User talk:DanielDraco/On Balance

Flip of a Coin

Completely agree with that section, and it's one of (if the not the) biggest reasons I abhor wizard-level combat. --Ghostwheel 02:19, 9 November 2011 (UTC)

The advantage of coinflip combats is that when every move has a high chance of killing, there is an immense tactical advantage to finagling one more action. Rocket tag may make the actual combats unsatisfying, but the precombat setup becomes more worthwhile if getting yourself a surprise round and your allies into position means almost certain victory. Since actual RPG combat frequently suffers from being a simple, yet long-winded comparison of stats, I'm not sure coin-flip rocket tag combat is actually a bad thing. --Foxwarrior 04:15, 9 November 2011 (UTC)

That's more of a problem with the base system (I full attack/move-attack/charge. Again.) rather than the principle of tactical combat. If you have different options that are viable and more-or-less powerful in different situations, you get combat that's a lot more tactical, dynamic and interesting, especially if you incorporate into it the pendulum of power.

Coinflip combat can also mean that if the baddies EVER get the jump on the PCs, it can result in a TPK, which is rarely/never the desired outcome. That said, by having combat be fairly short (4-8 rounds, I find, though the exact numbers can change), a surprise round can weigh heavily into the result of a combat. Just not be the only deciding factor of who wins and by what margin. --Ghostwheel 16:23, 9 November 2011 (UTC)

Fox has the right idea- coin flip means more tactical combat. What Ghost is describing is not tactical at all, but a seige. You are stuck in a seige by definition, because both you and your foes are entrenched in the combat and the only way out is to out last everyone. I personally hate those combats, and they are the reason I hate 4e. Focusing on the players end, if they are the ones planning the attack, the reward for good planning should be a short, easy victory with minimal losses. If they are the recieving end, they are already losing and need to turn the fight around or run. It sucks, but losing is supposed to suck. If there are no risks, the game is not worth playing.--Change=Chaos. Period. SC 19:32, 9 November 2011 (UTC)

Oh, I'm not saying that Ghost is wrong about tactical combat; it's certainly possible for longer combats to be tactically interesting, and I've occasionally found such in D&D at Rogue-level. I'm just saying that it's not the only fun way to do combats. --Foxwarrior 20:08, 9 November 2011 (UTC)

My mistake, but the point still stands.--Change=Chaos. Period. SC 20:20, 9 November 2011 (UTC)

A major issue I have with rocket tag, mechanically, is that not every character role is capable of it. That is to say, only spellcasters are capable of it. Now there's the occasional UMD-happy rogue who will spend his time pretending to be a wizard, and the occasional martial adept who minmaxes the hell out of a preposterous insta-win combo, but short of that sort of cheese, rocket tag is closed to non-spellcasters. If every type of character could participate, then I'd be okay with it.

But even then, it takes all serious planning out of it. A rocket tag plan such as, "Alright, you go in first and blind them. Then he'll go in and paralyze them. Then I'll use Wail of the Banshee," is really just a specific form of, "You make them save or lose. Then he'll make them save or lose. Then I'll make them save or lose." You can choose any arbitrary rocket and then go play tag. Beyond that it's nothing but trying to prevent specific obstacle like invisibility or a wall of force -- and that's just a simple question of whether or not you have the spell to deal with it. There's no real strategy, just re-skinning your weapon.

All that aside, though, the argument I make on the page isn't anything to do with tactics -- it's a matter of drama and story, and I think that's where it presents the biggest problems. Let me make something up to see if I can demonstrate my point.

DM: "You continue down the dark corridor, feeling your way along the walls so you don't run into them with each serpentine twist. As you round a corner, you see warm red light spilling into the hallway, and, once you've reached the next bend, you see the source. A behemoth, once a man but corrupted in form by science and magic, sits in an ornate chair, lounging boredly and reading a book; beneath his skin, red-hot veins pulse visibly, glowing bright enough to illuminate the entire room. His jaundiced and bloodshot eyes glance sharply up at you the moment you appear in the doorway, and he gives a slow grin. Rising to his full height, towering a head and a half above the tallest of you, the beast cracks his knuckles and, with a rumbling voice that is nearly sickening to hear for what little humanity is left in its timbre, he murmurs, 'This should be fun.'"

Player: "I cast Blindness."

DM: "...fuck."

