Hi Jay (thanks for taking this to the forum),
There are lots of factors that will lead to a good or bad tape pack.
Let's look at what happens during fast forward or rewind which are the worst case.
The speed that the layers of tape are laid upon each other determines the pressure of the air that's trapped between the layers. That creates a cushion of air that allows lateral forces to move the tape side to side as the tape is laid down. The faster the tape is being laid down, the higher the air pressure.
Obviously, if a reel motor assembly is seriously out of whack than all of your tapes would have a bad tape pack when moved to that side. But, it's kind of amazing just how far a motor has to be out of line before it effects the tape pack.
When I was experimenting around, I put 2 spacers between the motor and one of the 3 mounting screws moving that 1/3rd of the motor about an 1/16th of an inch on the take up side. To my amazement, I still got a decent pack in play. The tape was closer to one flange on one side compared to the other but if you looked at it from the side, it still looked pretty good.
That's a great example of the tape being smarter than the operator!
The most obvious reason for a bad tape pack is reel flanging. Reel flanging is when the reel is somehow out of line and the tape rubs it on each revolution. Most of the time, you don't need to be told that this is a problem since the noise it makes is real annoying but sometimes if it's just barely touching, and only at certain depths of the pack, it can escape detection. Most of the time this is because of bent flanges and if it isn't both sides, you can just flip the reel when it's empty and the problem goes away. If it's seriously bent, you're probably better off putting an important tape on a new reel since the flanging causes tape edge damage.
Sometimes the screws that hold the flanges to the hub slip and you'll get heavy vibrations when the reel gets full. Taking that reel apart when it's empty and rotating one of the flanges and/or flipping the hub can almost always help getting a more balanced reel. There are only 6 ways it'll go together and trial and error is the only way I know of to do this.
The last possible reason for getting a bad tape pack is the tape itself.
The tape guides have a tiny amount of wiggle room. If they were exactly 1/4", irregularities in tape slitting resulting in a slightly wider section of tape (by the tapes manufacturers) would cause the tape to either ride up one side of the guide or cup in the middle. Either one would cause tape damage in the long run and you'd have tons of oxide shedding because of that.
So, the guides are all a bit wider than 1/4". This however requires very careful alignment of the guides and heads.
In a perfect world, the tape would move right down the center of each guide, never touching either side and the heads would all be perfectly flat with no grooving and exactly aligned with the guides. If one of the heads is slightly out of alignment in anything but azimuth or height, the tape will try and climb up the slope of the head until the guides that are immediately before and after that head stop it's lateral movement.
Well, in the real world you can only achieve this kind of perfection some of the time. On my machine (which I constant fiddle with) I'd guesstimate that it happens about 1/2 the time in play with half the guides. Most of the time when the tape touches the guide edges, it's on the inside edge,.. in play. In rewind the tendency switches toward the outside edge of the guides. Your machine could be the opposite but it almost certainly has one tendency or the other depending on the tape direction. Maybe on a better engineered and refined machine this doesn't happen but stuff like bearing wear, surface wear (of the guides and heads), alignment of the shafts that bearings are mounted to, chassis alignment (which the shafts are mounted to) well, you get the idea. There are lots of elements that lead to the tapes lateral tendencies and in my case, I think it's bearing slop.
The result is that the tape rubs against the guide edges to one extent or another and that causes tape edge wear (along with scrape flutter). The more times a tape is played, the more edge wear happens. If other elements of a particular tape such as slight reel issues are present, the wear is increased.
Some edge wear has to be expected (maybe something like an ATR or 820/80 avoids this the most) as long as there are hard guides and the tape is on a material that will wear. This doesn't mean that the tape will sound bad but as the tape edge becomes textured and slightly uneven, the tape pack will suffer.
I have two tapes that have some edge wear and I actually get a better tape pack with them in rewind.
Why?
Just like I said, my machine tends to move the tape outward in play and inward during rewind. That shift moves the worn outside edge away from the guides and put a nice less worn side against the inside of the guides allowing the tape to move more smoothly.
I obviously had my tape path (maybe tension too?) poorly adjusted and the outside edge of the tape was shoved against the outside of the guides too much for too many plays.
I hope I've learned a lot about adjusting my machine since then.
BTW, those two tapes still sound wonderful and are still in the "heavy rotation" queue.