One of the more noticeable trends this TI was the overwhelming prevalence of high-burst strategies. This is a very understandable trend: you blow up an enemy, you are already at an advantage and it wasn’t surprising heroes like Tiny were valued so highly, along with burst-supporting characters like Elder Titan and Snapfire. But was this trend based solely on the realities of the International, or is there currently a problem with saves in general? Let’s have a look.
There are several support heroes in the game with saving abilities, but most notably we have Dazzle and Oracle a.k.a one pick and five picks at TI respectively. While the overnerfed Dazzle who lost most of his support identity in 7.30 is understandable, the lack of Oracle should have been a surprise.
The hero is exceptionally versatile, has access to both damage-dealing and damage-mitigating abilities and is arguably among the most useful late-game supports when coupled with high-impact late-game carries abundant during the International.
Similarly, heroes like Omniknight were almost non-existent while Shadow Demon was picked only ten times. The latter doubles as a setup hero with strong purge and was unlikely to have been picked as a save hero specifically.
Abaddon, another save hero who also has access to strong dispel was also quite unpopular with only eight picks total and the only “save” hero with a modicum of success coupled with relatively high popularity is unironically Legion Commander, with her 29 picks and ~51% win rate.
What do most of these heroes have in common? They’ve all fallen prey to a cast range inflation-deflation cycle in the recent past and are now in a very unfavorable spot that doesn’t allow them to fully utilize their kit against competent opponents.
At first there was Aether Lens, then came the Talents and finally there were Neutral Items. All of the above could and still can potentially provide extra cast range. This was a massive buff to most supports, but save supports specifically, who usually have a pretty straightforward job: don’t get caught and use your saving ability at the right time on the right target.
To make this balanced, a lot of heroes received noticeable nerfs to their cast range. For example, Abaddon Mist Coil went from 800 cast range to 575 over the last couple of years, while Oracle had his False Promise cast range reduced to 700/850/1000 from a flat 1000.
This wasn’t a massive problem, as at the time there were many different ways to solve these shortcomings and even end up in a position where you were ahead Cast Range-wise. Cast Range talents were boring but highly effective, while Aether Lens gave roughly 10% more cast range. It all changed in 7.28, where most talents giving this stat were removed.
It wasn’t a problem for initiating supports like Lion, who would usually end up buying Blink and would go point-blank towards the enemy to initiate anyways. It wasn’t a problem for most teamfight supports like Snapfire, who have a clear spell usage priority and above average tankiness. It wasn’t a problem for Elder Titan, who was the most popular pick of the International.
But it was and still is a problem for save supports who are always a number one priority target for the enemy and who usually don’t have access to counter-initiating abilities that would make going on them even remotely risky. Without the ability to stay very far away and without access to talents that would usually solve this problem, these supports were forced out of meta and we hope the next patch addresses this problem. While we discuss another problem with save in general.
Similar to most save heroes, two most iconic save items were also nerfed in terms of their cast range. Glimmer Cape and Force Staff are now both at 550 cast range, down from 800 and 750 respectively. The difference is 225 on average, the exact amount Aether Lens now provides, but for a position five support getting Aether Lens first and then getting Glimmer Cape can be quite problematic.
This problem was mitigated by talents and this problem was definitely nullified in cases where a support could get access to both a Talent, an Aether Lens and a Cast Range-increasing Neutral Item, but it is no longer an option in 7.30e.
We believe this is the reason most high-tier supports don’t default to these items anymore. Looking at the International, specifically, Glimmer Cape is below items like Medallion/Solar Crest, Aether Lens, Aeon Disk, Ghost Specter, Urn/Vessel and Eul’s in terms of popularity. Aghanim’s Shard was also purchased almost six times more frequently than Glimmer Cape, though it isn’t a support-specific item, so the stats are less meaningful.
Force Staff did considerably better, with 264 purchases, but do keep in mind that it is less support-specific as well and was often purchased on ranged cores going Dragon Lance into Hurricane Pike. It still works wonders as self-save, but it is definitely below par in terms of reliability when it comes to saving teammates.
It is very hard to react to an initiation with a Glimmer Cape. If you are in a position where you can use Glimmer without Aether Lens, you are usually going to be the target of initiation. If you are not initiated on, then by the time you can close the distance and press it on your teammate, they’ve probably already lost most of their HP to magic burst.
Same with Force Staff as a save item. We are not arguing towards not building these items and they are still very, very useful. However, we are arguing that they are in a dire need of a buff, along with save heroes, as it feels like the initiation advantage in the current meta and in the previous several patches is overwhelming.
Going for initiating and teamfighting supports with initiating and teamfighting items should be one of the options, not the only viable default and we hope that by the next Major Tournament Dota is going to be in a place where a save and counter-initiation-heavy playstyle will be allowed to shine.