Why Balancing Multiplayer Games Is A NIGHTMARE (ft. Riot Mortdog) | Design Delve
Vložit
- čas přidán 27. 05. 2024
- This video is sponsored by Let Bions Be Bygones, the new atmospheric future-noir thriller out now on Steam -- bit.ly/Let_Bions_Be_Bygones
In this episode of Design Delve J & Ludo sit down with Riot Mortdog to discuss the difficulties of balancing multiplayer games.
Support us on Patreon: / secondwindgroup
Second Wind Merch Store: sharkrobot.com/collections/se...
Music used in order of appearance:
Intruder - Stray OST
The Abyss - Hyper Light Drifter OST
Checking in - Celeste OST
The Notebooks - Stray OST
Secret Lab - Stray OST (edited) - Hry
Join us at Noon CT for the premiere of our new game developer hosted podcast, Dev Heads! czcams.com/users/liveOs-gt-15XKM?si=rOLNoJk4RrHj3UwX Hosted by JM8, Tina Sanzhez and Michael "Mikee" Tsarouhas.
❤️👾🎮🕹️❤️
No way you got one of the biggest and most transparent developers to actually talk about game design. That's actually so awesome to see since I love both your discussions and Mortdogs design philosophies.
This is my first time hearing anything from riots balance team, not sure how transparent I would call them!
@brycebedah9609 riot as a whole is a different story but Mortdogs streams are awesome for insight as to why the TFT team make their decisions
@@brycebedah9609Riot is a terrible company in most other regards, but Mortdog is a great example of developer-community interaction.
Too bad riot games are a bunch of chinese chills
@@brycebedah9609 It's not so much an issue of transparency as it is that they blatantly have no idea how to balance, and never have.
Perception of buffs and nerfs is so impactful, in every multiplayer game I've played people immediately gravitate towards things that were buffed in the most recent patch, even though in theory the buff simply means it was bad and now should be neutral. Same thing with nerfs, if something goes from overpowered to neutral, doesn't mean you should immediately stop playing it.
This phenomenon tends to affect bad players more than anyone else, funnily enough.
Reminds me of the fortunate shotgun that got "nerfed" and the usage rate dropped from 75% to 25% (I forget the actual numbers). The nerf to it? Item rarity was dropped from legendary to uncommon. No effect on stats, chance to find it, or anything at all... appearances are everything.
Ya but thats a very short term and temporary effect for the most part. Yes, people will experiment more after a patch, but give it some time and understandings will converge over time back to where it "should" be unless the experimentation actually bore fruit.
I think a large part of that is simply curiosity on what the buff is like that brings in lots of interest when it changes. Same with the nerf, but folks are generally less interested in a 'worse' experience than the one they already know...
This reminds me of a similar situation in Wolfenstein: enemy territory. The game had two submachine guns, one for each side, and players complained that one was stronger than the other. The issue was that the guns were actually identical mechanically, they just sounded different. The devs fixed this by "nerfing" the supposedly superior guns sound, which fixed the issue. Just another example of how perception of power can be more impactful than actual power.
I actually had a balance experience with Guild Wars 2 just before Heart of Thorns was added.
My friend played a Thief, and I had recently started playing with a theory-crafting tool for builds. So, naturally, being a good friend, I decided to make a horrible meme build for my friend to try. I started, of course, by replacing all of the attack and crit stats with HP and armor stats. I then replaced all of the damage moves with DoT moves, and then, I noticed something: in PvE, you could use food items to increase your DoT stats. Add to that a worthless trait that allowed you to share your “make weapon apply debuff on hit” abilities with nearby teammates, and I had made my friend a, “DoT-support thief”.
So, laughing my ass off, I give this build to my friend. They think it’s absolutely hilarious, and add a weapon combo to it that gives them a super-long evade at the cost of being unable to act, and we jump into PvE. The build broke the game, and my friend proceeded to use it through all of HoT (we were actually one of the sub-squad leaders (we were co-leading based on the task) in the first clear of Mordremoth because my friend could keep eyes on the battle as a whole while in the invuln state).
