[Feedback] Action and Reaction

Discussion in 'Feedback and Suggestions' started by YoYoTheAssyrian, Feb 6, 2014.

  1. YoYoTheAssyrian

    YoYoTheAssyrian Mushroom Warrior

    I would like to start with an interesting statistic. In Cardhunter going second gives you a statistically significant chance of winning. Now I’m not sure what the exact number is currently but player 2 wins about 53 to 54 percent of the time. Not a huge difference percentage wise, but when you consider that the database is tens of thousands of game results, it becomes apparent that card hunter favors going second. Now this may be because the AI bots are often reported as player one, but let’s examine some of the gameplay and design reasons why this happens.

    First of all there’s the obvious, reacting is better than acting in cardhunter. One of the most basic examples is moving into base to base. If you move a character base to base with an opponent’s character, the acting player loses all initiative. The reactor can choose from any number of options, running away, changing facing, attacking first, but needless to say, it is inherently risky and far less effective to move to a character rather than letting him come to you. In many other tactical games, this is not a problem, generally due to it being either impossible to run away from melee combat (Warhammer and 40k) or you incur significant penalties in the form of an attack when you disengage (D&D, War Machine) This has the combined effect in these alternate games of making combat “sticky.” It can be difficult at times to get into melee combat (for whatever reason) but once you get into base to base combat, someone is taking some damage. In Cardhunter, we do not have this, combat is the opposite of sticky, call it slippery after the super awesome elf trait. There is no penalty for running away from combat, and often it is the best possible decision. Slippery melee is one of the prime reasons for Reaction based builds and play-styles.

    Another aspect of reacting being better than acting is how it affects ranged combat as well. Consider a scenario; ranged attacker has a move and ranged attack, melee attacker has a move and melee attack. Let’s also assume there’s a wall in between them. If melee attacker moves around the wall, then ranged attacker who is now reacting will run away, and then shoot. It’s a complete failure for melee attacker, who now has to wait until he gets another move to try again. And unless he manages to pin ranged attacker in a corner, the cycle will continue forever, melee tries to move up and attack, ranged runs away and shoots him. Now consider if ranged attacker moves first. The scenario again switches to favor the reactor, melee attacker can run away around the wall, or he can move to engage and successfully get off his melee attack assuming he survives the ranged attack.

    This truism applies to every level of basic (moving and attacking) Cardhunter play. In almost every situation, moving second is better. Reacting is better, just inherently through the quirks of the system. So when considering how best to balance the game, and further how best to nerf or buff aspects of the game we think are out of whack, this basic premise of the system must be kept in mind.

    But wait! You say, this is Cardhunter at its most basic and abstract, this doesn’t apply when you’re using card (x)! And you’re right! This dynamic doesn’t apply to certain specific cards, but one of the few connectors across all those thousands of games is the basic system and how it rewards decisions.

    And now we get to the meat, considering the inherently defensive nature of Cardhunter play, how does this take effect when playing with all the tools and tricks available? Specifically, how does this interact with the top 3 complained about cards, Whirlwind Enemies, Winds of War, and Nimble strike? First of all, two of them are wizard cards that add to defensive play. Now WWE and WoW both have offensive uses, but mainly they are used to play keep away. Try to move up? WoW puts you right back where you started. Base to base with my wizard? Time put you anywhere else. Just seized a victory point? Not any longer, now you’re off of it. The primary effect of wizard positional control cards is to shift the balance even further towards the defensive. Not only do you have the basic advantage of slippery melee, but now you have a whole host of options that enables you to keep that basic advantage and build upon it.

    The prevalence and commonness of Wizard positional abilities in MP has profound consequences. First of all, it makes range one no step attacks secondary in importance. Delivering these attacks becomes contingent on whole host of factors, which often aren’t in the using player’s toolbox, or under their control. Any player will tell you that the attacks that get discarded at the end of a turn are the ones that don’t have movement component. Unless you can deliver them, which due to basic systems and control abilities is a lot harder than it sounds, they might as well not even be in your hand. As such we see many people engaging in a circular argument. “I only use control abilities because step attacks make it harder to keep people away from me!” “Well I only use step attacks because your positional cards make it impossible to deliver anything else!” This is a classic chicken and egg argument with no solution, especially as it depends on anecdotal evidence to support either side. Whatever the tactical situation, reacting gives a serious gameplay advantage. What benefits the reactor more? Step Attacks? Or control abilities?

