Hacker News Reader

If I were to guess, the results lag behind the prevalence. It takes many years for someone to get cancer. If you were to push back the graph by 20 years it may correlate strongly.

Edit: Also, the U.S. banned asbestos in 1989. That's the turning point of the graph.

The correlation is one of the strongest there is, and indeed, causal -- there's a almost perfect lag time of ~20 years [1].

[1] https://en.wikipedia.org/wiki/Lung_cancer#/media/File:Cancer...

The graph stops at 1970.

Are you suggesting that some post-Nixonian mechanism is in play here.

People die from lung cancer when they are a bit older, so it's shifted to the right relative to the prevalence of smoking.

smoking rates was dropping quickly since 1945, during the period lung cancer rates were going up significantly.

death is a bad stat to look at; incidents would be more useful