Arena Rex Forums

Calling Walrus: I have more thoughts about the whole attack thing. I feel like this is a super complicated topic and this time, I think a thread just about attacks in general is warranted, rather than about a specific interaction. This is going to be a very long post.

I am sitting and reading the March 2017 FAQ document, and pondering some of the answers. Here are some examples.

Q: Do counter attacks or opportunity attacks trigger Hermes’s Aegis of Fame ability?
A: No. Neither counter attacks or opportunity attacks are considered attacks for the purposes of Aegis of Fame. Models declaring these reactions do not have to spend favor.

Q: Can [Spite] be elevated to a Power Attack?
A: No, it is not a declared attack action.

The rulebook says:

Attack
Cost: Gain 1 fatigue.
Effect: Target an enemy model within the engagement range of the attacking model. Resolve the attack as described in the Combat section.

Combat section:
To resolve an attack, the attacker rolls a number of six-sided dice (d6’s) equal to their ATK and the defender rolls a number of d6’s equal to their DEF.


It all seems relatively straightforward, but the picture starts becoming a little hazy when we start mixing two things into the soup: Reactions and special abilities.

I quoted the FAQ about two special abilities earlier, because I feel like these abilities - Aegis and Spite - are the prime offenders in terms of confusing what an attack consists of and is considered to be.

And then you have two reactions that contain attacks as part of their nature - opportunity attack and counterattack.

To have a ruleset that makes sense and is robust, it has to be able to explain how all these things work together.


I feel like the start of an explanation has to be able to define what the different phases are of the various ways that models can hit each other.


Phase one: Declaration.

This seems to be a central one, as best exemplified by RRGs answer that "Spite is not a declared attack action".

I feel like you can declare two things in the game: Actions and reactions. You declare a move action, and you declare an attack action. You also declare a counterattack reaction or a move reaction.

Apparently you do not declare a Spite attack.

The main thing that I think is strange is that it seems like the declaration does not contain a choosing of a target. Because Spite has to choose a target to work. But Spite doesn't declare. So targeting has to happen at a later stage. This is also supports Leo's Pounce ability, which allows him to declare an attack, then do a reposition move, and then choose a target later.

But this makes the Taunt ability kinda weird. Taunt says that you may only declare an attack against a model with Taunt. This sounds like you pick a target when you declare an attack. But as Spite shows, that cannot be - unless the targeting is a floating thing that you can do whenever you feel like.

Something else that is extremely important that happens in this declaration stage is the paying of costs. You gain fatigue here - and this means that it is also at this stage that you choose to upgrade to a power attack. So Spite does not cost fatigue and cannot be upgraded to a power attack because Spite has no declaration step.


Phase two: Resolving the attack success or failure.

Could use a better name for this. But it seems like this is where the actual "making" of the attack starts. As I understand it, Spite starts here. And both a declaration of an "attack action" and a declaration of an "opportunity attack reaction" lead to this phase.

The first part of this phase has to be choosing a target, I imagine.

This is where it gets a little weird. Until now, the attack has not targeted any specific model. But the game rules are full of references to "When a model declares an attack against..."

As I feel I have demonstrated, this phrase doesn't really make sense. Either that, or Spite doesn't make sense. Because Spite is an example of an attack that is "against" something, but without "declaring". So these two things have to be able to be separated.

Taunt can work by being interpreted as "When an attacker in range of this model chooses the target of an attack, confirm if the attack orginated from a declaration of an attack action. If it did, then it can only choose a model with Taunt as its target". And the same for Fade, the counterpart of Taunt. Then Spite and opportunity attack get around Taunt.

And the same for Aegis. Aegis has to be understood as "When an attacker chooses this model as the target of an attack, confirm if the attack orginated from a declaration of an attack action. If it did, then the attacker must pay one Favor, or the attack is cancelled". That's the only way I can see it working. Then Spite and opportunity attack get around Aegis.

But... what about counterattacks? Counterattack shares the "When a model declares an attack against..." wording.

Counterattacks clearly are triggered at this stage too. This would lead me to ask whether a model is supposed to be able to counterattack against Spite or an opportunity attack. However, 99% of the time this will not be relevant. Because you cannot counterattack during your own turn.

It can only become relevant if Ur-kek attacks an opponent that has some ability to do retaliation damage, like Frigge. Then a Frigge without fatigue can defeat Ur-kek on his own turn. Spite triggers, giving Ur-kek a second attack this activation. Is defending Frigge now able to counterattack against this Spite attack?

If I was writing the rules from scratch, I would allow it, but in the current writing of the rules, counterattack seems to follow the same wording as Taunt, and so I would say that no, she cannot, because Spite is not a "declared attack".

The second part must be spending Favor, both by attacker and defender. Note that the FAQ specifically says that Spite is allowed to do this. So it must be here rather than in the declaration phase.

The third part must be actually rolling the attack and defense dice.

The fourth part must be calculating whether the attack is a success or a failure.