    Now nimble strike, and step attacks in general, subvert this system of defensiveness, and further, they are the only cards to do so (with a nod to things like cantrips, but that takes a two card combo). By combining a move and an attack, quite suddenly the reacting player is on the defensive. He can run away, but what if the attacker simply has another? You’ll be chased across the board taking damage and the inherent tactical decision will not have changed, the reactor will still be faced with an attacker base to base with them who has already accomplished dealing some damage. Since step attacks favor the offense, they have inspired quite a lot of hate in certain sections of the Cardhunter community. After all they are subverting the basic design decisions behind standard range one melee attacks. And the relationship between being moved to and moving to.

    Step attacks can also allow a person to attack from behind, thus preventing the defender from utilizing their blocks. This is one of the secondary issues people have with step attacks, but often, and in a vacuum, one can simply use positioning on the map and positioning within your party to prevent opposing characters from either reaching or getting behind people. The problem comes in when a build both counters your ability to position and utilizes step attacks, PaladinGP’s build being the prime example. Now some argue that the problem is nimble strike, and fine they’re entitled to. But the prime ability of that list isn’t that it runs NS, as many, many, many people do and are far far less successful. It’s that it prevents you from positioning yourself to receive incoming NS attacks. If a NS hits you from one of 6 front arc squares and you have a parry or any card that that you can use next action, then fine, it might be a bit unexpected, but you have used your counter card and prevented damage. Also, you have your other two character in support positions, either denying movement to your opponent, or providing additional oomph to your counter-attack. Most of the time, it is very hard to do anything else than hit them in the front. UNLESS, you have cards of your own that allow to either deliberately or randomly move your opponent’s pieces. Or if a player has exposed themselves, like by say grabbing a victory point. If you remove yourself from the protection of terrain and your party, then you deserve what’s coming to you. But really that’s just another example of how using a move card to move towards the opponent gives all the initiative and options to the reactor.


    So what’s the solution? Well that’s why this is a forum post, let’s get our debate on. But two things to keep in mind. Any nerf to step attacks will be a boost to positional control abilities, do we want that? Do they need that? And Second, any nerf to positional control is a buff to attacking, step attacks included, do we want or need that as well? One thing I’m sure we all agree on is that one of the things balance changes should accomplish is bringing the dynamics of turn order and victory closer together so they are no longer significant.
     
    amoth likes this.
  2. neoncat

    neoncat Feline Outline

    Just a point of information: the AI is always listed as Player 1 and it also tends to lose, which should skew the Player 1 / Player 2 win ratios in favor of Player 2.
     
    Bearson Onyx likes this.
  3. Jarmo

    Jarmo Snow Griffin

    You seem to base a lot of your argument on the MP map player 1/player 2 win percentages from Farbs' site. Well, did you know this about those?
    "I'll just add for now though that it seems AI players are always reported as player 1 by the API, and they tend to lose."

    I haven't done the math to try to see how much this influences the win percentages, but in any case it's good to keep in mind in relation to the subject of this thread.
     
  4. YoYoTheAssyrian

    YoYoTheAssyrian Mushroom Warrior

    Fixed, the 53% was intended as a jumping off point to a wider discussion, so let's have it.
     
    neoncat likes this.
  5. Jade303

    Jade303 Thaumaturge

    TL;DR. *EDIT* I meant to just put TL. I actually did read it but I am not going to talk about the body of your discussion.
    While there is no sense arguing that reacting is better than acting, part of Card Hunter's strategy revolves around priority (acting first) and forcing your opponent to react in a way so that you can react better.... and all of this acting/reacting happens fast and flips from player to player often.
    Without changing the integral structure of how the game is played, there will always be a first (or second) move advantage. For all we know the overwhelming number of computer losses may be obscuring the fact that human player 1's win more matches than human player 2's.
    I don't believe that it matters enough to warrant changing anything.
     
  6. YoYoTheAssyrian

    YoYoTheAssyrian Mushroom Warrior

    It's an interesting point, but again, you obviously did not read what I wrote. Priority is something I'm planning on talking about when I discuss offensive strategies in more depth, or rather, how a lot of cardhunter players despise offensive strategies and refuse to play them.
     
  7. Like Jarmo already said, it would be first useful to determine what the actual advantage is. I mean there is no point is discussing about the fix unless we first fully determine the problem. AI plays tons of matches and loses often, which will probably have some kind of an effect, although it shouldn't be as much as 3%.

    I think that one of the major issues in Card Hunter is what you talk about when moving next to your target. I think it's silly that if we have a warrior with tons of 1-range damage but only one movement card, that warrior will never catch a fleeing wizard. Either the warrior moves next to the wizard and the wizard just steps back, or the warrior stands there like a moron getting hit with all kinds of spells.

