I think that post was from someone that was considering using batteries to power the IoTaWatt and was interested in computing how much mAh capacity would be needed.
What I would add to that conversation is that the IoTaWatt uses very little current for ordinary work, but uses a lot when transmitting WiFi. The IoTaWatt uses the nodeMCU 3.3V supply, which in turn comes from the USB5V supply. If you look at the sampling data when a power supply is inadequate, it first shows up in those samples. It’s not really that noticeable in the overall power readings, but then again I haven’t tried to see how bad it gets before it doesn’t work at all.
When using a standard USB wall powwer supply, typically rated at 2A or more, the impact of the radio on sample voltage is not detectable. It’s worth noting that none of the powered USB hubs that I have work as well as a dedicated power supply.
To wrap it up, as I understand it the batteries may well be able to provide a lot of energy over time, but the mAh rating doesn’t offer any guarantee that they will be able to provide the needed current during radio operation.