I don't think this is entirely accurate. From what I can recall, the case against Landis was from two tests. The first showed that his T/E ratio was well outside the normal ratio for humans. Note that this is different from having elevated testosterone, in fact I believe his testosterone was actually lower than normal. It was just that his epiesterone was WAY lower than normal (these are normally about equal, his ratio was 12/1). As far as I remember this test was somewhat discredited due to shoddy procedures at the lab. The one that stuck was an isotope test which showed that the testosterone in his sample had a different isotope ratio than is found in humans. From this they concluded that it was synthetic and thus upheld the ban. I don't claim to entirely agree with all of their methods or even the results of the test, but I really don't think it's fair to say that he crashed, had a surge of adrenaline and subsequently tested positive.