    I wonder if it would have made more sense if Card Hunter was designed so that all basic movement cards were cantrip, but only for the character using them. For example I could move and hit with the same character during one round, but I couldn't chain cantrips so that I would move every character I have, one after another.
     
  8. YoYoTheAssyrian

    YoYoTheAssyrian Mushroom Warrior


    It's an interesting idea, but then my first thought was, I'm going to run all elf warriors and spam obliterating bludgeon. But in most all other tactical games moving and attacking are often done in just the way you describe. Activate one person, they move and attack, cast a spell, shoot, what have you, and then we move onto the next person. I can think of many exceptions to this, for example, Necromunda (great game btw) has weird melee combat that is super risky where even if you charge someone they can be the one who kills you, but it uses comparative dice rolls instead of igougo.

    Another interesting system is the blitz! mechanic from blood bowl. In that game, you cannot attack unless you start the turn b2b with another model. The exception is that once per turn you get a blitz action, where you can move and then throw a block. But only once per turn and with one model. I could see something like that working, because the information would be visible to both players, and both would use it in their decision making.
     
  9. Yes, well if something like that was implemented they would have to redesign a lot of cards. Many cards like encumber ones would become very powerful if you could always attach a cantrip movement to them.

    It is interesting to discuss about changes like these, but realistically speaking they will never happen, so it's kind of pointless :)
     
    Flaxative likes this.
  10. YoYoTheAssyrian

    YoYoTheAssyrian Mushroom Warrior


    I'm not expecting to change the devs nerf/buff policy, mainly this is about defining some aspects of cardhunter that are widely known by those who really get into the nitty gritty of the game. When we advocate to changes to the meta, whatever the cards, we have to keep in mind the underlying system. Currently the system favors defense and reacting. After all the devs have made it quite clear that we are no more and no less than a think tank (they read us, but keep out of our discussions). So I wrote some stuff we should all think about :-D.
     
  11. Farbs

    Farbs Blue Manchu Staff Member

    I just ran through data from 562,404 multiplayer battles that didn't involve AI players, and player 1 (as reported in this particular format) won 50.5% of the time.
    What I don't know right now is whether in a ranked match this format shows the player who plays first as player 1, or as player 2. These are a little muddled in places.
    What we can see from this however is that the AI does account for a large number of battles and a substantial portion of the bias apparent in the [META] figures.
     
    Megadestructo, Bearson Onyx and Jarmo like this.
  12. Jarmo

    Jarmo Snow Griffin

    Thanks for the info, Farbs! That's very good to know. Please let us know if you find out who is player 1 and who is player 2.
     

  13. Forgive my ignorance but is it possible that the tool is tracking player 1 as the higher rated opponent going into match and player 2 as the lower rated?
     
  14. Jade303

    Jade303 Thaumaturge

    The player with the highest ranking isn't necessarily going first (and player 1 goes first, I believe)
     
  15. neoncat

    neoncat Feline Outline

    This may help clarify the player number assignment:

     
  16. Kalin

    Kalin Begat G'zok

    Could we please, Please, PLEASE have some consistency in how players are numbered?
     
    neoncat likes this.
  17. neoncat

    neoncat Feline Outline

     
  18. amoth

    amoth Kobold

    Interesting piece of analysis that's made me think more about how i play and notice the strategies i use to cope with slippery combat. *

    It seems to me that either Card Hunter could adjust a lot of items to recognise the great power of cards that shake up positional play (i.e. make them rarer or costlier to play), or add more cards that further shake up the dynamics and open up more strategies.

    If reactive moves (like Dodge and Reflexive Teleport) were more common, that would be another path to countering step attacks. You could even have ones that only trigger when targeted by step attacks.

    The other obvious class of cards that are completely missing are reactive attacks - i.e. attacks that trigger when an enemy moves.

    Would that weaken the current positional super cards enough that 'normal' cards rise again in relative importance, or is this just a spiral of escalation that makes normal cards ever less relevant?

    *(It might help the discussion to set out some of the positional strategies, even though it's unlikely i'm teaching anyone anything here:

    I currently run a 3 elf priest team in MP who have to get into contact to deal damage. The main strategy is to just arrange to have more move cards than my opponent. That requires preparation that stretches over several rounds or luck with drawing extra moves. I often have to run my victim out of moves by chasing them with an elf who doesn't actually hold any attack cards before the one who does nips in for the kill. Moving to a spot that threatens two enemy usually means that only one can slip away. I also move to block escape routes, pin enemy against walls and exploit the enemy's need to get to the victory squares.)
     

Share This Page