We also took it into PvP, and I must say it was some of the best trolling I’ve ever experienced. People would target the “squishy thief” only to find out the hard way that that wouldn’t work.
And then the entire build got merged into the base kit for thief at the start of Path of Fire and we both kinda dropped the game, but it was extremely fun while it lasted.
Second wind has blossomed in to an amazing channel. Please don't stop making and creating such great videos.
unlike escapist I actually watch the other videos besides zero punctuation
"You take a card game like Hearthstone"
Legends of Runeterra: T^T okay
Who what now
One slight correction on the Vlad nerf story. He got two nerfs in patch V1.0.0.111 - a strict numbers nerf to his passive and removing movement speed buff during an ability. The movement speed nerf never happened, but the nerf to his passive did. So there was a real nerf explaining some of the winrate drop.
People still complained like crazy on the forums about the movement speed nerf though.
My favorite balancing story comes from Iron Galaxy and their work on Killer Instinct 2013. Way back when Jago was a particularly strong character, all the players pointed towards only his healing being too strong. Every time someone pointed something out it was his ability to recover a good chunk of health. Instead of immediately tuning the healing, they played around with Jago, understood why he was a little too strong, and came to the conclusion that it had nothing to do with his healing, he just had too many good options at mid range. So they tuned down a few of his moves to make him more punishable. Of course, everyone noticed the healing was unchanged so they still complained, but slowly after people played more and more of the update, they stopped whining. Jago was more reasonable. The philosophy they use here is finding the "symptom" of everyone's complaints, not the issue itself.
There's a saying along the lines of "the player base is great at finding problems, but terrible at finding solutions."
Man, I made a comment about this but now saw yours haha
It's been said by many but having only watched ZP on Escapist I've really been enjoying the different things on Second Wind, I think this and Cold Take have been a genuine surprise and glad I started watching them. Keep up the good work 👍
I 100% second this. I actually prefer cold take over zero punctuation these days.
@@lukesmith6817 I still enjoy his stuff, I think the Semi one more. Cold Take is mostly because of Frost's voice 😂
Riot employee using Hearthstone as an example and not Riots own card game Legends of Runeterra
As someone who put hundreds of hours into LoR, that was a stap from a thing that I thought was dead.
I genuinely think part of the issue is that balance is so nebulous to define because I have seen people online imply that a game is only balanced if the more skilled player will win 100% of the time, which I think is a very limited and flawed view, even if you could come up with a quantifiable measure of skill.
I'm imagining a system that could quantify skill to that degree with the result being players only playing on their own as the system would just say "Skilled Player Wins" when it comes to matches, since no actual execution is needed to prove otherwise.
Some credentials before I write this: I am a TTRPG developer. I’ve worked for Dreamscarred Press on Pathfinder 3P content, and more recently led the team that made the Liminal Space book for Lancer RPG. In these, players are on the same side… but balance is an issue.
If one player is significantly stronger than their peers, the game becomes less fun for those who are losing the opportunity to play. This happens most often in tables of mixed skill levels, but in this format you cannot just let a bad or inexperienced player lose. That isn’t fun, and there is more to do beyond throwing dice that you need to make sure they stick around for. To this end, we like considering the skill FLOOR as much as the skill CAP. That is, if someone makes a build for their character by throwing darts at a board blindfolded, how well can they still do?
Most DnD-like games have infamously low skill floors. What we did was do our best to nudge the power of “bad” options upward by a bit (so that they’re at least playing in a similar league of power level to their fellows), but that was only half of it. By far the more important thing was giving the players agency to swap out bad choices much more quickly than the base game. This let them experiment with something they think is “cool,” but also escape from a dead end build if they hit one. Not everything could get this treatment (due to needing to interface with the base game), so other tricks included making the options people get "stuck" with be more broadly applicable. that way, they could usually be used to support things that could be changed both during experimentation and after a swap.
