Project trade offs means that in case one (or more than one) constraint is no longer aligned with the original project plan, then the project manager will rectify the situation by playing with the remaining constraints.
The decision to make a trade off is really based on the importance of each and every constraint. For example, some projects have a fixed deadline that is impossible to change, other projects may have a tight budget that they can never exceed, while the remaining projects have a scope that should never be altered. The importance of constraints depend on the type of the organization undertaking the project, the client, and the type of project.
Now in every project, the importance of scope, cost, and time, is a number from 0 to 1, where the total of the importance of all the constraints is equal to 1. Let me give you an example so you can better understand, let's say you are constructing a tower for a very wealthy customer who places quality above anything else, and he wants the tower to be finished on time. The importance of scope I(S) will be 0.6, the importance of time I(T) will be 0.3, the importance of cost I(C) will be 0.1 (Total importance of constraints is 0.6+03+0.1 = 1).
Now in the above example, let's say you discover that the marble used on the floor and in the bathrooms was below quality. The tradeoff that you make, since cost is of the lowest important, is to get better marble, and get more labor to replace the old marble with the new marble.
Now let's look at another situation in the same example, you feel that your project is going to be late. Again, since cost is of little importance, then you add more labor to finish on time.
There are some situations where you have to make more than one tradeoff in order to deliver the project that the client wants, according to his constraints. Again, in the example above, let's say that the government blocked work on your construction site because you didn't have the necessary permit (of course, no one should run into this problem, but it's just an example). To solve the problem, you pay the penalty and you pay for the necessary permit, and you add a lot of labor, but still you can't meet the schedule so you have to extend the delivery date of the tower. In this case, there are two tradeoffs to deliver the project according to the scope: time and cost.
Finally, in some cases, you have to play with all 3 constraints, and that's the last resort.
In any project, you start making tradeoffs from the least important constraint, and you start playing with the most important constraint only when your project is in a desperate situation.
Personally, because I have been in the software industry for nearly all my professional life, I have discovered that the scope always suffers, regardless if its importance as a constraint. In fact, in nearly every software project, all 3 constraints are touched and altered during the course of the project.