmful (and I was expecting a list of the reasons why). If it's just one persons opinion, wouldn't it be more accurate to say 'Why I consider Bayesian A/B Testing harmful'.