Hi Lionel.
I have looked high and low for something quantitative in that regard but never found anything. I sure the makers know about this and have no interest what-so-ever in it getting published and no one else is interesting in destroying expensive batteries to gather useful data or if they have, they have their own reasons for not wanting to publish it.
I'm stating that on my experience of using or working with probably ~300 gelcells in my career. Early on gelcells were fairly robust. Some like Exide state their gelcells are more robust than AGM batteries and they could be. Gelcells can be made very robust by having extra electrolyte and less antimony for plate strength owing to supporting gel etc.
However, about 10 years ago the cells I work with; Panasonic, BB Battery, Power-Sonic, GNB, and Yuasa all got extremely touchy about the same time. A battery that we'd get new, install in a test and run down overnight would be found to have lost significant capacity in the exercise. This got progressively worse until we actually got some that lost essentially all capacity when severely drained overnight and then attempted to be charged the next day. These were batteries purchased from reputable places like Digikey, received, fully charged by us, and used and completely useless 48 hours later.
Perhaps in a specsmanship war the makers all upped their Ahr ratings in the standard footprints by more plates and insufficient electrolyte or some other cost cutting measure. Now I treat gelcells with extreme care and attention. Any I have around I recharge every 30 days if they have been out of circuit or leave floating.
Keith Cress
kcress -