I can't believe people are even debating going for two. Obviously, you go for two. When you have a chance to END the game, put it out of reach in a single play, you take it. Sure, if you kick the extra point, it slightly increases your odds of winning over not scoring any points at all. Certainly it makes things more difficult for Illinois if they score the touchdown.
But remember, the challenge of converting a two-point conversion (as opposed to an extra point) only exists for Illinois if they come back to score the touchdown in the first place. The naysayers are spouting these conversion percentages as though it were a given that Illinois would have scored. But it isn't - the odds of Illinois going for two and tying/winning the game are strictly contingent upon them marching down the field and scoring a touchdown first, which even against a tired defense is still a bigger challenge than converting from the two yard line. This is why kicking the extra point would only have increased the Gophers' odds of victory by a small amount. If they had failed to convert the two-pointer, a few percentage points is all it would have cost them. Compared to the reward of all but guaranteeing a victory, the risk is minimal.
This is a problem with a lot of conservative sports decision making. People overstate the risk and understate the reward. They get too caught up on what would happen if they fail. It's why teams punt more often than the math says they should. And it's why so many people on this board are insisting that kicking the extra point was the percentage play, even though it wasn't.
(The best percentage play would have been if Brooks had taken a knee, but that didn't happen. We're talking about the decision in the moment here).