Spell vamp has always been in a weird place: designed as a counterpart to lifesteal for spellcasters, it's a stat that only works on a handful of champions, but scales so well with them that it tends to be borderline broken. It's also a stat that's difficult to balance, due to its one-stat-fits-all approach: unlike lifesteal, which works on autoattacks and a select few abilities, and therefore works along pretty consistent rules, spell vamp has to deal with DoTs, AoEs, spell-based DPS and massive burst. In its current state, there's no real way to change it: buff its AoE power, and while you may make it a better choice on, say,or , it would become monstrously powerful on . Buff its single-target strength, and would become near-impossible to kill, even if you'd make better use of it on champions like or . On top of that, the stat is plain confusing: it's difficult to know what works with spell vamp because it's limited by arbitrary conditionals, namely its one-third reduced effectiveness on all AoE spells.
The big problem here is that spell vamp is trying to follow too many rules at once, when it's not made to work the same on everyone. Burst champions likeand aren't going to make use of spell vamp in the same way as DPS champions like and . Spell vamp can't be fully functional right now, because it makes little sense to group up all the kinds of ability-based healing you can have into just one unified stat. A Rioter mentioned a while back that spellcasters would benefit from different kinds of sustain: DPS casters would do best with constant, moderate spell vamp, whereas bursty champions would benefit most from tons of spell vamp within a very short window of time. With just the same component, , I think it's possible to cater to both types.
Hextech Revolver Edit
- New unique passive: Spell Vamp.
- Allows the wielder to heal for 100% of the magic damage they deal (before reductions), up to a cap of 12 + (1 × level) (+ 8% AP) health every 6 seconds.
While the reason why I reimplemented spell vamp in this manner will become clearer once you take a look at the items it builds into, I wantedto offer a reliable, yet limited amount of sustain in lane. Whether you're or or , whether you're ahead or behind, you'd still be able to sustain for about the same amount, so you don't get to overscale on it even if you generally do well with spell vamp, as is the case for and . While this healing doesn't carry very far into the late game, it's meant to offer about the same amount of healing as you'd get with the current Hextech Revolver in lane, minus the complications of having to calculate the healing you get based on magic damage, and depending on the nature of your ability.
While I tried to find an approximation for the kind of healing you'd expect to get from Hextech Revolver, chances are high this isn't the exact ideal curve for the cap on its healing. What's important here is that Hextech Revolver provides decent and consistent healing early on, regardless of what kind of AP spellcaster you are. The numbers may also seem small, but if you compare the healing you get from Spell Vamp to the healing you get from flat health regen, you end up with a significantly efficient item.
Will of the Ancients Edit
- New unique passive: Spell Vamp.
- Allows the wielder to heal for 12 + (1 × level) (+ 8% AP) every time they deal magic damage.
is the go-to item for champions who currently benefit from spell vamp, i.e. champions like , , and who all specialize in damage over time. While the individual healing may not look like much, it has the advantage of being applied to every instance of damage you deal, which means every tick of damage from a DoT and every individual ability heals you. Just as autoattack-based champions heal for a reliable amount on each autoattack, DPS-oriented AP casters would be able to heal consistently upon dealing magic damage as well.
As with, the numbers may be off, but the core idea here is to have an item that provides consistent sustain that's great for DPS AP casters (the more instances of damage you apply, the more you benefit), but that's not so good on bursty champions (the fixed healing means that you heal for comparatively less every time you deal a big chunk of damage). While the healing scales with AP, the source of damage doesn't need a strong AP ratio to scale well off of it, so it would also work well on AP bruisers, who unlike AD bruisers currently don't have any particularly good sustain options and often get out-sustained in lane.
Hextech Gunblade Edit
- Stats, active, recipe and cost reworked (see here for the rationale).
- New unique passive: Spell Vamp.
- Allows the wielder to heal for 100% of the magic damage they deal (before reductions), up to a cap of 120 + (10 × level) (+ 80% AP) health every 30 seconds.
While this may look confusing, as the above item looks completely different to(in another blog post I discussed how AD/AP hybrid items should be made either fully AD or fully AP, and so changed Gunblade accordingly), the focus here is mainly on the healing, which is the healing from made a lot burstier (again, this may not be properly balanced). Just as is made to be a great spell vamp option for DPS/AoE magic damage dealers, is, even now on live, the ideal choice for bursty AP champions such as , AP , and AP . Currently, its identity is diluted by its AD and life steal, but I feel swapping out the in its recipe for an would help reinforce its identity as a sustain option for AP assassins and assassin-y AP bursters.
While DPS champions would sort of be able to make use of this, they'd hit the cap pretty quickly, and would end up healing for far less than if they purchased a. Likewise, a burst caster could try getting a WotA for a bit more healing over time (if their spell rotation allows it), but their healing would be extremely slow compared to the massive burst of health Hextech Gunblade here would provide, which could be a life-saver in risky situations. Unlike WotA and Hextech Gunblade, I think it might be possible to remove the spell effect restriction on the spell vamp here, since no matter how much damage you deal through on-hit effects and items (including this item's active, but also the effects on and ), you'll always hit a hard cap before you'd be able to abuse the item's healing. If this does end up proving to be abusive, however, the sustain could easily be reverted to apply only on magic damage that applies spell effects, and special-case Gunblade's active to proc Spell Vamp.