This process took a lot of time and iteration, but was very much worth it in the end.
I think the problem with this is that raising the floor invariably raises the ceiling, especially in a game as hard to balance for as Pathfinder. Having more usable options means more chances for things to work together that really shouldn't.
Edit: Just to be clear, you certainly can balance the two. Psionics feels like it was built with this assumption and is probably the best balanced thing I've seen in Pathfinder. It just has the potential to go wrong, which I would argue Path of War did.
@@mewmeister8650 you’re not wrong, and it is a real concern. Heck, one of my main jobs was to be the simulation guru. I distinctly remember one conversation that went, roughly, “when you’re measuring your damage in Tarrasques Per Turn, we are outside of acceptable balance.” The trick is finding the point that you are aiming at and try to get there.
Dreamscarred Press was perfectly fine raising the power ceiling by three points if it meant the floor was raised by ten, so we took it as an acceptable trade (and specifically noted it to the customer base). Perfect balance in a system that complex is not reasonably possible, though, and the hard part ended up being knowing when to say “enough” and ship it.
Editor of perilous prints here and table top designer in the making. The problem with raising skill floors in such like the super hero games m&m etcetra is that they reduce the value of player choice towards more of a pure flavor direction and turns the game into math rocks the game. Which can harm narrative story telling.
Second making wide choices too good suffers from the the biggest problem between having low skill players and high skill players at the same table. The problem isn't a difference in strength but in ability. A character being min maxed to do 1 thing absurdly well is fine for the party (though not always the gm) the problem comes in when 1 character can do everything better than the rest of their party.
A party doesn't have to be balanced, but every member on a team should feel important to the team and so having a wide list of different and effective roles that can't easily be done by 1, character is probably the most important thing.
The reset mechanic does sound really cool though I imagine the min max players will be using it more than the struggling player will because min maxers tend to be more involved in the games mechanics.
Honestly makes me wonder if someone could make a game in which all players design a character together and then play x copies of said character take off
I get what you're saying, but at the same time, looking at a TTRPG like D&D or Pathfinder as just about balanced mechanics kind of misses the point.
I have been playing some form of TTRPG, mostly variations of 3rd Ed D&D, for 24 years now (since it was first released). When I go to make a new character, I know exactly what the most optimal choices are... and often do not choose them. For me, flavor always wins over all else.
And that's what a lot of those "less good" options do - they add flavor. If I want to play a pacifist mage who only deals non-lethal damage and battle field control debuffs, I can do that. If I want to make a fighter who specializes in the mancatcher, I can. And those options feel different than than the more standard, more optimal options.
From what I've seen, the real issue with inexperienced players is that they don't understand the need to communicate with other players. Picking abilities that mesh with those of your teammates is important, but that's a learned skill.
The consideration of skill floors and ceilings I agree with so much. It's pretty much a question of "how big is the difference between a character created by a player who has no idea what they're doing and one created by a player who knows EXACTLY what they're doing".
It's one of the reasons that I've never really liked Shadowrun. Because it can feel really disheartening when you try your damndest to make a good gunslinger and end up with a Pistol skill around 12, only for one of your buddies to "low-ball" his Pistol skill and still end up with 14 or more. All because he knows all the ins and outs of the system, knows exactly what qualities and equipment to pick, and you don't.
"Perception is reality" was essentially my experience with Dark Souls. When i first played it I was of the mindset that it was "so unfair" that a boss could just two or three shot me when I had to slog through 50-100 hits to kill them. My perception was that there was no way i could win because I was so much weaker than my enemy, but it wasn't the balance of damage and health, it was the balance of my own ability and the way I approached the encounter. That felt like a powerful realization.
