If you read carefully, the more jobs situation was 40 years later. This means that at least for some time, unemployment was higher as a result of the automation

If you read carefully, the more jobs situation was 40 years later. This means that at least for some time, unemployment was higher as a result of the automation. The thing to remember is that you can't mothball workers, you cant say "hey dont eat or live untill the job creation curve from this new tech catches back up, you'll be in demand again". So while the net effect of automation is more jobs, it still doesn't make the immediate (within a workers lifetime) impact necessarily better. Another thing the article mentions is a 4400% increase in the number of jobs. This is not population growth, (it is only 3 generations assuming child labor), so where did those people come from? Were they more skilled workers earning good wages replaced by automation? Were they unskilled workers whose previous jobs had been automated away (ditch diggers and the like)? In one case you have a net loss in liquidity in the hands of people, this is the net equivelant of lost jobs. In the other case you have a potential gain in liquidity. Once again, not making a case for or agains the automation, just discussing if the employment is equivelant. To use your analogy its not like me asking if we should continue with the rugby at all. It's more like me saying "Hey guys, I know that the mechanism is different, but can we also look at the fact that 100 more teams in the cup might dilute the quality of play?"