Not impossible.
Collect all reports of battery failures since the Evo launch. Then try to correlate these events with the use of SBC kernels.
Correlation does not make something 100% true. By that argument, 100% of the fire-phones are EVO's. So we can conclude that the EVO is a fiery deathtrap.
Since we don't have access to the fire-phones pre-fire. We have no proof that it was SBC that caused it. Not saying that SBC isn't dangerous. It may cause fires. But these 6 examples aren't necessarily caused by SBC.
I have a BS in EE. I get the difference in voltage and current. I'd put SBC dangers at "likely" based solely on the brief physics explaination despite the fact that I have no idea about how these charging algorithms work (I haven't used or looked into SBC at all).
However, I also understand probablity and statistics. We already said that theres no where near 60,000 SBC users... so the sample size is going to be quite small. Correlations at this point may be just very odd probability.
PROVING the potential dangers of SBC are going to be based on physics and engineering.
The "the truth must be known" argument goes both ways. It's dumb to argue "This kernel is 100% guaranteed safe" for more reasons just beyond this possible way to damage batteries. However, its also dumb to argue "omg this kernel will 100% destroy your phone". Allowing the user to make an informed decision is important, presenting it in a fear-mongering way is not.
Airplanes crash, its part of real life... And therefore on EVERY flight (in america anyway) they are required to go over emergency procedures before the plane takes off. Its presented in a "hey, this can and DOES happen" way. not a "hey, this is GOING to happen" way.