There will always be exceptions in battery life due to the inexactness of being able even by the makers to tell you how long they will last.
The longest lasting battery I ever had last in a car was 11 years out of a 2 year warranty Walmart battery being a perfect example. The problem of course being you can't see that in any fashion when picking one. I presently have a 'reconditioned' original SuperStart 2 year one that is now on its' 6th year, it makes zero sense but the battery just keeps coming back again and again.
When I sold car ones I saw that a far higher percentage of the upper end more expensive ones fail to make their predicted lives and why I buy a good battery but never the 'best' one. Say, car, most batteries regardless of the price or type or warranty will tend to fail in 3-5 years and doesn't matter how you maintain them, it's locked in the warranty numbers. At least here in Texas...........contrary to most belief, here they die far more in the summer than in the winter, it's killer with underhood temps of 200 degrees, batteries most definitely don't like that. That won't apply to bike batteries of course, them usually being remote to the engine and in more open air.
The amount of shaking going on of the battery figures in there bigtime, the long life I get comes from cars that sit a lot and the batteries are not subject to vibration all day long. The vibration is what shakes loose sulfation to then pile up in bottom of cells until it connects to short them out.
Why if the bike has none vibration reducing pads can make the battery last much longer. And why the AGM idea works on bikes.