Attack success = Attacker moves on to phase three.
Attack failure = Attack ends.
Attack failure if the defender counterattacked = Defender moves on to phase three.


Phase three: Resolving the attack damage.

Do damage and stuff. This is plenty complicated, but not the point of this thread. I haven't found anything particularly weird here.

...

So my conclusion here is, again, that I think a few things about the rules are a bit vague when you start looking at the specific details of specific interactions, at least if you expect consistency.

The main thing that I feel doesn't make sense: That Spite, the so-called "not a declared attack", shows that declaration of an attack action and the selecting of a target must be separate events, and therefore "declaring an attack against X" is not one event, but two events that happen in chronological order. Because Spite has a trigger which replaces the declaration, but does not replace the targeting.

...

Obviously you don't need to read a thread like this to play the game. This all falls under the classic GW adage that in a game of this type, not every combination of rules will make sense all the time - when you find questions that the rules can't answer, just roll a D6 for it or agree with your opponent what you wanna do.

But for people like me, who go "that isn't good enough - I wanna dig deeper and figure out a logically consistent way to resolve these questions using codified rules", it leads to this kind of analysis. Which is what I think happened with Magic, because it started being played for money. You end up with a "normal rulebook" of 30 pages or so, which covers all the elementary rules you need for playing the game, and then an "expanded rulebook", which will be 300+ pages, because it needs to work out how to resolve all the complex interactions that the basic principles just aren't detailed enough to address.

I'm posting this mainly as a note to myself about how I interpret the intent behind the rules, but also in case anyone wants to throw out an alternative interpretation.

For example, you might say that Spite doesn't "declare" an attack, it "makes" an attack, and therefore it has a "making" phase, and the targeting happens there, or something...

_________________
My Arena Rex painting log

I think the Declaration aspect relates to the paying of the “Cost”. Usually gaining a level of fatigue. However, Spite does not have to pay this cost, as it skips straight to rolling and resolving.

Counterattack and opportunity attack, aren’t declared as “attacks”; rather they are unique declarations, and only become “attacks” upon successful dice resolution. Therefore, they get around being an attack for the purposes of Aegis of Fame, due to the timing window.

Simple Hypothetical: Hermes (active model) declares attack action against Hagal (ready, reactive model), and gains 1 fatigue > Hagal declares counterattack reaction against Hermes, and gains 1 fatigue > Dice roll of ATK vs ATK > Roll results in 1 net success for Hagal > Counterattack, is now treated as a successful attack, in addition to a declared reaction > apply net success to (now) attacking model’s damage tree > -2 for Hermes ARM > Hermes ticks off 1 box from his Vitality bar.

Because the counter attack (or opportunity attack) isn’t treated as a successful attack, and Hagal isn’t considered an “attacking model” until after the resolution of the dice roll, it gets around the timing trigger of Aegis of Fame, which is prior to the rolling of dice. This is why Spite doesn’t declare, (so gets around Taunt and doesn’t have a Fatigue cost), but is still an “attack” prior to the dice roll, so Aegis of Fame applies, and the Favor must be paid.

I would say no, you can’t declare a counterattack. The trigger for counterattack specifically states the opposing model must declare. Spite has no declaration.

Ok, I think I have better ideas for terminology. I would say that overall, any event in the game that will result in attack dice being rolled can be divided into an "initiation phase" and a "resolution phase".

"Initiation" would be the declaration, phase one, in case of an attack or counterattack. For Spite, it would be a trigger instead of a declaration.
"Resolution" would be all the rest, consisting of two separate sub-phases: "resolution of combat", phase two, and "resolution of damage", phase three.

Although I am thinking that if I were to do a proper flowchart thing, selecting a target would have to be its own sub-phase. Because which target you select determines whether or not it is possible to declare a counterattack reaction, and a counterattack declaration leads to a different type of combat resolution than normal combat.

_________________
My Arena Rex painting log

I disagree with treating counterattack and opportunity attack the same here. They resolve completely differently.

As you said yourself in another thread, the whole section of how to roll ATK vs DEF is in a section called "This is how to resolve an attack". Therefore, opportunity attack must be an attack. Otherwise there are no rules in the rulebook for how to do opportunity attacks.



I agree with your conclusions about counterattack and Spite, but not about opportunity attack, as outlined above. Opportunity attack resolves 100% like Spite. They both just resolve a normal attack. So if Spite has to pay, then opportunity attack does too.

Edit: Or to put it differently... if opportunity attack doesn't have to pay, because it isn't "considered an attack for the purposes of Aegis" as the FAQ entry says, then that seems to me to be a "just because" ruling that isn't consistent with anything, since it doesn't explain *why* it is not considered an attack.

And I feel quite certain that if anyone were to ask if Spite has to pay for Aegis - I wonder if anyone ever has? - that they would get the answer that it doesn't, for the same reason.

_________________
My Arena Rex painting log

Yeah, targeting can't be part of the combat resolution because opportunity attack targets differently, but combats normally. So might as well have four phases.