--DanielDraco 23:14, 9 November 2011 (UTC)

Rocket tag is pretty overstated in my experience, at least once it's separated from the "15 minute workday" and "focus fire" problems. That blindness in your example is pretty hardcore (despite being single target), but it's not unlimited in use. It's also filling slots that need to be budgeted for other effects when it first becomes available, and later when you have more slots to spare it's falling behind on the save DC because you are higher level and your foes have higher save bonuses. In situations where you are 1) on the clock, and expect to have more encounters than you can reasonably assign hardcore high level combat spells before recovering them and 2) are fighting more foes in an encounter than you can target with any single hardcore high level combat spells you don't have these encounter ending problems in as meaningful ways. Yes, they cast this spell and deal a lot of damage to the encounter, but then they proceed to under-perform or use other resources to keep up with other characters for the rest of the encounter or in other encounters. And it's not like the spell is an actual auto-win at low levels, the enemy has to actually fail their save and this is not guaranteed. It admittedly breaks down at higher levels due to caster DC boosting and high levels of save divergence (so you can target weak saves with non-top level spells), but it actually does work okay to mix rogue and wizard and even fighter level characters in lower levels of the game under certain conditions.

What this actually means is that single enemy encounters and short encounter days are not particularly compatible with low-mid level wizard balance point games in which anyone else is supposed to contribute. The system just doesn't support them well because of the prevalence and ease of use for single point of failure attacks. Any time these become more costly or harder to pull off and the disparity is much reduced. Plenty of people want single foe fights and short encounter days (see: overland encounter tables) though, because it's iconic or cool, and putting those in is a good reason to either pull wizard level characters out or restrict the game to just them to prevent overshadowing. - Tarkisflux ^Talk 00:51, 10 November 2011 (UTC)

Even if they can't play rocket tag all day, they can still play it for a couple encounters per day. The very fact that they can completely nullify any given encounter within one or two actions is the problem. Yes, sometimes their I Win spells will be used up before they get to the boss fight. But the odds are that nearly all of the combat spells they have on hand are either insta-win or battlefield control -- after all, why would you prepare anything else? They might have some debuffs thrown in there -- mostly meant to soften up tough foes for the insta-wins. So if they're depleted of spells that will let them play rocket tag by the time they face the boss...well, at that point they're pretty much entirely out of helpful spells and will be cowering in the corner for the whole fight. Or if they saved their useful spells for the boss fight, then they play rocket tag after all and ruin the whole thing. --DanielDraco 02:54, 10 November 2011 (UTC)

Given the way you've defined balanced and fair and the recent clarity added to your blog/post/rant/thingy, there's not much I have to add. At some point you have to make a somewhat arbitrary design decision to allow certain types of tradeoffs and to support certain themes, and deciding that measured bursts of single point of failure effects are unacceptable because they actively undermine themes you care about is a perfectly reasonable design position to take. I'm glad I have a better sense of the types of games you want to play, and have a better sense of where you're coming from in the future. - Tarkisflux ^Talk 06:16, 10 November 2011 (UTC)

Blargh, edit conflict. What I wrote:

First SC--you're wrong. What you're referring to isn't "tactics" at all. It's logistics or strategy. But tactics is the actual tactical movement and use of abilities in combat. As DD mentioned above, coinflip combat takes all the tactics out of it. It's just "tag, you're dead". I explained what I mean by the "pendulum of power" here.

Also, Tark, I find that most spellcasters of a decent level save all their slots for combat, and then have scrolls for utility without giving up much/any combat potential (and let's not get into the 15-minute workday). Plus, with any properly-equipped spellcaster is going to have DCs that outstrip the saves of most monsters of their CR. Furthermore, at higher levels you use lower-level slots either for utility, or use save-less spells (grease for example, letting the rogue freely SA, or haste to buff your attackers to ungodly levels, etc).

Strongly agree with DD's last blurb too. --Ghostwheel 06:22, 10 November 2011 (UTC)

Tactics are the plans one uses to enact a strategy. Individual movements and abilities are assets or actions, not tactics. If I were to just punch you in the balls, it would not be a tactic, it's an attack. If the strategic goal was to punch you in the balls, the tactics would be the plans to set up the achievement of the goal, like paying Felicia Day to walk past you and wink, causing you to present a more accessable target area, then sneaking up behind you and getting you to turn around, allowing me to punch you in the balls. If stratedgy is the goal and outline, tactics are the subpoints and actions are the fine details.

