Outliers, not in any way related to the general trend and a bad argument in general.
Numbers please.
worst case, average case is more than an order of magnitude lower for cardiac related death:
2/100.000 for cardiac related death by excercise.
1278.9/100.000 normal
1879.4/100.000 obese (combined, so about 1.5 increase)
So yes, outliers. Though the numbers are difficult to compare since the timespan of the research is different and the age groups could be off. But you asked for numbers.
One is just implicit.
Some of the longest living people in the world (e.g. in the island of Icaria) drink and smoke like it's 1977 and they have always been.
https://www.nytimes.com/2012/10/28/magazine/the-island-where...
Perhaps less stress and better lifestyle (quality of life) is the key to fewer cancers and such, as opposed to cutting smoking and drinking -- even with the increased burden on the lungs and such.
BTW, exercise kills as well. There are also known cases of heavy smokers/drinkers that made it to 100.