Exactly. I find it amusing when players complain when a weapon/spell/whatever has been buffed or nerfed when that sort of thing isn't something I necessarily notice, particularly since I don't do PvP in those games. I remember going to fight Midir and absolutely expecting it to take me days of attempts, but I got him on my first attempt (1 attack really should have killed me, but I had the merest sliver of health so I could heal after I got hit). The perception is why I take a break when I've been stuck on a boss for a while. Knowing what does and does not work for me is very important. I'm playing around with a faith-based build in Elden Ring and 1) my stats are all over the place given how I'm leveling up and 2) because of that it feels quite different to play. I feel weaker even though by levels I'm not too bad. My quality melee character at the same level would absolutely beat the crap out of this equivalent faith-based character.
Everyone I try to get to play souls likes is I tell them the game has to break their perception on fights. You can't take hits, your Estus equals the number of mistakes, time to kill is half of your skill, etc. It doesn't play like a normal rpg
Knowledge based gameplay is a thing in RPGs. Biggest problem is to make it discoverable enough so players can take on tasks organically.
@@Pecisk that is very hard to do in ROGs since most of them are so systems reliant so the player is made to think they are limited by the systems themselves. Like, in the new FF7 games, how do you explain the stagger system and the material system through gameplay? I think some styles of RPG get a pass. I do think it’s important to teach through gameplay wherever possible. In my opinion it shows a devs understanding of game design the more they can do that.
I wish more designers had Mort's consideration to the 1/10000 chance for someone to be having a mindboglingly misserable experience as something that should be prevented rather than sticking RNG on a pedestal and shrugging that most players will be fine. Even if it's never happened to me personally, it just feels almost criminal to willingly subject some players to that experience, intentionally.
I think some of the biggest problems with game balance is when games fail to help the player understand why a certain tactic/build/whatever didn't work. It makes it really easy to play a game and destroy people with one strategy, then find someone who knows how to counter it and decide that their race/hero/champion is just OP.
The "perception" point reminds me of that game where they had two guns (the Thompson and MP40) that were team reskins of each other, but because one *sounded* more powerful, players thought that one was stronger. Feelings are so hard to account for in the balancing game!
Well, one thing to understand is that a lot of developers don't play the same version of the game players do, and most are in fact terrible at their own games. They also very, very, rarely look carefully at their own code, and not all devs are honest with each other on the same page.
The simple reality is that the first sign of a problem is when players tell developers there are issues, and their spokespeople say such is impossible, or that their game is hacker proof. Rarely do developers even check their scripts to see if people might have found a way to get into them. This is why script hacking... basically messing with the number string that is used for "randomness" for so many types of games (computers being incapable of truly being random, and rather using a script... oftentimes a long tone, to fake it) is one of the most common things you see.
At any rate what certain devs claim, or even genuinely think, is impossible, oftentimes is not. Perhaps they think something is "just an alternate skin" but someone messed with the animations and/or attached hidden riders to it. This could be something like slightly faster projectile speed, a slightly more accurate sight, or a script that simply favors the high end of the potential damage yield. There have been a fair few "skins" especially those with "custom animations" that have had gameplay impact despite claims to the contrary, and I don't believe for a second that it's usually been accidental especially in games with competitive aspects.
I know one game.. sadly I can't remember the title, but it was an MMO shooter, where some of the skins "accidently" changed projectile weapons into what were effectively now hitscan weapons despite the animations and weapon class making that counter-intuitive, but it was very easy to abuse them if you realized this. Shockingly these weapons were the killing tools of choice for some of the devs and their peer group though being listed as lower tier premium weapons in the stat lists so players assuming the numbers were accurate wouldn't buy them. It might have been Planetside 2, but I'm not 100% sure on that, it's been a while.
I recall the original Puzzle Quest. A common conplaint is that the CPU cheats.
Its a match 3 game, so you can sometimes trigger a cascade where one match leads to lots in a row with just one move. The player got about as many as the CPU did, with it being slightly weighted in the players favour.