What does this mean in context? Rocket tag is the supreme tactic for combat if your goal is to end your target with no losses. If I were to kill a dragon, I would want to have it boil down to a few actions as possible. Dramatic combat is stupid combat, and thus I do not support it.

Also, in regards to the link, as a player and a GM I call bullshit. I've been in the exact same combats you described, and I feel the same way every time: trapped and cheated. Having near loss forced upon you as an empty threat is neither rewarding or encouraging- it is hollow and fake. Players I've dealt with like their victories real and earned, be it through trickery or strategy. On the flip side of the coin, loss stings and generally sucks, but it is supposed to. Survivors plot vengence, the dead seek to do better in new incarnations, but either way it leads to great stories.--Change=Chaos. Period. SC 06:54, 10 November 2011 (UTC)

Rule 14. --Ghostwheel 07:20, 10 November 2011 (UTC)

Ghost, I'm not interested in arguing the point here but nothing you wrote actually contradicts anything I wrote, and if you think it does then I have apparently failed to write clearly. But that's neither here nor there, since I'm actively removing myself from the conversation at this point. DD's piece isn't an argument for or against a particular balance point in general, it's an explanation of his playstyle preferences and the conclusion that wizard level content doesn't support those things. And he's right. When it was less complete before it looked like he was overlooking something, but he added additional details and indicated that some forms of balance weren't forms he was interested in allowing because they fail to support narrative structures he wants in play. Since I'm the only person on the wiki I can think of who actively endorses fighter, rogue, and wizard level play (though my own actual play preference is often, but not exclusively, wizard) and I agree with his reasons and justifications for choosing to exclude wizard level material, there's nothing here for me to argue against. He's holding a perfectly legitimate position as far as I'm concerned and he's holding it for accurate reasons, and I don't have anything constructive to add at that point. - Tarkisflux ^Talk 07:50, 10 November 2011 (UTC)

Alright, so rocket tag is a tactic. Whatever. You've won the debate on semantics. But within your argument, you've highlighted what (to me at least) is the major issue in the debate we were originally having. "Rocket tag is the supreme tactic for combat if your goal is to end your target with no losses". Well yes, that's exactly the problem with it.

Premise: Players use rocket tag

• Option 1: Foes also use rocket tag

• Result: Very little tactics/strategy beyond "keep launching rockets"; no drama; possible TPK; etc.

• Option 2: Foes do not use rocket tag

• Result: Party wins. No challenge is presented.

I don't like either of the possible results, so I'm not going to accept the premise.

And as for dramatic combat being stupid...why are you playing an RPG, then? The whole point of RP is the story. Besides, it's not as if I'm asking players to artificially choose inferior tactics so that it's dramatic. That would be stupid. The drama is found in a legitimate struggle -- and it's the DM's job to limit the players' options so that a legitimate struggle is possible. Sure, if the players are exceedingly creative and they prepare, they can still often win without even entering initiative. Personally, as a player, I've always found those strategies that took fifteen minutes to prepare much, much more satisfying than the wizard saying, "Hey, guys, I can just blind it." You put more effort into it so that when it works, you feel like a clever badass, not a bored god. --DanielDraco 15:48, 10 November 2011 (UTC)

Point of Order - Game Testing

When you say that people who are in favor of the SGT are committing the logical error of conflating single combat and team combat you are flat wrong. The SGT completely ignores team combat, on the premise that testing that sort of thing is actually really hard. It requires substantially more tests to make sure that the various group configurations work and the game doesn't break as soon as you start trying to play it in a different way, and often occludes actual failures of classes because other characters are covering for them. To avoid those issues and simplify things, the SGT just measures whether or not a class can keep up in combat on its own. And the SGT doesn't make any attempt to hide that bias, and its failures for support/utility classes are known and admitted. It's often written off as a non-failing since those classes are "boring to play" (an assessment I do not agree with), but it's an admitted failure when it comes up. And while this can lead to unexpected synergy with other classes and over performance issues, it does a reasonable job of putting a floor on actual single character contributions to the party regardless of party configuration. And for high rogue and wizard level games, it is more important to know this than to know the likely power ceiling (which is more likely to be measured with actual party testing).

