Exactly. And all the article itself is example of marvelous statistics: it's not really useful to know how many people died while taking a selfie or how many se

Exactly. And all the article itself is example of marvelous statistics: it's not really useful to know how many people died while taking a selfie or how many selfie-deaths were caused by running in front of a train. Implication is that it's about "selfie-taking trend is dangerous", but it's not. Much more interesting would be to compare how many death, caused by extremal activities were accompanied by taking a selfie — it's quite possible, that people die of running in front of a train as much as they did before, they just take selfies while doing it now.