Will of the Ancients
Spell Vamp 20% Spell Vamp --> 0% Spell Vamp
New Passive: Your spells and abilities heal you for 15% of the damage dealt, calculated BEFORE your opponent's resistances. 33% effect for AoE Spells.
 
The hope here is that there's a lower spell vamp value and less healing generally in exchange for not needing so much to get started.

For those that may not be able to access the changes I will list them here. Note that Abyssal Scepter has already changed in recent PBE patches. Keep in mind that this isn't entering PBE until next cycle which will be in a week.

Blasting Wand

Cost: 860 G --> 850 G

Needlessly Large Rod

Cost: 1600 G --> 1250 G

Ability Power: 80 AP --> 60 AP.

Context on the Needlessly Large Rod change.

NLR being at such a high price / AP point means that almost any item that builds from NLR would have to have greater than 100 Ability Power on it when combined with any other Ability Power component. This proved to be a pretty huge constraint in using NLR as a stepping stone for certain items - going to retarget NLR as a smaller item. While this does have some implications on lane timing (You no longer need a double kill to get NLR first back, for example) - It was either this or inflate AP as a whole across the board. This has some... interesting ramifications in terms of the stability of mid lane purchases.

Rabadon's Deathcap

Recipe Change: NLR + Blasting Wand + Amplifying Tome

Total Cost: 3300 G --> 3500 G

UNIQUE Passive: 30% Ability Power Amp --> 35% Ability Power Amp

Zhonya's Hourglass

Total Cost: 3300 G --> 3000 G

Ability Power: 120 Ability Power --> 100 Ability Power

Luden's Echo (Summoner's Rift)

Total Cost: 3100 G --> 3000 G

Ability Power: 120 Ablity Power --> 100 Ability Power

Movement Speed: 7% Movement Speed --> 10% Movement Speed

Context on the Luden's change

Luden's is a terrifying item in terms of the general poke power it adds - Luden's probably needs a more solid identity that 'raw power' - so we've been pushing the item to be more and more about being mobility focused rather than spam spell focused. I suspect we'll eventually need to do more to the item but it'd most likely be more of a slow burn.

Archangel's Staff (Summoner's Rift)

Recipe Change: NLR + Tear of the Goddess + 1030 G

Total Cost: 2700 G --> 3000 G

Ability Power: 60 Base AP --> 80 Base AP

Seraph's Embrace (Summoner's Rift)

Ability Power: 60 Base AP --> 80 Base AP

Rod of Ages (Summoner's Rift)

Total Cost: 2800 G --> 2700 G

Health: 450 Base Health --> 300 Base Health

Mana: 450 Base Mana --> 400 Base Mana

Passive: 20 Mana Growth / 200 Maximum Mana --> 40 Mana Growth / 400 Maximum Mana

Passive: 2 Ability Power / 20 Maximum Ability Power --> 4 Ability Power / 40 Maximum Ability Power

Context on RoA change

RoA is currently leaning a bit too much towards favoring AP Heavies with the statline - while not really solving Mana / AP concerns for mages who want to pick it up. Going to adjust the item to more favor the Mana and Offensive statistics of this item, especially over time.

Nashor's Tooth

Total Cost: 2920 G --> 3000 G

Attack Speed: 50% Attack Speed --> 40% Attack Speed

Ability Power: 60 Ability Power --> 80 Ability Power

Context on Nashor's change.

We're also pushing more items up to compete in the 80 Ability Power range - While Nashor's is a fine purchase for people who already have strong on-hit procs - the ratio of Attack Speed to Spells favored these characters probably a bit too hard.

Rylai's Crystal Scepter

Recipe Change: NLR + Amplifying Tome + Giant's Belt

Total Cost: 2900 G --> 3000 G

Passive Single Target: 35% Slow for 1.5 seconds --> 40% Single Target Slow for 1.5 seconds

Passive AoE: 15% Slow for 1.5 seconds --> 40% Slow for 1 second

Passive DoT: 15% slow for 1.5 seconds --> 20% Slow for 1 second

Note: If a spell is both an AoE and a DoT - they use the DoT value.

Context on Rylai's change.

Rylai's is currently good in a few scenarios, namely - do you have a spammable single target spell and a train pattern - a large part of this is the fact that the AoE / DoT portions of the slow are so weak in comparison to the single target slow. Changing the tactic here to make those slows have less overall duration rather than penalizing the direct effect. This is a very powerful push in terms of Rylai's overall power - but we'll be looking at modifying and cleaning up of the slow rules as well either now or in the near future.

(Also added a self-only freeze particle effect so you can watch everyone become ice cubes when you're slowing them. FOR CLARITY.)

Haunting Guise

Total Cost: 1480G --> 1500 G

Liandry's Torment

Recipe Change: Haunting Guise + Blasting Wand

Total Cost: 2900 G --> 3000 G

Ability Power: 50 Base AP --> 80 Base AP

Context on Liandry's/Guise change.

Liandry's Torment is designed to be able to fit as a High Health / Low Resistance shredder (hence the combination of health damage + flat penetration). However, the lower AP on the item frequently meant your kit didn't actually work. Aggressively pushing the power of this build to be able to compete with the more immediate damage type builds - like proc builds.

Will of the Ancients Xypherous has mentioned that he is definitely considering changing Hextech Gunblade in the future. I also asked about revolver and he said there will be no change to it.

Spell Vamp 20% Spell Vamp --> 0% Spell Vamp

New Passive: Your spells and abilities heal you for 15% of the damage dealt, calculated BEFORE your opponent's resistances. 33% effect for AoE Spells.

Context on Will change.

You need so many things to get a Spell Vamp build to work. You have to be resistant enough to not get bursted. You have to deal a sufficient amount of raw damage in order to heal. You probably need to be ahead to outscale the opponent's resistances. You need a ton of base damage. Therefore - this change is trying to reduce the number of dependencies needed in order to 'turn on.' You won't need Pen to heal from this item on champions. Spell Vamp builds are fairly problematic to control once they get rolling - so they tend to be pretty binary. The hope here is that there's a lower spell vamp value and less healing generally in exchange for not needing so much to get started.

Morellonomicon

Recipe Change: Codex + Idol + Amplifying Tome

Total Cost Unchanged

Context on Morello's change.

Wanted to increase the number of potential purchase options at the ~1200 G level.

I'm attempting to make Codex + Book or Wisp + Book potentially attractive choices compared to NLR, especially on your first back. You'll get more interim power than the NLR - at the slight cost of strongly telegrahing what you're going to build and locking yourself into a build. It's mostly an experiment to see if players like having this kind of choice between 800 + 400 component vs. ~1200 mid-tier.

Athene's Unholy Grail

Recipe Change: Codex + Chalice + Amplifying Tome

Total cost unchanged. Same reasoning as Morello's

Void Staff

Total Cost: 2295 G --> 2500 G

Ability Power: 70 Ability Power --> 80 Ability Power

Context on Void Staff change.

Void Staff's raw efficiency is pretty high compared to some of the other items on the list.

While % Penetration will always be a great multiplier for damage for any damage oriented mage - you don't really need it to also be such a great source of Ability Power. This pushes Void Staff's efficiency more in line with some of the other items with comparable multipliers. Also adding a bit of AP to ensure that players end up at roughly the same amount of AP they did pre-patch.