I think this is why the issues are so ... confusing:
You're tempted to think: A plastic pool is, more or less, insulated from the ground around it. Why should I have to have a grounding network if the deck is isolated in this way? It seems backwards .... and to serve no purpose.
Yet, that insulation is the very reason we want the grid. Let's look at it again.
Assume that, for some reason, either the deck or the water become energized. With a plastic liner acting as an insulator, you have just allowed a voltage difference to exist. Voltage difference = shock hazard.
Bond the two together, and the two are kept at the same potential ... no more shocks. Just as important, with a good bond, the fault is more likely to trip the breaker. In this arrange, it is JUST AS IMPORTANT that the 'water be bonded' as it is to have that grid.
While this approach doesn't solve every issue, it is hoped to lessen the risk. It appears this whole situation came about the moment we started using those plastic drop-in pool forms; before then, it appears that there was enough conductivity through the pool walls to prevent this from being an issue in most cases. The prevalence of plastic piping has probably also contributed to this 'voltage gradient' issue.
I wouldn't be at all surprised if the day comes when we look at deliberately treating the water for conductivity - making it sure it's very conductive - as well. We might even require the use of conductive pipe.
Remember .... it's not just a question of pool equipment. Indeed, I have one customer who, for religious reasons, deliberately constructed a pool without directly connected filters, pumps, or metal piping. Yet, there are still the lights in the pool room, the radio / sound system, and all manner of possible introduced electrical sources. Even though that particular room has no receptacles - there's no way to prevent someone from running a cord.