Additionally, your suggestion that the only way to know an outcome is to run an actual test isn't wrong, but it's extremely misleading. If you want to actually know how something works overall, you need to run LOTS of tests on the same configuration to weed uncommon/unlikely but completely possible probability events. You can run a party of monks against a balor and win if the characters are properly equipped and the balor only rolls 1s and the players only roll 20s. And with what you've written that should be a perfectly acceptable outcome that proves that they're doing ok, but it's actually an outlier data point that would disappear completely if multiple tests were run. And you need to do multiple runs with every actual combat to weed these sorts of things out. Class balance is a matter of statistics, not of one-off events, and for that reason a generic thought experiment is vastly superior to an actual test in terms of time and energy expended (unless you're running a simulation on a computer or something). - Tarkisflux ^Talk 18:53, 16 November 2011 (UTC)

In support of that point, I'd also like to point out that the generic thought experiment also results in sdomething a lot closer to what would happen when YOUR group plays. The problem I find with the SGT is that it ignores terrain for the most part, which is essential to my campaigns and fighting style, thus I don't use the SGT, as it is written for rather generic combat. The thought experiment when done by multiple minds also has a higher tendancy to reveal falacies. One of the above examples that prooved the article worse than useless to me was the use of Blindness on what I'm guessing was a giant, which supposedly ends the fight. As a fighter, bullshit. While true in some games out there, not true in mine. Blind something, and it will thrash out, possibly with severe consequences. The written encounter was placed in a library, with has things like shelves and book, which make wonderful debris, weapons and even cover/traps when weilded by the blind.

In conclusion, I agree with Tarkis that you need multiple tests to access any rule, be it a class or spell, but more importantly the tests must be relevant to how you play.--Change=Chaos. Period. SC 20:48, 16 November 2011 (UTC)

Alright, so maybe the SGT acknowledges that it doesn't perfectly measure how a class performs in an actual game. (Although, if it weren't working under the presumption that the results of single combat can be extrapolated to team combat it wouldn't exist, so I'm not "flat wrong" about that.) My point is that it doesn't even measure it enough that the SGT is worthwhile. I'll clarify my wording later.

Did I say that the results of one test would objectively demonstrate the balance of a class? I'm fairly certain I didn't. I didn't intend to, anyway. I didn't even say that the side that wins is the side that's more powerful. The playtesting that I proposed is for the sake of getting a feel for the power, because takng a look at a class and thinking about how it's supposed to work rarely yields the same results as actually trying it. It's a subjective measure, and I'm not hiding that -- it's essentially impossible to get an objective measure of balance that actually reflects what you want it to reflect. I proposed amount of wasted time as a possible one that equally measures flat damage and things such as debuffs, but even that isn't really ideal. The only way to measure everything all at once is to try it out and subjectively examine how it goes. Yes, it introduces the issues of outliers, but you can generally tell when you're getting shitty rolls or good ones; within a few tests, you should have a good feel for how much each class is going to contribute.

Yes, there are some variables that are ignored, like terrain. You're never going to get a perfectly complete picture of how a class plays until you actually play it in a campaign. I think I'll emphasize that on the page somewhere. I might just remove the section on pre-game testing, because there might not be any worthwhile way of doing it. Whole-party testing is much too complicated when you're trying to figure out the power of a single class, and balancing for the SGT just leads you to balance every character for single combat so that team combat becomes nothing but a cluster of duels.

Blindness is pretty much an insta-win by standard rules against most enemies. If it isn't so in your games, you're probably using house rules or an unusual assortment of enemies.

Pointing out potential flaws in my reasoning is fine, but Jesus Christ, people, don't go for the jugular when it's still in alpha stages. Just because I put it on the page, that doesn't mean I'm entirely happy with it and it doesn't mean I'm going to keep it. That idea for how to playtest was a new thought I was toying around with and fleshing out to see if anything useful would become of it. These are notes -- they're going to be unclear and they're going to lead to some dead ends. When they're no longer just notes, I'll move it out of my userpage.--DanielDraco 01:13, 17 November 2011 (UTC)

Not trying to beat you up about things, and I apologize if it came across that way. I thought this was more of a blog/rant and was intended for comment and criticism, but if you would prefer that withheld until it's in a more finished stage I'm happy to do so. So I'm going to assume that your last comment is a request to withhold criticism, and do that until I hear otherwise. - Tarkisflux ^Talk 03:17, 17 November 2011 (UTC)