There are a lot of icing charts out there, but here's one, and it's pretty clear that carb ice can form in some situations which are somewhat unexpected:

I've gotten carb ice in Anchorage on a clear, cold day, such that the 150's O-200 Continental quit on short final into Merrill Field. I don't think I've forgotten the carb heat on a Cessna since then. Quite a learning experience for a barely post-solo student!
I've gotten carb ice at Laramie on a cool but dry Fall day in a PA28 Archer with an O-360 Lycoming, such that the amount of available power was significantly less than the throttle setting--learned not to believe Piper's POH's advice to only use carb heat if desired and made it every time the throttle is reduced for landing.
I've gotten carb ice over Nebraska at cruise in a Skylane with a stock O-470, dark night, high clouds, seemingly no significant moisture, solved with just a bit of carb heat.
I've gotten carb ice over northeast Iowa on an 80F cloudless day in early August with my O-360 Lycoming, solved with full carb heat and full throttle.
But with a carb temp gauge in the TR182 I partnered in the late 70s (TO-540 Lycoming), I never got carb ice because I pulled the carb heat on time. Since installing a carb temp gauge on my O-360 Lycoming (part of my Insight G1), I have not gotten carb ice for the same reason.
The key to avoiding carb ice is to apply carb heat soon enough, while the engine is producing enough heat to accomplish the job. If the air coming into the carb throat is a few degrees above freezing, that's the time to pull the carb heat. If it's a lot higher than freezing, or below freezing, there's less likelihood of carb ice. But there aren't any guarantees, so if there's any indication such as rough running or much less power for the throttle setting, regardless of the carb temp gauge indication, pull the carb heat and go to full throttle to melt it out. It's bad stuff!
Cary