What makes a good strategy game?

[javelinrl]

Might seem a bit off topic here since RLs aren't exactly strategy games but bear with me. I am the creator of JavelinRL which is quite a mess of ideas - it is party-based, uses D&D rules (d20 3e) and it differs enough from the rogue-like formula that I often have people asking me why it is called a RL at all... (but hey if even Dwarf Fortress is often called a roguelike I believe mine is one too). So I'm really not pleased with how all the ideas put into the game have turned out so far - I think there is a lot of potential but all of the disparate elements haven't *clicked* well enough together yet.

In this regard I'm working on the next release on-and-off which is going to change a lot of things - it already has mouse support, which makes the game more enjoyable for the non-hardcore crowd out there and right now I'm doing an almost-complete redesign of how the strategy/management layer of the game works. Hence, this post here. I'm trying my best to allow Javelin to be played mostly as a rogue-like/RPG by players who are only interested in combat, exploration and tactics but at the same time I have this whole system for city building, capturing enemy towns, producing units and such that are mostly seen in strategy games. By default a lot of these things are handled automatically unless you decide to take control over them.

So I imagine there's a lot of overlap between roguelike fans and turn-based strategy gamers as well, which is why I've come to you with this topic. I know I am myself a big fan of both genres! Also roguelikes do have their fair share of strategy and making the incorrect long-term decisions will guarantee you'll be dead before reaching your final objective in most games. RLs aren't easy games so you have little chance of winning unless you have good strategic planning and thinking along the way, including but not limited to how to spend character/attribute points, item selection, branches to visit, enemy types to avoid, hero builds...

So this is an open-ended topic that would help me determine if I'm going in the right direction with my current redesign efforts and hopefully raise good ideas and points for the community as well for the sake of discussing the genre. What do you guys think is the place for strategy in roguelikes (traditional ones and others like DF)? What makes a good turn-based strategy game? Over the years, which strategy games have got you glued to your seat turn after turn until 3AM on a work/school night? How did they achieve that? What was your favorite part (or parts) of the experience? Which games (roguelike or othewise) just didn't get strategy right and why did they fail? Any RLs you know that have more strategy elements than your typical dungeon-delving hack-and-slash game?

[getter77]

Number one thing to bear in mind is Time---the Length of a winning session, broadly, should greatly inform how granular all decisions can come to be in terms of gunning for a win or tilting towards a loss.   Nobody wants to throw down on something that lasts for a great many hours that you can, without often knowing, "lose" within the earliest phases and with no recompense to mount a satisfying comeback.   The newest X COM struggles hard on this negative-only snowball effect and only now with some of the more ambitious mods can it approach something more focused on spectacle and tantalizing rewards than time traps that just discourage and render moment to moment meaningless vs the bigger design.   4X outright is terrible on this front almost to the core, prostrating before an altar of false gods to Spreadsheet Maths and Grind For Opportunity Cost in lieu of all else.

[javelinrl]

This is a great point, especially for multiplayer games. There always has to be some sort of negative loop for the winning side so that the more this side wait to take advantage of their position, the more they start to lose the position. Not only does this presents an interesting strategical time constraint but it also gives a chance for the losing player to come back and win the game, resulting in that a player never really feels that the game is over and he is just waiting to lose.

One amazing game in this aspect is Starcraft 2. Lets say that you and your opponent get to 100 supply (army size, roughly, for those who don't know the game). Then you both enter a battle and you lose - now your opponent is 80 supply and you are 40 supply. Pretty scary, right? What happens next is that often the winning opponent will retreat and defend (a tactic the commentators call "when ahead, get more ahead") but if he waits too long it's possible that you both will have the time to gain another 100 supply. Now.. 140 supply (you) versus 180 supply (him), is not so scary anymore, right? You went from being half his supply (50%) to now being almost even (140 is almost 80% of 180). These numbers are not perfect but they illustrate how the game works, roughly.

This also is very interesting because it puts the winning side on a dilemma. After losing the battle the losing side will almost always have to retreat and defend but the winning side has to decide - do I attack now than I have the advantage or do I retreat as well and "get more ahead"? For people watching the game it's easy to look at numbers for both sides and know what is the right choice but for a player inside the game, who has no information about the other player's army, it's a very tough call to make. What if the other player has a hidden defense army back at home, or just a few strong units there? Should he risk attacking now and possibly lose the game if he is wrong? If he goes back, how long does he have to grow stronger before his opponent catches up with him and his advantage is negated? This is not only something that applies to Starcraft but any game with a proper negative feedback loop.

