Quite a lot of what has been written in this thread is absolutely true.

One thing to remember when installing bolts is that the aim is to stretch the bolt (usually an elastic, recoverable stretch, but sometimes a stretch where the bolt does actually yield, and will not return to its original length upon removal). Applying a torque is just a means to this end.

It is only by making the bolt longer that the bolt can actually clamp the joint together, and in many cases this clamp load is critical, because you want the joint to seal, or to resist slippage.

Pure torque tightening is at best a crude method of tensioning bolts. Even with top quality torque wrenches, new cut threads, top quality bolts and skilled fitters, you can expect a +/- 25% scatter on the actual clamp load you get.

Chief among the reasons for the scatter is variations in the amount of friction in the threads, and beneath the head of the bolt.

This uncertainty can be largely by-passed by using a torque + angle technique. The torque snugs down the joint, and the angle then applys a repeatable amount of stretch to the bolt, which in turn means that you get a more repeatable total tension in the bolt - which, remember is the aim of the excercise.

As the torque + angle will have been calculated using data on the geometry and stiffness of the bolts, the geometry and stiffness of the joint, and inherently reduces scatter, there's absolutely no way to go back and work out an equavalent torque value to use.

Even if you do three 90 degree turns with +/- 10 degrees tolerance on each, you will be applying a more consistent preload than if the joint were being tightened with a torque wrench alone (11% error rather than anywhere within the 25% error band I mentioned above)

If you use a protractor and do it properly, you are tightening the bolts with really quite a good accuracy.

In terms of deciding whether to re-use bolts or not, it depends what the FSM says about the particular bolt. Personally, I would prefer not to rely upon measuring diameters of bolts, because it needs an accurate measurement, as the reduction in diameter is a Poisson's ratio type of contraction and is much less than the corresponding change in length of the bolt - i.e., you need far lower measurement precision to determine if a bolt is OK if you measure length change.

An even better method of bolt installation which is beyond home use is where the bolt angle and torque are measured simultaneously during installation. By comparing changes in torque level as the angle turned increases, you can determine the point where the bolt begins to yield, and stop the machine there. This method both uses the bolt to its utmost, and has an extremely low scatter in terms of bolt pre-load.