I don’t want to be the wise ass here but as a scientifically educated person I feel I can sometimes help people understand how things work and why they work or don’t work. There’s no hocus-pocus involved.
Assume you have approx. 2200uF in caps in parallel over your battery leads, which is a typical setup in a lot of cases.
Now let’s consider a fully charged 12S system: when charged at 50V, these caps can store 2.75 Joules of energy, you can look up the equation if you like.
Suppose you want to accelerate uphill and your weaksauce-filled-Multistar-pack voltage sags to 40V, your caps will discharge 0.99 Joules of energy.
Let’s say the voltage sag takes only one second: 0.99 Joules expended over 1 second corresponds to 0.99W of power (Power is Energy over Time) available to your system.
At 40V that equates to a whopping 25mA additional current during that ONE second.
Call me old-fashioned but that just won’t cut it.
So: I agree that caps help smooth out voltage ripple (which is in the millisecond range) but I stand firmly by my statement that they will NOT help with a battery that can’t cope, and we shouldn’t be leading people to believe so either.
(EDIT: thanks to @DerBrecher for pointing out a small error in the energy calculation which is of course quadratically related to voltage)