On the other side of the equation are games like Heroes of Newerth, which I've been playing a lot recently. It's a pretty good game but there is no negative feedback loop. Once your team gets a few kills on the opponent team, it will grow stronger, and as it get stronger it will get even more kills - creating a positive feedback loop for the winning team. Once the winning team is strong enough to start destroying buildings of the losing team they will grow even stronger and the other team even weaker. HoN games can last up to an hour and a half and no one wants to spend all this time just waiting to lose as you see your opponents get stronger and stronger, so people will always type "cc15" after a few deaths, even if the game is just beginning! "cc15" means that you should give up the game when it reaches 15 minutes, which is when the button to give up is enabled. It is still possible to win a losing game because HoN is pretty cool but I've always thought it had this huge fault in it's design and the fact that people want to give up the entire game when it's not even 10 minutes in shows that...

[getter77]

Thea is also probably a game to examine in this sort of regard, especially as it tries to marry and bridge the various design sensibilities---"tries", but still at least has glimpses alongside an emphasis on narrative and world building.

[doulos05]

A good strategy game is a game in which a competent player who is thinking at the same level of abstraction as the game will intuitively recognize the range of possible outcomes from an action. That's because the reverse will also be true. They can predict the range of actions which can produce a desire outcome.

Victoria II is a prime example of NOT to do this. 1) The layer of abstraction is... muddled at best. You're kind of the king of a 19th century power, except you have to keep track of individual populations down into the low hundreds (sure, you can ignore that. But if it's meant to be ignored, it should be hidden from the player). Furthermore, a lot of the knobs you can fiddle with (militancy and consciousness, for instance) don't have immediately self-evident outcomes, and certainly not intuitive ones.

A better example of this would be a game like Starcraft 2. Need more minerals? There's a clear and intuitive action (well, set of actions) that do that. The combat is similarly clear.

So what's the targeted layer of abstraction. I'm in a similar bind to you, in that I hope to expand Qaf to range from dungeoneering to kingdom building. But those are vastly different levels of abstraction. As a dungeon delver, I care whether that blow severed his weapon hand or opened up his stomach (it might affect whether that armor is salvageable or not). But as the captain of a force of 100 swordsmen, I don't care about how each monster is put down, only that they have been and at what cost to my own men.

The best implementation I've seen of this yet is in a tabletop game called Battletech. The rules themselves are clunky at times, but the implementation is beautiful. They have strategic combat rules where each counter represents a platoon of troops and damage is highly abstracted. But at any point in the game when two units are within a certain range of each other, the players can agree to "Invoke Battletech rules". The rest of the combat is frozen while these two units drop down to a highly detailed resolution model (it tracks individual missiles from ammo bin to impact point). At the end, you up-convert the units and resume the battle. I think you have to do something similar. I also think it should be a sharp switch (even display-wise) so that you 'reset' the player's abstraction level to the appropriate level.

TL;DR: Think hard about the layer of abstraction you want and make sure that no decision the player is making falls below that abstraction layer. That's what makes a good strategy game: good abstraction.

[javelinrl]

Doulos: good points! I plan on having an in-game manual that the player can check to see what each city building does, for example, as I have with most of the things in game already. Other than that I just tru to make things very clear ("do you want to hire a unit of this type" or "pillage location for x gold").  Other thing that is already in the game is that the player can turn on "strategic mode" (name borrowed from Master of Magic, which is a superb strategy game even today) - in this mode instead of starting battles, it will show you a prompt ("this is an easy/moderate/difficult/... battle. Do you want to enter the battle or skip to the results?") - this way players can always do battle with their "main party" but always skip battles with their "henchmen army" if they want, or just join both and decide when to fight and when to skip. In the future I want to make it easier to change it in the game's interface but the feature is already there. Inside battles, they can also choose which units should be controlled by the computer so it's possible to only control your heroes while allied mercenaries (or even some heroes) fight automatically.

In Javelin the abstraction level is pretty well-defined. Unlike most RLs it has an "overworld" and a 'battle mode" like most JRPGs, instead of having most monsters in the overworld itself. Many RL fans don't like this but on Javelin it is important for 2 reasons: 1. it is an easy way to control your party, instead of having control of one hero and the others following you around; 2. since Javelin is very strategic I wanted to keep the entire world in only one screen, more less, this way you can look at the entire game world at once and know what is happening everywhere - this way, you really need a big scale of the world map instead of something like Dwarf Fortress which has a relatively "zoomed in" view of only one area. Back to the point: Javelin has 3 levels of abstraction: world, battle and dungeon. You're in only one of them at a time so I think it's pretty well defined and clear for the player, without much overlap between the scales.