1. Initiation phase
2. Targeting phase
3. Combat phase
4. Damage phase

Where the only one that is resolved the same regardless of origin or variables along the way is the damage phase.

_________________
My Arena Rex painting log

I still am of the opinion that it is not an "attack" until the criteria is met. Until the point that a dice roll resolves with net successes, something is not a "successful attack". so I.E. it is specifically an opportunity attack reaction (note, not an attack). Once the criteria is met, it then becomes a "successful attack" as well.

The rules for opportunity attack expressly tell you that it modifies standard combat procedure (+1 bonus success). That still doesn't make it a "successful attack" until the dice roll resolves with net successes.

No it does not. It resolves as ATK +1 bonus success vs opposing Def. This is expressly different than standard combat. In addition, opportunity attack costs 1 fatigue to declare, unlike Spite, which has no fatigue cost.


Also, don't forget that Walker has also expressly stated that opportunity attack and counterattack do not apply to Aegis of Fame. This sets a strong precedent to interpret these as "not attacks until..." during the relevant timing windows.

Hot damn. You are totally right, I forgot about that entirely - both about the +1 and about the fact that it does in fact refer to the ATK vs DEF in the opportunity attack rules separately. I stand corrected.

That does change things. I think I agree with you now. So basically there are three different kinds of "combat types" - standard combat (ATK vs DEF), opportunity combat (ATK+1 vs DEF), and opposed combat (ATK vs ATK). Each associated with one type of declaration.

So then Aegis is meant to trigger off... selecting Hermes as a target in the "standard targeting phase", which is something neither of the reactions go through, since neither of them have freedom to choose a target. But both a standard attack and Spite would.

Alright, that works. Thanks for the help! I'll have to try and make a flow chart tomorrow...

_________________
My Arena Rex painting log

I like this characterization of "combat types". Makes it a bit easier to wrap your head around 3 different things all sharing the same word.


Yes, I agree. I think the "targeting" aspect is the key here.


Good deal, I'm looking forward to see what you come up with for the flow chart.

Me too. It makes sense, really. Both of the reactions don't have a normal "choose target" step - they automatically target the model that triggered their reaction. So it makes sense that the effects that are related to whether or not you are able attack them - Aegis, Taunt, Fade - aren't envisioned as being relevant here. Because they want to mess with the choice, and there is no choice to mess with.

_________________
My Arena Rex painting log

OK, here is the flow chart with pretty colours and all. Version 1.0.



The idea is that when the rules refer to "When you declare an attack against...", they are referring to what I here call the "Initiation phase" as a whole, marked as the green outline on the top. This phase consists of two steps - the Declare/Trigger step, and the Target step.

During Declare Action, you pay fatigue costs, and therefore you also upgrade to power attack. This also explains why Power Attack is listed as its own action in the rulebook - it's more like an alternative attack declaration than an upgrade. And only available in the box marked as "Declare attack action".

Noteworthy: Leo's Pounce triggers after Declare, and resolves before Target.

During Choose Target, you choose the target. This is where Aegis, Taunt and Fade kick in. Since Spite has to go through this "Choose target" step, Walrus is right that Spite logically has to pay the cost for Aegis, even though it feels weird to me. The reactions do not go through this step.

If we want Spite to not have to pay for Aegis, Spite would have to have its own "Choose Spite target" step that would be different from "Choose attack target". But I don't see any basis for such a reading as I read the rules.

(You could also just remove the two Target boxes after the reactions, and argue that the reactions skip the Targeting step entirely.)

Once you have Declared/Triggered and Targeted, the Initiation phase is over, and you move on to the Resolution phase, marked as the blue outline at the bottom. Which consists of two steps: Roll combat, and Apply damage.

As Walrus pointed out, there are three different types of combat that roll dice in their own way. All three types of combat include a pre-rolling step though, which is Use Favor, where both attacker and defender get to add Favor dice. Once the dice have been rolled, there are always two possible outcomes: No Success, or Success.

Normally, only the attacker has the option of being successful, but unique to the Counterattack Combat Roll, in this roll the defender can be successful too.

If the result is No Success, then the flowchart ends. If the result is Success, for either attacker or defender, we immediately move on to the grand finale, the shared step of Apply Successful Attack Damage.

As I understand it here, any time the rules say "Immediately after a successful attack...", they refer to a timing point at the start of the Apply Damage step, before damage tree boxes are selected. (Edit: Basically during or immediately prior to step 1 of the damage application sequence I posted in the following post.)

If I am correct in this understanding, then Sereqet would be able to use Envenom any time after she has made a successful combat roll, regardless of whether her successes originated from an attack, a counterattack or an opportunity attack. This feels right to me, personally.

Other types of attack, which aren't included in the flowchart, but could be, would be Wild attacks and Living Hazard attacks.

Does this all make sense to you, Walrus?

_________________
My Arena Rex painting log