Sources of Tactical Depth

[Username17]

The OP makes an interesting list of behaviors which indicate lack of depth, but the 'knot-cutting' post sends rest of the thread into a spiral of "you're doing it wrong" back and forth. What was the intended thing to scrutinize here?

If I understand Frank's argument correctly, what he's saying is that any situation in which the optimal behavior can be determined in advance is a situation lacking in tactical depth. However, that could be said of nearly any tactical situation, provided you already have all the pertinent information. If you don't know whether or not your foe has a Teleport Redirection effect which dumps you directly into an antimagic field on the Negative Energy Plane, the question of what tactic to pursue against that foe becomes pretty meaningful.

To put it another way - if your players don't know whether or not attacking a foe with a ranged or melee weapon is preferable, they'll have to experiment to find out which is best in that particular situation, which lends a great deal of tactical depth to the situation. Suddenly, it's not necessarily best to have the highest initiative, because you might be the one who finds out that the monster disintegrates anything within 5 feet as an immediate action - but if you DO have the appropriate strategy, winning initiative has a clear-cut value above and beyond the others.

echo

Pretty much. Also with a new page, we can hopefully get back on track.

A "deep" tactical situation is pretty darn similar to a "confusing" or "mysterious" tactical situation. A tactical situation is deep, if the player's contribution of decisions matters. And it matters because of two criteria:

• The other players don't necessarily know what choices the player will make.
  and
• The choice the player makes actually impacts the course of events.

In the OP I mostly talked around the first criteria. If the scenario reduces to Top Decking or Move Spam, the other players in fact know exactly what the player will do and their contribution to the game is meaningless. But the second criteria is equally relevant. If the results of the player's choices are all basically the same (usually because the effects are trivial, but if they all make you win the gold victory star it's just as bad), it doesn't really matter what the player chooses and thus it doesn't make any difference whether the other players know what the payer will choose because they don't care what he chooses.

There is also an effective loss of choice that comes from too much mystery. If the players don't know what range of effects an action might have, the choices they make are pretty much meaningless - the only response the players care about is whatever crack ass tirade the MC goes off on in response. In short, questions like "The tunnel branches, do you go left or right?" are not meaningful choices in any sense of the term until the players get more information.

-Username17

[fectin]

So essentially, tactical depth needs a situation to be both deterministic and unsolveable?

[Username17]

So essentially, tactical depth needs a situation to be both deterministic and unsolveable?

No. It can still be probabilistic. It needs to be predictable and unsolvable. If someone says "Jacob did X" then the players need to be able to predict what the immediate results are going to be. Even if that immediate result is determined by a die roll or card draw, the players need to be able to understand what results could occur and what the chances roughly are.

And it needs to be unsolvable. Or at least, unsolved.

-Username17

[echoVanguard]

And it needs to be unsolvable. Or at least, unsolved.

-Username17

Doesn't having it be the latter simply delay the problem? I'd posit that for true, long-term depth, it needs to be unsolvable even for known parameters, probably by the inclusion of some kind of random mechanic (e.g. dice).

echo

[Username17]

And it needs to be unsolvable. Or at least, unsolved.

-Username17

Doesn't having it be the latter simply delay the problem? I'd posit that for true, long-term depth, it needs to be unsolvable even for known parameters, probably by the inclusion of some kind of random mechanic (e.g. dice).

echo

Chess is a solvable problem, but it has retained a fair amount of tactical depth because no one has solved it yet. There is a computer program that cannot lose a game of Checkers, but no such program exists for Chess yet. It's actually not even clear what the solution to Chess even looks like (although my personal suspicion is that it is "White Wins").

Since any game you write is going to be played a finite amount of times and having a finite amount of computing used against it, the game seriously doesn't need to be an unsolvable problem - just a complex enough problem that it isn't solved.

A game is therefore like a cryptographic code. It doesn't need to be unbreakable, it just needs to be hard enough to break that it is not broken within the time frame that it is important.