I've played table-top Battletech a few times before but I never used those "zooming in" rules. Cool stuff!! I imagine they would work pretty well, almost like mini-games inside the main battle... I guess Javelin's "strategic combat" is kinda like that where you can either control the battle move by move or just get an immediate result (like footsoldiers in BT), except that these happen in the world map and not inside a larger battle map with the other mechs! One of my favorite memories of BT is deciding missile hit locations - it feels sooo satisfying rolling locaitons once for each missile that hits, almost like you can hear the explosion each time you roll the die... and the face in your opponent's face when you hit a reactor...

[javelinrl]

Getter: I've found a pretty good lets play of Thea! I've only watched the first episode but it is quite informative. The game is very impressive for an early access game but a bit heavy on the micro-management for my personal taste. I'll try to watch more episodes when time allows, it seems like a pretty unique strategy game, very worthy of taking a few notes from https://www.youtube.com/playlist?list=PL0k9dWZr1FDdZhLY6_ViS-AwJKVPzFX3u

I also really love the idea of having card games within strategy games. This is why I'm modeling Javelin's city building mini-game to be a sort of trading-card and deck-building game, even if it won't be obvious for the player that this is how it is coded behind the screens. The idea behind this is that the player needs to work "with the hand he's dealt", so to speak, instead of always being able to follow a set strategy like in the Civilization games, where you can always build any unit/building you have access to and always follow the same path on the research tree. I've always thought this is not really how the real world works - as a ruler you can't always decide what your people will do, you can only try to steer progress in a way that suits your own plans... a card game is more like that since it draws on luck instead of letting you have absolute power over all decisions.

[doulos05]

The zooming in rules are part of the BattleForce rule set found in the BattleForce boxed sets and now in CBT: Strategic Operations and CBT: Interstellar Operations. Pretty fun stuff. Honestly, my favorite version of BattleTech is the Alpha Strike Ruleset from, well, CBT: Alpha Strike. All the fun of BT in 1/4 the time. Remember how a lance on lance fight used to take 2-3 hours? Yeah, in AS you can fight a company vs. company fight (with some conventional elements reinforcing both sides) in the same time.

[getter77]

Heh, that's early days Thea---game has been fully released since for a good while now and had a fair few updates and an expansion or so to polish and broaden the lot.     But Arthenex's old vids do give a fair bit of the gist of things.

[javelinrl]

Doules: if you haven't heard of it yet there is a pretty good open-source implementation of Battletech for computers. You can do multiplayer with friends, etc. Not being a huge fan of the game as you seem to be I'm not sure which rules version it uses or how faithful it is but from an outsider's point of view it seems to take everything pretty seriously. You can also play single-player against the AI, of course!

http://megamek.org/about

[Krice]

So I imagine there's a lot of overlap between roguelike fans and turn-based strategy gamers as well

I don't know if it's strange, but I don't like strategy games at all. As a developer I try to move even further away from tactical combat in a sense that it's too easy to make the player character yodel through the game in guaranteed safety if you apply tactical solution X to Y.

[javelinrl]

I really have to disagree with you. Any tactical game that has an optimal route is a bad tactical game, same with the strategy genre. Both tactics and strategy should be about adapting to the current scenario that emerged from complex, dynamic and independent factors - and so they're by definition impossible to beat with a one-fits-all-formula. Of course some tactics are "better" than others depending on the situation - sometimes it's better to charge towards your enemy and other times it's better to fight from behind cover; if all you got is tank characters it pays to engage the enemy directly, etc. If you've been playing any tactical or strategy game that isn't like that then you're just playing a shitty game and it's all that boils down to.

Of course any complex games, including most roguelikes have balance issues but that is something to be patched, not something that makes the game terrible. Older strategy games pre-Internet were not that easy to be patched so they could be more "flawed" in that sense but nowadays that's a lot easier to remedy.

My game development bible, for both RLs and other games is the DCSS design guide (which they call "philosophy"). Most devs here are probably familiar with that document but in any case, there's a section specifically about this concern over there: https://github.com/crawl/crawl/blob/master/crawl-ref/docs/crawl_manual.rst#crusade-against-no-brainers

[doulos05]

Doules: if you haven't heard of it yet there is a pretty good open-source implementation of Battletech for computers. You can do multiplayer with friends, etc. Not being a huge fan of the game as you seem to be I'm not sure which rules version it uses or how faithful it is but from an outsider's point of view it seems to take everything pretty seriously. You can also play single-player against the AI, of course!

http://megamek.org/about

Oh, I love MM. It's pretty much the only way I can play battletech given where I live. They have pretty much every option from the current ruleset implemented, but not alpha strike (my personal favorite) or BattleForce (my second favorite). I still play it all the time, but I wish someone would write a working implementation of AS or BF...