It depends solely on application and the science of annealing and/or tempering. Example would be you can take all the brittleness out of high carbon steel by annealing it. Application and “fatigue” are the main factors in choosing the proper alloy for the job soft or hard.
There really is an exact science of these combinations depending on application. Take Rails for example. If Rails were too soft they would only last a week of heavy Train traffic. They would just mush out. Yet at the same time how they react to temperature expansion and contraction can even be a huge factor.
I watched the local line put 50 miles of new rail in one time only to pull it all back out and replace it again. The metallurgy was not correct so when summer came it expanded far too much and was buckling the rails causing the spikes to pull and derailments.
You could literally look down what was supposed to be a straight shot and it was wavy by four or five feet every few hundred yards. Looked like a snake...
The “Temperature” of a continuous railroad track is an important factor in whether or not the track will buckle in compression, or even fracture in tension. It is defined as the track temperature at which there is zero thermal (tensile or compressive) stress in the rails due to ambient temperatures. It is strongly influenced by the ambient temperature when it is laid. The stresses change with the number and degrees of curvature in the track as well as ambient temperature in service. Here’s a good explanation: https://www.railtemperature.com/Research/Ensco_Temperature_prediction.pdf