But not weighted enough. I remember reading an article where the devs said the player needs to get 70% of the cascades in a game to feel like they got an equal amount as the CPU. Perception is everything.
The player will assume a cascade in their favour is something going to plan and a CPU cascade is unfair. By having them equal they made players feel like they were being bullied.
Ahhh Puzzle Quest. I loved that game so much.
I am legitimately only upset about the final boss just gaining massive amounts of spell resistance stupid fast. My Wizard was useless against him, doubly so because I relied a lot on Haste combos. ;_;
I never got to finish the game as a result.
First person shooter games have the same issue. NPC that actually plays like player is very annoying. Players play game to feel powerful, not slapped around by always perfect counter player.
@@Pecisk We've all had that moment where something BS happens in your favour and you think "Ha, see how YOU like it!" despite the fact chances are you get those moments way more often than the computer does.
The impact of perception on gameplay makes a massive difference on whether something is viewed by the community as balanced, broken, or underpowered. With games with a ton of weapons in particular, it feels like people will see a video of a particular weapon getting super powerful in the right situation, and determine it's "overpowered" without considering how it functions outside of that very specific situation. The same happens in reverse where people will look at a weapon or skill meant for use in specific situations, and decide that because it's not useful at all because it isn't as useful in general gameplay.
This is where the trickiest part of balance talk comes in, in my opinion, because if developers listen to the extremes on one end or another, without really having a look at things, they risk making an already balanced feature unbalanced because the community was too focused on using it in an unintended way. This can easily lead to something that was previously balanced becoming too good because now it's good everywhere on top of its niche, or becoming too useless because it was previously only performing well in its niche.
I've seen this a lot in particular with fan made mods for games, where they attempt to balance a feature or skill from the base game that was previously overpowered in a specific setup that most players would only use if they watched a guide to make that setup. The mod creators focus so much on keeping that extreme from happening that they make the components of said extreme useless outside of it.
I had a strangely introspective relationship with League of Legends, back in the day. Waaay back in the day, Udyr jungle was overpowered, but Trundle jungle was kinda mid-tier. On Udyr you could buy all the gold-per-minute items and still stomp until mid-game. And he synergized with everyything you would eventually built them into. Trundle was more item dependent, slower to launch, and basically worse in every way. Until late game.
But, I won more games playing Trundle. Because I was actually consciously or unconsciously aware that I was not unstoppable, and so I took fewer risks, played more optimally, and would more routinely actually make it to late-game with more objectives. The perception of power made me a worse player, and my brother would beg me not to play Udyr, even when I could routinely three-buff any other jungler under the sun, and starve them out, I took too many risks.
Essentially playing the weaker jungler required me to play in un-punishable ways. And when there's five pairs of eyes on the other team, they will spot punishable behavior eventually.
Im not super much into multiplayer games. But it is a constant battle to get a game balanced. There will always be a group of gamers in every multiplayer game that wants advantages. and if they find out a particular character is better they will abuse it. We will always be one step behind people that want to get advantages.
The hardest part about balance? It shifts with player knowledge. Things become good, then bad, then good again for no reason other than understanding use cases. You end up not knowing which part of the discovery journey any single feedback is on
Don't forget to join us at Noon CT for the premiere of our new game developer hosted podcast, Dev Heads! czcams.com/users/liveOs-gt-15XKM?si=rOLNoJk4RrHj3UwX Hosted by JM8, Tina Sanzhez and Michael "Mikee" Tsarouhas.
A big thank you to @MortdogTFT for joining us in the Design Mines for this episode! It was a pleasure to peer behind the curtain on how TFT is balanced. Somehow even more chaotic than i thought!
I’m a musician and I normally use these videos to help me think in new ways.
Art is art.
Not sure how I can use this in another medium, but it was SUPER interesting.
Loved it.
