Sometimes that crush and breakdown is what you want. In all cases, the rifling lands make that crush inevitable and in the spot you least want strength loss a foot or two up the barrel. We may want the shoulder of the bullet to crush and glide into the bore without very high resistance because the higher the resistance, the higher the pressure on the base needed to overcome it and the faster the rate of powder burn/pressure rise occurs, which in turn puts a lot more smushing power on the bullet base and can make it fully rivet to the confines of the chamber neck inside the case. Base riveting is something which almost never happens in a controlled, consistent fashion while maintaining consistency of form with what the front end of the bullet is doing. It's like balancing a stack of plates on a pogo stick and then launching the whole mess up into the air: The forces will never be in line enough for the stick and stack to come back down again in perfect alignment.
Sometimes we want to do our damnedest to prevent shoulder crush, like in a revolver, but that's a whole 'nuther thing. A balancing dose of tin will make a strong intermetallic lattice which is malleable but not thixotropic; in other words doesn't go to putty once crushed but merely squishes as much as it needs to without becoming weaker in the process.
Trying to fit a bullet to a .308 Winchester throat compared to a .30-'06 throat (or the opposite) is a nightmare. If you a tough, close-fitting bullet into an '06, it takes a lot of extra pressure on the base to force the bullet through and does murder to the base geometry. The harder the bullet, the worse the outcome. There's a balance where a relatively malleable and crushable bullet will deform plenty while engraving but in places where it doesn't destroy the bullet's balance and form. The challenge with the '06 is the neck clearance is large, the throat entrance is large, and the bore is comparatively small. The bullet needs to be fitted so it doesn't rattle around like a BB in a boxcar and hit the throat all crooked, but it also has to get squeezed a lot to accomplish this. The neck is long, bullets are typically heavier than .308, and there's a LOT of bearing length that needs to be fat and needs to get drawn through the throat funnel. Much to go wrong. The .308 has a parallel freebore and two smaller transition tapers, so it is possible to make a tougher bullet that only has to go from .310 to .308/300 instead of .312 to .308/300 and the neck is so short that the typical bearing length is short and the throat itself can be used to guide the bullet from the middle and front instead of relying on support from both ends, effectively eliminating the need to fill the case neck to the chamber walls for support at the back. This eliminates quite a few problems at high velocity, particularly the issue of a crushed/weak alloy having to contend with high twist forces and abrasion further on up the pipe. Can't go TOO tough, though, because as pressure drops off that relax point happens and lube blowout/gas cutting/muzzle end leading can occur, which isn't good for accuracy. Using extremely hard alloy here also ruins things at high velocity for the same reason (loss of obturation) and has the added problem of being very brittle and easily abraded by the driving side of the lands.
Doing a little actual math reveals some interesting things. Using the BHN formula and Quickload models, it is possible to develop an approximate and somewhat oversimplified slow-motion movie of what is happening at launch. Take water/quenched 2.5% antimony/.5% tin alloy with enough trace other stuff so it precipitation hardens to about 20 BHN over a month. 20 BHN can take about 29,000 PSI of force before it begins to permanently deform. Now, take a 56,000 PSI PEAK pressure .308 load using Reloder 7, a fairly quick rifle powder for high-velocity applications and a charge that develops in the 90%+ of the maximum allowable peak pressure. Recipe for failure, right? Nope. The bullet is moving, and the front end has begun to squeeze through the throat already and the back end has moved a third of an inch before pressure reaches the point on the base where it permanently deforms. Guess what? By then, the front of the gas check is entering the throat the most of the bullet is quite safely inside the throat and barrel with full contact everywhere except for some stripes near the front of the nose in the groove area, the lube groove spaces, and the gas check shank which is centered at the throat entrance by the harder gas check anyway. After that, add as much pressure as you like and it isn't going to harm the bullet because it has nowhere to go but straight ahead. This situation can be made even better by selecting a bullet which starts out being able to align to center as it moves forward, BUT has some space here and there so the contact and resistance occurs incrementally from near zero on up instead of starting against the load of being a match fit with no jump. Also consider that the bullet at 1/3" travel is moving at an instantaneous velocity of 350 FPS and accelerating rapidly, so it has INERTIA to carry it through the highest point of engraving resistance...and that force is subtracted from rather than added to the force that needs to be applied to the bullet's base by expanding powder gas to cram the bullet into the throat. This means that the alloy will crush sooner and more easily on the front than the back and will allow the base of the bullet to deform essentially none due to gas pressure, i.e. NO RIVETING while the front and middle squeeze down several thousandths and in some cases elongates the bullet from the back to the front. Also, remember this is a bullet made of 10.5 BHN putty which has been artificially reinforced by quenching and precipitation hardening to nearly double the test number in "hardness", and the front goes from 29,000 PSI strong to more like 15,000 PSI strong as soon as the shoulder and front bands begin to deform into the throat. Yet, we don't crush the back end so the alloy is STILL 29,000 PSI strong. All this lets the bullet engrave easily without distorting the back, and doing that gets the bullet straight into the bore without bending, riveting, or collapsing at its grooves. The only hurdle remaining is keeping obturation up with weakened driving surfaces at 2,400 fps and 180,000 RPM a couple inches from the end of the muzzle. Remember, crushed alloy gets instantly hot, add barrel friction and a barrel that's hot to begin with, and the fact that lead alloy loses strength at a rate that's the square of temperature, so the bullet might really be something like 8 bhn at the drive side of the bands, at the muzzle. This still works though because if you calculate through F.W. Mann's exhaustive formula the pressure on the engraving/torsional bearing area of a bullet like the MP-30-180 Silhouette for a .004" land height and four grooves, the pressure is only about 10,000 PSI (a lot less than that many pounds on a lot less than a square inch, actually, but the same "per"), and the bullet metal at BHN 8 will take over 11,000 PSI of force to deform it. If your lube and alloy blend keeps the abrasion down, you can JUST pull off the launch without losing obturation and getting the accuracy-destroying drive side wear, loss of obturation (leaks), lube blowout, leading, and a damaged bullet which won't fly straight.
Now, try the same thing with 22 BHN Linotype alloy and see what you get. Spoiler alert, it's not even minute of berm accurate and will teach you right quick that there's more use for a Chore Boy pad than cleaning stuck grease out of pots and pans.
Further, try NOT water-quenching that alloy and instead add a .001"-thick, 30 BHN, self-lubricating, heat insulating jacket. Instant good results. Why? The slicker than lead coating reduces engraving force even more than a stiff but crushable alloy alone, so the 15,000 PSI strength of the air-cooled, 10.5 BHN base is enough to hold the base in form until the gas check goes in the throat and also adds surface strength and heat insulation to keep the alloy strong while it shoulders the rifling twist and hauls ass out of the muzzle without leaking or getting washed out.
I forgot where I was going with all that, but maybe if anyone actually read all that it will at least make what some of what Fiver wrote make more sense.