If making a lifeboat is the real goal, then let's colonize the moon. It's much closer, and about equally inhospitable.
The best option is not screwing up our planet. If asteroid strikes is the concern, then funding a planetwide network of residents living in underground bunkers with supplies to last decades is probably cheaper and more effective than going to Mars.
> That's like saying: "don't make backups, just don't screw up."
That does not even belong in the same category. Data backups in computer systems are cheap, and people/organizations not doing them even cheaper!
On the other hand, doctors don't make backups of the gravely ill[1] before a dangerous surgery. That is beyond our current capabilities, so "do not screw up" is as good as it gets.
Yet another example, civil engineers do not make backups of skyscrapers[2] before doing maintenance work, even major maintenance work. While technically feasible, the economic cost would be prohibitive.
My gut feeling is that a "backup planet" would fall somewhere in between cases [1] and [2].
The top mosty likely existential threats, including "what the universe could do to Earth" would leave it more habitable than Mars.
If "lifeboat for species" is the goal, then going to Mars is a solution in that direction but not a particularly good one - building an underground colony in Antarctica or a self-sustainable isolated underwater colony would achieve the goal better, be reachable quicker, and at a lower cost. However, 'lifeboat for species' right now is not an explicit end goal for anyone who would be capable to fund that.
The best option is not screwing up our planet. If asteroid strikes is the concern, then funding a planetwide network of residents living in underground bunkers with supplies to last decades is probably cheaper and more effective than going to Mars.