Game I've never heard of in my life has untold millions of players has become a daily thing
I would say that the statement "There are no balanced games" is less about balance being impossible, and more about managing your expectations as a player to avoid the "scrub" mentality. There's a lovely article excerpted from 'Playing to Win' describing the mentality, but a common refrain is "That's not fair, it's unbalanced" from players who are hung up on their own ideas of how a game should be played instead of playing the game as designed and trying to find counter-tactics.
Yep. Balanced is a kind of platonic ideal that doesn't exist in this reality, the job of a game dev is to make sure that it's balanced enough that player skill is the controlling factor in a win or loss(Which IIRC Sirlin covers by saying if a game truly, factually is that unbalanced, it's probably not being played seriously, i.e. to win). Maybe there's a realistic question of whether Cammy can beat Akuma even with a huge mismatch in skill, but a roster fighter can just shrug that off if there are at least 3 viable options at the top, especially if you allow a loser to counterpick to potentially recover
hell yeah, my favorite dev as a guest at my favorite videogame design video series!
Even single-player balance is a TON of work in a complex system. I've developed comprehensive balance mods for FTL and Brotato which each took hundreds of hours, and that's only *after* the hundreds of hours the devs spent making the vanilla experiences strong to begin with.
I remember when both me and a dev team got humbled. I was playing Smite when the game was in Beta and they'd just put Loki in the game and he was dominating every game because he could go invisible and you couldn't detect him and at the time did 2x backstab damage+ extra damage coming out of stealth. I argued he was too powerful and part of the issue was the invisibility with no way to detect and either the game needed a way to detect invisible things (like Dota 2 has as the main Rikki counter) or Loki to be seriously nerfed. With some info the community had I worked out some beermat maths to work out how Loki was melting everyone even tanks and proved his burst damage was already above any other assassin. A dev stepped in to explain my calculations were wrong and due to how armour works and Loki's blades not getting armour piercing on special attacks or attacks from steal the damage stacking I was calculating was way off. I argued something seemed off even if my numbers were way off. 3 days later the same dev came back to the thread having looked at the Loki win / loss numbers and said there was huge evidence of Loki winning a high % and having looked into it we were both wrong. Turns out some-one had forgotten to code Loki's damage correctly so every attack was doing pure damage that no armour could mitigate and they'd have to try and fix this then go over all the data again.
I would love for robin walker to come on to second wind. I love his philosophy on balancing for pubs specifically.
In an interview on a a now defunct gaming website. he said one of the reasons he added random crits was that people would blame the game and not their under performing teammates. (alongside the other reasons). People go "Fuck that bullshit crit" when they die instead of "fuck you medic" when they die
Randomness can really shift perspective really. although that is a double edged sword really.
@@Sniperbear13 yes. Random crits are hated by the tf2 community. But in my opinion they just need to be reworked.
I personally see randomness when done right as a necessary evil.
Stuff like mtg for example is made fun due to the RNG. And cards can be pushed to be more interesting because both the resource (mana) is randomized and so is the resource type. Since all cards are in the deck
In more "balanced" games like hearthstone, cards are limited by the fact that mana is guaranteed every game at the same rate
@@abrenmam hence why its a double edged sword. Randomness can remove some skill factor and just screw people over, or make things more exciting and entertaining.
having a really good deck in MTG might not matter if you just cant draw Mana or just keep drawing nothing but Mana. while there is ways to Mitigate it, its still a crap shoot.
Yeah that checks out for TF2. It's a very casual and chaotic shooter and it just wouldn't be the same without its elements. Random crits could possibly be adjusted away from being positive reinforcement loops though, your critrate goes up with damage dealt.
After years of playing D&D, I eventually realized players don't want balance. They want controlled imbalance.
And the perfect analogy is Jenga. The balance exists only to set up that thrill of seeing the tower fall over.
They want fun. Sometimes that fun is being able to steamroll everything. But I think DD has talked about the fact you want players to have some struggle to overcome, but player has to feel powerful. