We can certainly open this back up for discussion. Ammo went in at the same time that I tweaked Chill Touch to only have a 50% chance of inflicting Drained, so a lot of things changed all at once.
Though both Chill Touch and Devourer's Caress have lower damage/effect numbers overall than they previously did, that was part of a downward adjustment for all ammo-using attacks. Provided you're using appropriate ammo, that overall adjustment should be balanced out by the increased base damage and effect power from the ammo, or at least that was the intention. Unfortunately, I suspect there might be a problem with the Cure aspect of Devourer's Caress. I just did some quick testing and it doesn't look like the Cure amount is ramping up as you use better ammunition, though it does ramp up properly as you match more standard keywords. Most other effects do seem to be increasing with better ammunition, but Cure is a little special. Maybe there's a problem with just that effect, or with Self/Beneficial effects on attacks that mix Beneficial and Damaging together. We'll have to do some more digging to see exactly what's going on, but that could account for much of the feeling that this combo was nerfed.
I only made two intentional changes to the combo. The first was that in reducing Devourer's Caress to account for ammo, I focused pretty much all the reduction on the Cure aspect, instead of spreading it out with the amount of Exhausted or the base damage. Combine that with ammo not boosting Cure back up and you have a double whammy, but I could potentially return things a little closer the original balance between Cure and the other effects/damages for that attack.
The second change was the main one that's been brought up, lowering the percentage chance for inflicting Drained with Chill touch from 100% to 50%. The basic idea was to make things just a bit less predictable so you'd have to keep a closer eye on things, but it's true that there's a 1 in 16 chance of failing to apply Drained after 4 attacks, which could be such a high risk that players might not be willing to even risk rushing in. One option would be to raise the percentage to 75%, at which point there'd only be a 1 in 64 chance of failing 3 times. That's a lot more dependable without being a guarantee.
I could also raise it back up to 100%, and it wouldn't be the only case where a weapon has an attack that dependably sets up another attack on the same weapon. That said, this is clearly one of the most powerful of those combinations, though changes like ammo and line of sight may have cut that back a bit. Plus, whether I raise the percentage to 75% or 100%, I'd probably trade it off against providing a smaller amount of Drained with each success, so the effect would tend to fall off a little more often.
At the end of the day, if this is a little overpowered, but at the expense of burning through expensive charges (they're very quick attacks), that could be okay.
LOL!!! How did I know this response would be coming!