-Username17

[Lago PARANOIA]

The OP makes an interesting list of behaviors which indicate lack of depth, but the 'knot-cutting' post sends rest of the thread into a spiral of "you're doing it wrong" back and forth. What was the intended thing to scrutinize here?

I apologize for that. What I was trying to get at is that having completely alternate solutions destroys tactical depth and that games where this is theoretically a possibility (such as D&D but not Chess) can have this as a maladaptive response because it means that people aren't interacting with the intended game. I didn't expect for it to spiral into a bunch of 'I'm a pretty pretty princess Mary Tzu and look at me proving my intelligence aren't I smrt?' wanking. Which is why I stopped posting after that, because I am ashamed at what I unleashed. Fuck you, you wannabe Napoleons. You're not clever.


But getting back to Frank's real point, the unsolvable part of the equation is key. It looks easy enough to construct it, which is why a lot of people mewl 'stop being a lazy DM!' when you bring this up, but this is actually very very hard. Having a balanced game helps a lot, but it's not even close to being able to make a game unsolvable, as we've seen from 4E D&D.

Gygaxian DMing aside, the typical shake-ups as far as TTRPGs are concerned is to continually release new material when it looks like players are about to solve the game. You know, like if primary spellcasters are overpowered, release Book of Nine Swords. I don't especially approve of this method of obfuscation because it greatly increases the chance that someone will introduce material that fucks the game up in other ways. And of course having the DM introduce on-the-spot tactical depth like 'these trolls have fire resistance' has its own set of problems. The only way that consistently adds 'unknown knowns' to the game without fucking other things up is randomness. I think that D&D needs a lot more randomness in key phases of the game. I'm a huge fan of WoF, but I'm also worried that more randomness than that will disengage people from their players. The changes need to be implemented DM-side. For example:

- Randomized Battlefield effects. The DM rolls on a table and a scenario gets applied to the situation. This applies even to encounters that are created beforehand like in a module. For example, on a roll of 23 you could make it so that half of the monsters defect to the PC side and attempt to assist them. On a roll of 36 it turns out that the monsters are infected with some kind of illness and if the PCs aren't careful they can get it, too.

- Randomized treasure. We've tried having player-run magical item catalogs in 3E and 4E D&D and it failed miserably. Now while it's a good way to move books and I can understand you not wanting to ditch it, in terms of gameplay randomized or at least semi-randomized treasure is the way to go.

- Randomized campaign worlds. I think that this will be really, really difficult to implement (and may not be possible at all) but not only should there be a feature to generate a campaign world FFTA/Schrodinger's Cat style where Bloodpool might be next to the capital city, but DMs should also totally be able to roll and see 'oh, hey, a plague hit Fillmore last week'. It might not SEEM as though it's a source of tactical depth, but in good groups it's actually a really big deal whether a zombie apocalypse can show up in the Holy City Glyover.

[Almaz]

TORG did the "randomized battlefield effects" bit, although more on a round-by-round basis, and I wholeheartedly agree that a randomized battlefield circumstance, be it per-encounter or per-round, is a great thing to add. Even if it acts in such a way that only one or two characters significantly experiences the randomization, that can change the experience for everyone.

TORG's version was more cinematic while simultaneously smaller in effect... it was like set directions. For instance, a card might suggest something like "the heroes fall back under enemy pressure!" and the heroes get a bonus to any defensive actions they take, but the villains get to take more shots at them, dropping a huge incentive to move for cover because the villains will pile on the pain if they don't. Or maybe it was just "the heroes get crushed!" and they suffer a broad defensive penalty. Or sometimes the villain would go off and grandstand for a round instead of doing anything useful, because he's busy gloating, and all the heroes are strong. Et cetera. In any case, sometimes you don't care and just go "pshaw, a puny modifier? I ignore it!" But sometimes it makes the difference between actions.

