"That does not change the fact that even well made and designed springs change with repeated flexing."
You are bringing in fatigue failure here - which I was specifically excluding. *Very* different principles apply...
We are talking about STATIC failure (or permanent deflection) here. Put in the shooting vernacular, "the springs get "tired" if left compressed for some extended time period". We all probably know people who think they need to change their magazine springs every year or six months to prevent the springs taking a permanent set ("getting tired") or that you should not leave a gun cocked to "preserve" the mainspring.
That particular shooting urban-legend is what I'm talking about.
Here's how springs are designed (as normally done):
STATIC Failure:
1) Choose material - can be relatively high-carbon steel or an alloy. Not counting corrosion, the ingredients of the material will not change with time, so that is not the explanation for the "getting tired" effect, if it exists.
2) Choose physical dimensions: This can be L x W X H for leaf springs or wire diameter and coil diameter for coil springs. In either case, not counting for wear (which doesn't happen in the static case), there is no change in geometry with time, so that is not the explanation for the "getting tired" effect, if it exists.
3) Stress Evaluation and Spring Constant: This step is done in conjunction with Step 2, in that the amount of force required for the spring's operation and the equivalent deflection are related to the physical dimensions. Once the proper decisions (and calculations) are made, you can expect the spring to exert sufficient force to do the required task (feeding-rounds, for example), without exceeding the allowable stress in the spring material.
If the internal stress exceeds the yield strength, the spring will take a permanent set, and if it doesn't, it won't. If the yield strength is exceeded, due to poor design, it will happen the *first time* the spring is deflected and the time duration of the applied force has nothing to do with it.
Leaving rounds in a magazine or leaving a hammer cocked will not have an adverse effect as long as the spring is designed properly. As stated, even if it is NOT designed properly it still doesn't affect the lack of time- dependency involved in permanent deflection (or failure) - it would fail at the first application of force, not experience a failure at some distant time in the future.
Difference in STATIC failure (deflection-stresses only) and FATIGUE failure (repeated stress) and related analysis:
The STATIC strength and FATIGUE strength are related in some ways, but are definitely not equivalent either in theory or practice.
You could write a book on fatigue failure analysis, and many people have. Among other interesting aspects is that you can sometimes remove material from an object, thus raising the average stress (STATIC case), while at the same time significantly improving the fatigue life of the part (number of successful cycles).
Fatigue design/analysis also has an aspect of art, as opposed to science/mathematics, as OKFC05 implied ("...not a complete description of the physical process").
As my favorite machine-design text states, "It is very important for the designer to develop a "feel" for stress concentration so that he will know intuitively when it exists and what to do about it."
This goes back to my original contention that the quality of the design-engineer is at least as important as the specific type of spring (and related systems).
