Self discharge is not always linear. So where a certain cell design may exhibit a drop-off of X % for the first days/week, a plot will often show an exponential decay. In some cases, this is directly related to ESR (or Re) of the design.
The fuel rods appear to be merely 18650 cells and what we also don't know is if they were ever subjected to an overcharge or over-discharge state. The built in BMS should be robust enough to detect a VCO on charge/discharge (or thermal runaway), but the damage done from even a single cycle can cause excessive dendrite growth on the anode/cathode and even the best separators are vulnerable.
We also don't know the charge rate, not just the number of cycles. Charge rate is directly related to MPV and can also cause a loss of capacity due to the active material(s) being damaged. I suspect these cells are rated at C/6 or higher.
Many, if not all, batteries have a limited number of cycles although super capacitors generally can be in the hundreds of thousands, if not millions, depending on many factors (DOD, rate, temperature, etc.). But, the energy density limits on super capacitors makes them very limited on energy sensitive applications (they do excel at power based applications).
