Mostly because you can run experiments on simulation to see "how much does it change if I do..." The difference between a 5cm thick and a 3cm thick was rather

Mostly because you can run experiments on simulation to see "how much does it change if I do..." The difference between a 5cm thick and a 3cm thick was rather surprising to me that it changed it that much for what a 250°F cooking does for some duration. It also provides a "cooking for 70 minutes says it should be X in simulation while the oven is providing Y" that something is off between the simulation and (unmeasured) reality and one of the assumptions of either the simulation or reality is off.