Demigod doesn't use a Rock/paper/scissors approach when balancing indivdual Demigods as a whole and so the method of reasoning you've detailed doesn't apply here.
I'm surprised to see you write this, since my entire post was spent trying to explain this very concept to you, who insisted that on the bad-ness of two "scissors" being able to overcome one "rock." So uh, we agree. DG is not really a RPS game, which means there is no RPS system of counters that comes crashing down when doubles (builds) are introduced the equation.
Anyway, the problem isn't stacking Demigods, it's in stacking builds. I've already gone through this, but I'll detail it again in the hope you understand better. We'll look at Snipe. The ability, based on it's atributes, is most effective when used as a finishing tool - Snipe a fleeing enemy, as the further away you are the more damage you do. It can also be used to help bring down a Demigod someone else is attacking. This is clearly the abilities intended purpose - long range, modest damage. Let's add four more snipes into the picture. You now, if all are the same builds, have 5 long range attacks all doing modest damage. If they focus fire a single opponent, depending on that persons HP stacking but regardless of their build, they have a real chance of being able to take them down without placing a single person on their team at risk of any kind. This is not what Snipe was intended to do - it's no longer used as a long range finish tool, rather a long range method of wiping out a single player. This is where my complaint comes from. Take any Demigod defining ability - Towers, Snipe, Hammerslam, Spit, Heal, Ooze, Pounce, Fireball, etc - and multiply it's effectiveness by the number of people stacking a build off of which the ability forms the base and you have a problem. 40 Towers on a single map? Good luck with that. 5 times the surviveability of Sedna's Heal? Hope you like losing. 5 Snipes? Hope you weren't planning on doing anything other than running back and forth from the Health Crystal. Detail counters to these types of build stacking and you'll notice you're forced to play specific with methods to stand a chance - only a handful (more than one, another way of saying multiple, as in several combinations - as in not a R/P/S system) of combinations would work. And if use of items enable the ability stacking team to counter their supposed 'counters' then we have a real problem on our hands.
Snipe is the one and only ability in the game that I think is slightly problematic in that respect, and only if there are 4 or 5 of them. Any less and its not. Other than 5 snipes, your claim that "demigod-defining abilities" ("DGDA) become stronger when combined with themselves as opposed to different DG-defining abilities is unsubstantiated. You can take two different DGDAs and be just as powerful and more flexible than just stacking the same DGDA. If you've lost before to stacked DGDAs, then I surmise that was far more to do with you being outplayed then the double.
This goes back to what I was saying before, scrubs misattribute their loss to some breach of protocol which should be banned, where in fact there was no breach and they simply were defeated by a strategy that they have arbitrarily determined to be objectionable.
You've contradicted yourself here by listing X and Y as counters, then stating there are no Demigod counters as such, rather Counter builds. As you'll haven't noticed, thats exactly what I'm saying - the problem I have is with the Builds, not Demigods themselves. So I'm not sure what you're attempting to argue here, but you need to be more consistant in your arguments.
Sorry, I should have been more explicit with where I refer to builds and DGs. But if we were to get very specific, we should actually be referring to "skills and tactics." Certain skills and and tactics counter other skills and tactics. Certain skills and tactics are resident to specific builds. Every build is tied to a specific demigod. Because of the wide variety of skills and tactics available on each DG, it is difficult to imagine that within a reasonably sized game one could invent a doubled team of DGs that make a certain DG completely unviable for the enemy team, therefore introducing an element of pre-match RPS. This, combined with the failure of the "imba stacking DGDA claim" leads to the conclusion that doubles are fine.
If a stacked team is just as fine as a normal mixed team, where's the harm is allowing people to disallow them in their games? It's just another team combination right? So what's the big deal? House rules, and all that.
Oh I don't care what people do in their own custom games. To me this is more an argument about what the standardized rules for a hypothetical team automatch feature should be.