It's more that people believe your statistics to be naïvely over-optimistic based on a few things.
1) counting each Tour individually, so Armstrong being counted amounts to 7 busts.
2) the large number of other riders known to have slipped the net.
3) that a few of the riders mentioned are only tipped off because they confessed, and your 10-year timeframe is too restrictive because Riis and Armstrong were outside of it, notwithstanding that you treat BC's history in track cycling as the same as a 10-year timeframe in road cycling, with much more exposure, much more money, a much deeper field and much more doping cases.
4) that only Landis' positive was intentional, as the UCI attempted to suppress Contador's and successfully suppressed Armstrong's. Pereiro got a rushed TUE, which makes him no different to Froome.
5) in the eyes of the general public Indurain is clean despite a borderline confession. Ditto Cunego if we include other races. Including marginal cases like Indurain affects the outcome of the data quite considerably, especially when you count each Tour individually as you have done.
6) the wilfully restricted dataset (Tour de France winners only) reduces the value of the conclusions drawn from them. Add in other podium riders at the Tour, other GTs and races and the number of suspected dopers who have not faced any sanction increases greatly. I mean, Rominger? Jaskuła? Escartín? Andy Schleck?
7) Pierre Bordry is not working for AFLD anymore.
8) We aren't dealing with Lance Armstrong, and the past has given us lessons to learn from. The likelihood of the kind of massive scale team-wide doping of the 90s and early 00s is small; smaller, more tightly controlled pockets are seemingly the order of the day, laut CIRC. Easier to circle those wagons.
9) Sky are tied to a media empire that will enable them to control the story in the public eye the majority of the time and will make it easier to suppress any bad news.