Getting completely rid of spell vamp as a stat is a bit more complicated than just changing a few items, since there are champions and even a mastery that all have spell vamp in some way or form. This post isn't about retuning the sustain of these champions, since that would require analyzing their kits and suggesting potentially deep changes to their abilities, but it does raise the question of what spellvamp means to the likes of, , and : how is their sustain meant to work with their abilities? How much are they healing for each of their spells? How is their healing supposed to evolve, and what would changes to their healing impact on their playstyle? Here's what I propose to get rid of spell vamp, at least in name only:
- allows the user to heal for 0.4 / 0.8 / 1.2 / 1.6 / 2% of the damage they deal, instead of giving 0.4 / 0.8 / 1.2 / 1.6 / 2% life steal and spell vamp. This is a buff to the mastery, and perhaps a significant one, as you'd heal from certain on-hit effects and items, so it might warrant a numerical nerf. As a side note, I dislike masteries as a form of out-of-match power, which even now isn't always too visible, so it might be worth removing the system entirely.
- Runes of Spell Vamp removed. As with masteries, I dislike runes for providing not-too-visible power in a manner that puts newer players at even more of a disadvantage, while offering little in return, and so would like to remove the system. Even if that doesn't happen, though, it would probably still be worth dropping spell vamp from runes, as they don't usually lend themselves to healthy laning.
- (+ 1% per 6 bonus AD) of the magic damage she deals, reduced to one-third effectiveness on , and no longer gives spell vamp. This isn't immediately obvious, but this would allow Akali to heal from the magic damage on her basic attacks, and would no longer allow her to apply her spell vamp to 's active. Considering she's heavily dependent on the item already, and doesn't autoattack as much as she should late game, this would probably change her for the better, if only slightly. Again, I'm not trying to change Akali's playstyle or power in any real way here, though I think she should probably have her sustain reevaluated, and would have her power affected by changes to Gunblade. 's Discipline of Might allows Akali to heal for 6%
- here for the rationale). The TL;DR here is that I want to rework Lee Sin, as indicated in the linked article, and in doing so remove his sustain in favor of a significantly better shield, plus a few other things. The alternative is to simply have Lee Sin heal for a percentage of the damage he deals on autoattacks and spells, with reduced effectiveness on and . This is the exception to the champions mentioned here, as I'm pretty sure that Lee doesn't need spell vamp and have an idea of what he could do with instead. no longer grants spell vamp (see
- here for the rationale). My opinion is that Morg's spell vamp doesn't really contribute to her power in a manner that's too visible, and I'd rather take that sustain away in favor of more power towards her utility mage/support playstyle. The alternative would be to just make her heal for 3 / 5 / 7% of her spell damage, tripled on , which just goes to show how much her kit dilutes her spell vamp. is removed in favor of another passive (see
- V5.8, to the benefit of his gameplay and power progression, though I still think his ult needs to change), and that might include taking another look at his ultimate. However, this change at least would make his actual healing a lot clearer to understand. no longer gives spell vamp and instead allows him to heal for 5 / 6.5 / 8% of the damage he deals through abilities, tripled on the primary target of and . This is pretty much the exact amount of healing Ryze actually gets when he activates his ult, give or take a fraction of a percentage point. His spell vamp and spell AoE-ification sort of work against each other, so the stats listed are a lot less impressive than they look. Still, that sustain is enough to qualify as really, really strong, particularly when Ryze reaches a full build and gets to lay down a ton of constant damage. Considering Ryze's kit is really, really old and has a few problems (it's nearly impossible to catch up against him when he's ahead, plus he's fairly oppressive against melee champions), he may require a rework (EDIT: which he did in
Healing Amplification Edit
- Healing amplification, such as here, mainly because of his low focus on interactivity and decision-making, and while he should continue having a lot of sustain he probably could do without excessive amounts of healing amplification. Alternatively, the other option here is to have healing amplification increase your healing, but have that extra healing count towards the cap. unique passive and , all amplify the healing from Spell Vamp, and this additional healing does not count towards the cap (so you reach your cap at the same rate, except your total healing is increased by your healing amplification). While this shouldn't have much of an effect on , it highlights the issues with live and spell vamp, namely his propensity to overscale on it. He definitely needs a rework, which I'm proposing
Other Items Edit
- Nothing much to say about here that removes its spell vamp). (I proposed a rework to it
What do you think of these possible changes? My main goal here wasn't just to make spell vamp more popular, but also to make it less abusive on the champions who currently scale too well off of it. Did I succeed? How do you think these changes would affect champions, itemization and gameplay? Would you like to use these new items? If there's a part of this that's confusing or that could be phrased better, let me know as well. Beware, though, this post is deceptively long, so don't take it upon yourself to read everything at once.