I would suggest having the battlefield effect be something "hot swappable" if you're rolling on a table and the results are highly specific like betrayers, etc. So that, like some of the suggestions for Winds of Fate rolls, you could quickly generate a variant table and roll off that. This would help the results make a little more sense sometimes. For instance, for an orcish ambush in the forest, you might string together the "war" and "forest" and "surprise" circumstance tables, whereas for hunting a shadowy monster in a farmer's village, you might take the "farm", "terror" (read as: terror mission), and "abomination" circumstance tables.

I have no comment on randomized treasure other than "yes."

[echoVanguard]

Okay, so we've now identified two types of independent depth factors: unsolvable and not feasibly solvable. There is of course the risk that as player access to computing resources increases, situations that may have once been unsolvable in active play can become solvable, although discussing it is kind of pointless until it becomes a reality (however, see http://www.telegraph.co.uk/news/worldne ... andal.html).

echo

[Swordslinger]

Gygaxian DMing aside, the typical shake-ups as far as TTRPGs are concerned is to continually release new material when it looks like players are about to solve the game.

This is pretty much the best solution. Give monsters lots and lots of unique abilities. Have your PCs constantly finding ways to beat new things to keep them thinking of tactical counters. And as soon as they figure out how to beat something, either mix that monster in with others that change the scenario or start using a different style of monster entirely.

[Lago PARANOIA]

This is pretty much the best solution. Give monsters lots and lots of unique abilities. Have your PCs constantly finding ways to beat new things to keep them thinking of tactical counters. And as soon as they figure out how to beat something, either mix that monster in with others that change the scenario or start using a different style of monster entirely.

While this is very important as a source of a tactical depth, you can't solely rely on it. Aside from the fact that having four hundred kinds of 12th level monsters destroys versimilitude, you simply do not have that kind of space nor do you have the time required for playtesting. The problem gets even worse if you make monsters more complex. Now not only do you have fewer monsters per book but it also makes it harder for a DM to figure out how to run then.

This is why I pretty much ignored you and wotmaniac in the WoF thread when you said that this problem could just be solved by telling the people to get more creative with monsters. Even if all of the actors in the game behaved as intended (which obviously isn't the game), the fact remains is that after a certain point that this is IMPRACTICAL. The game should be able to be competently run on one monster manual and there should generally be at least three different flavors of monsters per encounter. There's a limit to complexity.

[Swordslinger]

While this is very important as a source of a tactical depth, you can't solely rely on it. Aside from the fact that having four hundred kinds of 12th level monsters destroys versimilitude, you simply do not have that kind of space nor do you have the time required for playtesting. The problem gets even worse if you make monsters more complex. Now not only do you have fewer monsters per book but it also makes it harder for a DM to figure out how to run then.

Which is why I like a system where you can create custom monsters easily and quickly. Less than 50% of my 4E "monsters" come from the manual. The rest are custom.

The nice thing about 4E style is that you don't need a ton of weird races to make things tactically interesting. Hobgoblins alone can have a bunch of special types with different abilities to make a hobgoblin force sufficiently interesting and retaining verisimilitude, so long as you accept a world which has a wide variety of fighting styles beyond those that the PCs have.

In fact, the flexible 4E system really lets you make a series of interesting monsters without compromising verisimilitude because you no longer have to worry about the fact that minotaurs make terrible wizards, and go ahead and just make a minotaur spellcaster if your plot needs one.

[Lago PARANOIA]

The nice thing about 4E style is that you don't need a ton of weird races to make things tactically interesting. Hobgoblins alone can have a bunch of special types with different abilities to make a hobgoblin force sufficiently interesting and retaining verisimilitude, so long as you accept a world which has a wide variety of fighting styles beyond those that the PCs have.

Oh, wait, you're serious; let me laugh harder.



Look, even if you slap 'Bladeling' or 'Goblin' onto a faceless brute, that doesn't suddenly shake up the players' tactics. The cleric still spams (errata'd) Righteous Brand and the Rogue doesn't decide to use something besides Deft Strike or Piercing Strike for the first time in 100 battles. Let alone do something simple like creating enough complexity that a player may decide to dip into their ration of Daily Moves.

