Shaman chain heal rules
So, I was taking a turn in a game in which I am playing as Tribe. I was making a recovery move in which I was trying to maximize healing done by my shamans, and it appears that the following rules apply to the bounce behavior of the heal move:
1) Range restriction of 2 squares for the initial target and subsequent bounces
2) If multiple wounded targets are within the 2 square range for a given bounce of the heal, then
a) KO'd targets are preferred over non-KO'd targets
b) lower HP targets are preferred over high HP targets
The problem I have seen is that a lot of healing potential is wasted by the current function of the chain heal. While it is obvious that a KO'd unit should be a target of the heal, often (due to AOE damage) there are multiple significantly damaged units near to the KO'd unit. Because KO'd units receive 1/2 of the healer's power in healing instead of 2x (before effects such as the chain bounce), a KO'd unit should nearly never be the primary heal target of the shaman. Even if you target another damaged unit to receive the first full heal amount, the heal will prefer the KO'd unit to damaged units, meaning that instead of (for instance) healing a damaged unit for 200 before reviving the KO'd unit to 25 HP, a total of 225 healing, the bounce will go first to the KO'd unit for 50 HP before healing the remaining unit by 100, a loss of 75 potential healing done.
I realize that the 75 healing done is minor, but in circumstances involving AOE damage this could mean a gain or loss of an AP for the opponent, and could therefore be significant. My question to the developers is whether or not this behavior is intended, and if it's likely for a tweak to this mechanic to occur. The Tribe has more significant issues to deal with (rare AND weak AOE instead of easy + weak or rare + strong AOE against TF2, for instance) but I thought this was interesting and thought I'd start a thread.
Thoughts?
The sequential jump calculation is also really annoying because it means that you can't jump through a full HP target. i.e. no chains of sequence
Initial unit
Full HP unit
Non Full HP unit.
It's my view that there should be an option to be able to specify targets for chain attacks, then there would be no asymmetry and you don't have the sub optimal target algorithm confounding with any other factors when balancing the game.
The current rules allow experienced players to control what they hit and which of their units may be hit. I feel it is important to preserve that. Slight burden of knowledge issue but it's not hugely consequential
The sequential jump calculation is also really annoying because it means that you can't jump through a full HP target. i.e. no chains of sequence
Initial unit
Full HP unit
Non Full HP unit.
It's my view that there should be an option to be able to specify targets for chain attacks, then there would be no asymmetry and you don't have the sub optimal target algorithm confounding with any other factors when balancing the game.
These chains can happen, as long as there's nothing else pulling the chain away from the full HP unit.
The sequential jump calculation is also really annoying because it means that you can't jump through a full HP target. i.e. no chains of sequence
Initial unit
Full HP unit
Non Full HP unit.
It's my view that there should be an option to be able to specify targets for chain attacks, then there would be no asymmetry and you don't have the sub optimal target algorithm confounding with any other factors when balancing the game.
These chains can happen, as long as there's nothing else pulling the chain away from the full HP unit.
I'm sure they can happen, but they sometimes don't, even if they "could". I see this as a problem!
Example of it not happening:
Challenge Collection -> Healing waves -> Do the following AP:
1. Move VM back 2 spaces
2. Move VM back 2 spaces
3. With shaman, heal the Axe Thrower directly below
4. With shaman, heal the Preistess directly above
5. With shaman, heal the Axe Thrower directly in front.
On AP5 the heal ends with the Axe thrower, even though it "could" jump to the VM.
How it "could" happen.
1. Move VM back 2 spaces
2. Move VM back 2 spaces
3. With shaman, heal the Axe Thrower directly infront.
4. With shaman, heal the Axe Thrower directly in front.
On AP4 the heal makes the chain you'd expect to have in my example above. This is what I mean by it "could" happen.
The current rules allow experienced players to control what they hit and which of their units may be hit. I feel it is important to preserve that. Slight burden of knowledge issue but it's not hugely consequential
I agree it's useful, and the wizard is fairly easy to predict so you can plan against it but, and I'm sure Hamlet can confirm/deny this, there are situations where the chain would be different if you happened to be on the other side of the map. I realise this is still predictable, but I think it's really inelegant and elegance is always high on my list of priorities!
Right, because on AP5 the jump the Shaman "plots" is from the Axe Thrower up to the Priestess (because up is favored over down) and then to the Impaler which is the closest target from there.
I agree that having these be totally deterministic is important so players can plan moves (also because you can't have random turn outcomes in HA). In fact, I might actually prefer chaining in its current from rather than being a little more optimal, because a more advanced algorithm would be more mental work to process when you were planning moves.
The only problem is the asymmetry. It comes up rarely, but should still be fixed.
There is asymmetry in that when firing straight upwards it will always go left instead of right or something like that. But having to click multiple targets would be just as inelegant
Interesting information. Disappointing that effort hasn't been put in to make the chain heal a "smart" chain (see games like World of Warcraft for precedent), but given the scale of the game it's not surprising.
It needed to be very smart in wow because there are many more targets and the range on jumps was something like 20 yards
See the appendix to my guide (sig) for exact chain rules. It doesn't actually favor lower-HP targets otherwise, but it does favor KO'd targets. The problem is that it computes the jumps sequentially; i.e. it determines the first target without reference to what the potential second jumps are. In order to do something more complex they'd have to iterate over all possible pairs of jumps.
My Hero Academy mechanics guide: http://iam.yellingontheinternet.com/2012/08/10/hero-academy-mechanics-gu...
I talk about games and many other things here: http://twitter.com/HamletEJ