This morning, I managed to read an article on the Guardian. The authors took WADA supplied data on 26 Olympic sports and concluded that cycling is the worst offender in terms of indications of substance use in investigated samples over an 8 year period. The dataset had Year, Sport, Samples, Total Findings (positive) and %Findings as parameters.
I took the same data and made some quick plots. Make what you want of it.
% Findings - If you plot some of the most popular Olympic sports, baseball/softball looks worse in terms of % findings in samples.The other thing you can note is that many of these sports are seeing reduced findings over the years, including cycling. No such assessment can be made of Football (soccer for you Americans), Weightlifting or Athletics.
% Findings in samples investigated, where 'findings' are defined as an Adverse Analytical Finding or an Atypical Finding. Both are defined in the article.
Findings vs Samples Taken : For cycling, you can see that the number of samples taken has steadily increased over the years. After 2008, the number of findings however decrease.
Similarly, here's one done for Football, Tennis, Athletics, Weightlifting and Baseball. Except for baseball/softball and weightlifting, the other sports have all seen increasing samples taken until 2008, then they have taken a dip. Why this common trend?
I didn't have time to plot the other sports out.
The bottomline is that things look quite uncertain here and I'm not ready to call cycling the worst offender for getting caught with adverse findings. But a couple of points -
1) For one, yes, Football takes a lot of samples. If these are all in-competition samples, then the picture misses out on a large number of out-of-competition samples which have the capability of catching more "surprise" findings.
2) As far as stringency of tests, nothing from this data tells us how strictly doping controls are implemented. What are thing things looked for in the samples? Which sport looks out for more 'banned substances' than other sports? Which countries are lax, which countries give a damn. Do they follow a single standard? Do we know?
3) Finally is the classic conundrum of false positives. All this data we have only describes that certain findings were found in the samples. Who's to say they weren't later disputed and then turned out to be "false positives"? What we need to know is the probability that an individual who has really cheated can be reliably tested to be positive in a laboratory test. All this testing is useless if you don't have a reliable test. Another parameter I would have liked to see in the dataset was False Positives but its not there.
So as usual, the world isn't so simple, atleast for me.