The whole point of tactical depth is to encourage players to in fact change up their tactics. Generic Soldier mobile with 'Eladrin' slapped on it makes it a bit more annoying for the meleefolk to chase them down, but it doesn't exactly make them think about what they're doing. 4E's racial system, 5MoD aside, does not accomplish that and I'm totally convinced that you're shilling.

[Swordslinger]

Look, even if you slap 'Bladeling' or 'Goblin' onto a faceless brute, that doesn't suddenly shake up the players' tactics. The cleric still spams (errata'd) Righteous Brand and the Rogue doesn't decide to use something besides Deft Strike or Piercing Strike for the first time in 100 battles. Let alone do something simple like creating enough complexity that a player may decide to dip into their ration of Daily Moves.

That's not because of the monsters. It's because 4E doesn't have complex PCs. If you hand the PCs more options, then they have more options to counter the monsters set up. The idea of roles was a pretty terrible one. All characters should be able to play at least 2 roles, if not three, and switch them as the situation demands.

[echoVanguard]

The idea of roles was a pretty terrible one. All characters should be able to play at least 2 roles, if not three, and switch them as the situation demands.

This is a pretty perceptive statement. I think having sufficient options is another key component of tactical depth.

echo

[spasheridan]

What I enjoyed about 4th ed was the first few levels. Our group swapped characters a bit and the new powers that you gained were meaningful. There were fun choices about when to throw down dailies, when to use action points. With good pacing you had 3 "days" of adventuring each level usually and then your choices as a group really changed.

So day 2 and day 3 of level 2 might be similar, you might still have a few neat combo's using the warlords action point aura or the cleric's new healing combo. Then the new leveled up team had a few new interesting combo's to work out and see how they behaved.

I think it remained enjoyable up until level 4 or 5.. then each new power was either the same as another in the team's aresenal or actually WORSE than an existing lower level power. I don't think there was much replayability because I think we had basicly won the system - commanders strike with the fighter for more marks, action point within 5 squares of the leader for the uber win, ranger for the alpha strike damage. IT really became a rinse, repeat system and a new party would do the same.

3rd ed never really hit that with my groups, the variability with prepping new spells made even the same group play differant each day.

As far as knot-cutting, about the only awesome tech we discovered was Make Whole. We would find awesome shit and cut off a small piece and burn what was left. Like an arsenal from an old castle. A bunch of barrels. Sacks of food. Then a while later we would need to float down a river or equip a militia and we would bust out our splinters and create whatever tool we needed.

Compared with the 3rd ed options of summoned astral constructs bull rushing save-or-die undead into lava pits or tunneling throqua's under walls, stone shaping walls to trap monsters inside air-tight caverns, or the ever popular scry-or-die tech the 4th ed knot-cutting always felt a little lame.

We managed 2 4th ed campaigns before the dreariness drove us off. It just got boring. 3rd ed never bored us, we just leveled up to a point where the complexity of a session was too annoying... "OK, we have these 8 spells on the party, then we polymorph the cleric into a war troll. Let's pause for 15 minutes while we recalc all stats. Next round a dispel hits the group.. roll for 10 minutes.. ok, let's all recalc our stats. Now that utter-cold flame strike spell comes down, let's split that damage into 3 components and apply all resistances, make saves. Ahh.. new buffs from the wizard, let's recalc again for the summoned monsters that didn't have haste already..."

When the fighter attack and rolled a 6, we would spend 5 minutes making sure we hadn't forgotten the last +1 to hit he needed to get his swing in...

A few sessions like that and we would gladly start over at level 1 just to get 3 encounters done in a single session.

[Winnah]

I've seen this in a couple of games.

Button Bashing:

Complete lack of tactical depth. The character throws every ability at the problem as fast as possible in the hope they do something right. Not just the behaviour of novice players, Button Bashing can be a response to a challenge where the characters normal tactics function poorly or fail to have any meaningful effect.

Usually a response to some high level brute monster with lots of resistances and immunities. Contingent liches and their ilk can trigger this, but have methods of routing encounters before characters resort to chaotic action spam.

