You’ve mixed up different points that weren’t connected together.
Stealth detection:
Spy, and similar addons, can only detect Feral/Rogue if they carelessly entered stealth in close proximity to you. If they are experienced players - these addons won’t detect them. So even with this addon - you still don’t know that there might me a stealth class around, and by that - unprepared.
With your idea - it’ll be possible to always know that there are a Rogue/Feral around, and how many are there. That means that any class can at least keep themselves in combat to avoid sap, use AoE a lot, use stealth detection abilities/potions, etc - and that will cut stealth advantage almost completely.
There is a big difference between guessing/not knowing and being completely ready.
Bonus and intervals:
The problem is not for how many players it’s going to change. The problem is - it won’t be able to change when that change is suppose to happen, because of being on interval.
In your idea the bonus is for an area in a zone on a shard. Let’s assume that there is less Horde and more Alliace in the area with that Elite WQ for a lot of AP, so Horde gets 30% and Alliance gets 10%. But Alliance finishes WQ and disbands, Horde gets more players for WQ - bonus should change or this system doesn’t make sense right? But it won’t change in time, because there is an interval.
This kind of dynamic system that you suggest won’t work with intervals, it needs to be in real-time to be accurate and have a chance to work.
Even this mixed example is not a reason for a system to not work, or for stealth to be easily known and ready for.
GW2 WvW is different to WoW’s shards. In GW2 there are pre-determined servers and separate maps, instanced, and there is a queue that is also a trigger for recalculations. It’s a huge epic BG basically.
To make something like that work in WoW open world - you need to add queue for open world locations…
With this, technically your idea could work only if it would calculate bonus for a whole continent, like Kul Tiras/Zandalar, and all zones/areas on that continent will have the same bonus. Loading screen for the continent will act as a queue and a calculation trigger.
But because of phasing - it’ll ether be the same as current system and will calculate bonus for the whole region based on zone population in the region, or it’ll have the same problems with recalculation/intervals/etc, as we’ve discussed above.
So interval is a minute? Every minute system has to get and compare data from all players/shards/zones/areas and recalculate bonus for each area in every zone on every shard. So just huge lag spikes every minute? And all that for a few % that will never feel fair?
Even if we imagine an ideal situation of game servers upgraded not only to solve current lag problems, but also new from that system - because of bonus constantly switching it’ll be another system set up for disappointment. You start WQ thinking you’ll get 20% bonus, next minute it changes for 10% while you still filling that bar or whatever, and the next minute other faction left and you get 0% when you about to finish it. So now, you ether have to wait and hope for a change on the next minute, or accept no bonus and that you’ve risked entering that WQ area and fighting with other faction for nothing. Same for any vice versa. It’s going to be a confusing mess.
As I’ve said before - unfortunately tinkering with a bonus to make it into some complex system is not worth it, because there is no gain and a lot of needless problems.