This is my argument against using Variance, as well, as it begs the question: Variance to what?
A fair coin will have a 50% chance of coming up heads today, tomorrow or a 100 years from now.
You can't say that about an ever changing Universe.
Variance to the standard you are working from, all things are levels of certainty, some accept it before others. For example, you state the universe is changing, in what manner? the laws that govern them do not, how variable is that, not very much? I don't need to recalculate the strong and weak nuclear forces in the same way I don't need to know every variable or data point in the earth's history, it's moot.
We collect date, make predictions and re test, if our predictions are 99% accurate I would agree with them, that's the alpha level. Said another way, the chances of receiving the value we did (that co2 is causing the warming) happens 1 out of 100 times.
We know the sun's output, we know it's relative mass, what it produces, it cycles, would the cycles be different in the past? perhaps, but we know the variables that make up the sun, we can predict output that is.
In this scenario we know the sun's output pretty accurately, we know the effect of wobble, orbit, ocean currents, jet stream etc.. we have fairly accurate data, I mean we can shoot laser beams from pin point to pin point km's away, drive vehicles on mars and make extraordinary calculations and predictions, to picoseconds even.
I don't need to meet every human to know we can't fly, there are laws governing this shit, it's repeatable and testable this isn't ancient greece.
I feel people who talk about this from a denial standpoint (no offense intended) haven't gotten to the higher levels of academia, real academia, chemistry, physics were people are at the forefront of actual knowledge, making computers, rockets that self land, the precision required and dedication to reach those levels (as those in climate science have done) is astounding. It's not social studies, it's hard science.