[TheFlatline]

Look, even if you slap 'Bladeling' or 'Goblin' onto a faceless brute, that doesn't suddenly shake up the players' tactics. The cleric still spams (errata'd) Righteous Brand and the Rogue doesn't decide to use something besides Deft Strike or Piercing Strike for the first time in 100 battles. Let alone do something simple like creating enough complexity that a player may decide to dip into their ration of Daily Moves.

That's not because of the monsters. It's because 4E doesn't have complex PCs. If you hand the PCs more options, then they have more options to counter the monsters set up. The idea of roles was a pretty terrible one. All characters should be able to play at least 2 roles, if not three, and switch them as the situation demands.

That's all well and fine, but the added options have to be at least as useful as the powers that the players already use now. Which, if they are, result in power creep. And even if they are equally useful, you pick one to spam, and fall back on the other in the event that you can't spam the first any more.

To make "moar powerz!" a valid option, you're going to have to make the current go-to powers more situational, or make the new powers vastly more useful in a situation that the PCs can create.

Take sneak attack from the rogue. It's vastly more useful to use than a standard attack, but requires a specific situation- that the PCs can create, before it can be used. It's valuable enough that rogues will go out of their way to create a sneak attack opportunity.

Reworking the ability system to favor high risk/high reward or situationally beneficial powers would require a fundamental reworking of the core classes- and classes like the mage would be nearly unrecognizable (the best spells may take several turns to pop off or something).

It's not a difficult concept: if you have a power that takes 3 turns to set up, it's damage output should be noticeably more attractive than the default button mash for 3 turns. If it's less attractive, or about as attractive as the default, players will shy away from using it, because the added risk isn't worth the reward.

The more I think about D&D, the more I want to cap the core classes at like level 10. Maybe even level 5 or 7 (though that's pushing it). Shrink the PrCs down to 3-5 levels maximum, and eliminate entirely multi-classing penalties. Therefore, by 20th level you might have a fighter/mage who can fight as a world class fighter and cast as a world class mage. Let PrCs be super-focused (martial fighting styles are a 3 level PrC dip to master, casters might have 5 levels to max out a school), and start from there.

The problem I see with multiclassing to give yourself more options is that you end up never being as good as the equivalent level of a dedicated whatever. There's 20 levels in all the core classes, which means if you multiclass, you lose out on benefits that core class would give you. But if there's only 10 levels in fighter, once you hit 11th level, you're as good of a fighter as any fighter you're ever going to see. Now you can focus on more martial flexibility, or put yourself on the path to being a cleric so that by the time you hit 20, you're a great cleric and a great fighter.

There was a Swashbuckling Adventures core book that was shit for 3rd edition (I think it was a 7 Seas D20 book), but I always thought the 3-5 level PrC dips for different fighting styles were awesome ideas. They were shallow enough that you didn't have to dedicate massive portions of your character's career to them, and with a 10 level base cap system, it's feasible that even a wizard could take a martial style to help protect himself (though probably not efficient).

[CatharzGodfoot]

Randomizing monsters and the game world is entirely the wrong direction to take.


Randomizing the game world has no effect on tactical depth. The entire campaign exists in that world, so either the players quickly get used to it or the world stays baffling and unpredictable. Neither situation increases tactical depth. If changes in the game world are meaningful, you've also unbalanced the characters for the entire game. It also kills things like world building.

Randomizing monsters won't improve tactical depth easier, because although it makes the game less predictable ('solvable'), it also reduces meaningful player choice by making the worth of each choice less clear.

That said, if you randomize monster powers (or, equivalently, the GM assigns them classes), but give a good indication of what those powers are, the meaningfulness of player choice remains while still mixing things up. This is somewhat similar to WoF, where you can know the opposition's overall capabilities without knowing their options round-to-round.

Another option is extending the duration of combat to allow players to figure out their enemies' capabilities and then act on that information. Unfortunately, combats tend to take too much player time as it is.

