If you're playing 1D10, then a good rule of thumb is that 2:1+1 will kill your enemies more than you'll lose guys.
So if you're willing to accept a slow, grinding advance, 2-3:1 is "good enough". What you need higher odds for, is a rapid advance, the sort where you break through and stay face up doing so.
But especially in France, you probably can't get those kinds of odds without burning an o-Chit, and you really don't need them. It's only 3 hexes from the Belgian border to Paris, so 2-3 combat wins can get you to the gates, and you should have enough reserves in a normal game to just move up other guys when the attackers flip.
So yes, I will often accept lower odds attacks, but as with everything else, what counts as a "good" attack is dependent on your strategic situation; how many losses you can afford to take, and how quickly you need to move.
The 4:1 thing, at least when I was using it as a rule of thumb, is mostly for the opening phases of the Blitz into the Soviet Union. There, it's easier to surround and blast units with air power, and you need to advance much further to get to anything vital, and you're in a hurry turn wise; wait too long, and the Soviets start doing things you don't like such as railing away factories, building all their mils, and otherwise making your life difficult. Once you're committed, it's really to the factory line or bust, and that's a long way away. So you need to stay face up, and that means fewer high odds attacks.
But if you're say, charging forward in France, or you're the Americans attacking in Italy, or the Japanese trying to break the Chinese mountain lines (Note, this is with the old maps, I'm still green with the newer ones which make for more maneuvering war), you can often do quite well with lower odds attacks.