Right but (I may have edited while you posted this), keep in mind that this is exactly how WHM functions now. You bank Misery for the buff window. But you can only bank ONE Misery. If you have a full Blood Lily now, you can't cast 3x Solace to get a second Misery. Each one you cast is costing you 1/3rd of a Misery until you cast the Misery you already have. That is, a player couldn't spend 30 sec or whatever before the buff window chain casting Cure 2 to bank 5 Miseries to unload in the buff window. There's no allowance for that unless you can stock several Miseries, and you can't. So you have to wait for the buff window to use the one you have.
In live, you have two options at that point. (1) is to refresh Dia and then chain cast Glare to fill out the rest of the buff window. (2) is to cast any GCD heals and have damage potency losses and see your DPS decline. Obviously, unless people are dying, this is the worse option and so not really an "option" at all. (2) is being wrong/bad.
If Cure/Medica spells generated 1/3rd Blood Lily, then you'd have the same situation. You could only stock one Misery going into the buff window. Once you use it, now you can Dia + Glarespam OR you can Dia + heal. If you heal, you can heal twice more to get in a second Misery. This would do the exact same damage as casting the 4 Glares, as all of this would occur in the buff window. But it would not do MORE damage, since you'd be sacrificing 4 Glares to do Cure 2/Medica x3 + Misery. Since they are damage neutral, the outcome would be the same. Glare x4 would be marginally safer on damage (more chances to get a crit) while the Misery would be "all or nothing" (if it crits vs doesn't crit), so the Dia + Glarespam would give more reliable damage, but the Cure x3 + Misery would be damage neutral on long fights. (They would be expected to give the same damage overall regardless of crit values or whatnot)
So there wouldn't be any "wasteful healing to prep damage". Well, no more than there is right now. That is, this change would make things no worse than they are in live with people burning Solace/Rapture to stock a Misery for the burst window.
The only real differences are (a) inside the burst window you'd have an option other than Glarespam, and casting a GCD heal (if you could do two more and a Misery after it still within the window) would not be a DPS loss, and (b) outside of the burst window, you would break up the Glarespam more frequently since you'd have other buttons to push and be getting Miseries more often by doing so. The best part is, if someone wanted to continue playing as they are today (never touching Cure/Medica), they could do so and their damage output would be (on average) identical.
I'm telling you, as a non-optimizing, not-hardcore, green-parser (rare blue on a good day, purple in 24 mans...no idea how that works, but I don't parse so just see my name captured on other people's runs from time to time), this would not make things worse for people like me.
It's strange to me you think this would widen the gap between low and high skilled players. I don't see how unless we made it where you could stock more than one Misery, which we'd be not doing... I'm a bit confused is all. I can only see it lowering the gap, not increasing it, as high end players would mostly play identically to how they do today.
I could have sworn sure they (Yoshida and Starfox, in one of the first LLs) noted exactly that factor in explaining why we were seeing double-SCH so often even in T1-4 (rather than only as a fix to Infirmity in T5, since Lustrate at the time healed for 25% HP regardless of anti-healing debuffs); since SCH had effectively auto-regen and a much less wasteful 3/4s of a Bene every minute, instead of 1 per 5 minutes... yeah, SCH was kicking WHM's butt overall simply because WHM had to drop significant damage (and mana) to do any healing while SCH didn't -- i.e., a gap in oGCD output. It wasn't particularly unknown or undiscussed even back then. It was ostensibly part of their reasoning behind adding Tetra and Asylum (or simply, a greater increase in oGCD throughput to WHM than to SCH).
I don't recall a lot of SCH/SCH comps back then, but I was paying far less attention to the meta. All of the ARR-era videos I've seen of Coils clears and Extreme clears had 1 WHM + 1 SCH. I'd have to see this Liveletter. I'm not saying I don't believe you, I just don't remember that being a thing back then.