I'm not aware of any move that makes it impossible for any one player to win Settlers of Catan, while still extending the game to 2.5 hours. There are at least four basic strategies that work to win, and it's nearly impossible to block all of them to the other players while still pursuing one yourself.
I have seen people win without ever building a third settlement.
i think the point of the parent is that the settlement phase at the start of the game can be pretty decisive. E.g. some people always choose crossroads with numbers that have low probably (like 11 or 3) or with only 2 aligned fields (instead of 3). Of course, you might still win the game but you're problaby too annoyed about your decisions in the first 5 minutes of the game than playing and having fun (and having a chance at winning).
Especially in games where you collect ressources to gain victory points, making the wrong choices at the start can be very devastating (because of the runaway effect).
If you choose a weak starting position, it's not going to take 2.5 hours for your opponents to win in Settlers of Catan. But if you do have an obviously weaker board position, no one will put the robber on your 2, 3s, 11s, or 12, except in rare situations. That's part of the built-in handicapping for those with obviously stronger board positions. Those 5/36 rolls don't actually produce 5/36 of the time, because most of the time the robber is blocking one of them.
If you choose a weak initial position, the expansion-first player gains an advantage over the city-first player, especially in a 3-player game. And they will both know that your only option is a card-first stealth victory. The latter is usually only viable in a very balanced 4-player game.
If you don't understand probability for 2d6 dice rolls, yeah, you're going to have a bad time in a lot of games, some of them lasting 2.5 hours or more. And if you haven't bothered learning that, why would you learn from the stronger opponents that you play?
I mean, why continue at that point? That just sounds silly.