Randomized treasure changes the tactical depth of a combat only as much as your tactics change when fighting a trolll with a flaming sword instead of a troll with a spear of ice.

[Swordslinger]

To make "moar powerz!" a valid option, you're going to have to make the current go-to powers more situational, or make the new powers vastly more useful in a situation that the PCs can create.

Of course. Given 4E was the example, it's easy to see how it could be improved upon, given that many classes have at-will powers that all attack the same defense and have the exact same range/targets. Usually melee 1 vs. AC.

Just giving people a mix of melee attacks, ranged attacks, area attacks and attacks versus different defenses goes a huge way towards giving you reason to mix it up.

Take sneak attack from the rogue. It's vastly more useful to use than a standard attack, but requires a specific situation- that the PCs can create, before it can be used. It's valuable enough that rogues will go out of their way to create a sneak attack opportunity.

Sneak attack should look similar to 3E sneak attack where it won't function on certain creatures. This means that if you throw humans at the rogue, he's trying to line up sneak attacks. Against an iron golem though, he has to mix up his tactics.

The problem I see with multiclassing to give yourself more options is that you end up never being as good as the equivalent level of a dedicated whatever. There's 20 levels in all the core classes, which means if you multiclass, you lose out on benefits that core class would give you. But if there's only 10 levels in fighter, once you hit 11th level, you're as good of a fighter as any fighter you're ever going to see. Now you can focus on more martial flexibility, or put yourself on the path to being a cleric so that by the time you hit 20, you're a great cleric and a great fighter.

The main problem with multiclassing is vancian casting progression and ability scores. In another thread, Lago or K (I forget which) mentioned removing ability scores from the attack equation. As far as missing out on options, you'll always be missing out, unless you want all PCs to eventually be the exact same character. Missing out is beneficial to a team game because it leaves another niche for other party members to fill. Ideally you want even maxed classes to be lacking things that another character could potentially have. So two people playing wizards or fighters aren't playing the exact same character.

[Username17]

Randomizing monsters and the game world is entirely the wrong direction to take.

I would amend that to "partially" the wrong direction. In that it would be insufficient and if you did it enough it would be really annoying.

Randomizing the game world has no effect on tactical depth.

That is not true. It has a positive effect on solvability, but a negative effect on predictability. So if your problem is that the game is too solvable, adding random elements makes the tactical landscape deeper. If your problem is that the results of actions is not predictive enough, adding more random elements makes the tactical landscape shallower.

-Username17

[MGuy]

I agree, randomizing battlefield conditions would make the game much more interesting. I say that because a lot of the better encounters I've run for my groups were in hostile or just plain odd environments. Even introducing a cliff, lava pools, etc on the battlefield produces something a bit more interesting during fights. I can only imagine how a "battle tide" system in which certain actions are randomly "better" than others on a given round would work. Hell if it is a standard part of the system I'd even assume there would be a way to influence it with certain class abilities (such as a Ranger's or a Bard's) I'd imagine.

Randomizing monsters on the other hand doesn't sound appealing.

[mean_liar]

I'm a big fan of incomplete information: most games are bad for this because your character makes his action decision the moment its his turn to act, and therefore its rare to get caught short outside Ready actions and the like.

By simply hiding some information a la Burning Wheel's combat system, you introduce a lot of depth to play that didn't exist prior.

[echoVanguard]

I'm a big fan of incomplete information: most games are bad for this because your character makes his action decision the moment its his turn to act, and therefore its rare to get caught short outside Ready actions and the like.

By simply hiding some information a la Burning Wheel's combat system, you introduce a lot of depth to play that didn't exist prior.

Can you elucidate, mean_liar? A cursory search didn't turn up details.

echo

[mean_liar]

In Burning Wheel everyone chooses multiple combat actions in secret (corresponding to a round of combat), and those multiple actions are slowly revealed in turn, in order of execution throughout the course of a round.

Its similar to Broadsides and Boarding Parties, if you've ever played that old Milton Bradley game.

[fectin]

kind of like Killer Bunnies, only less tedious?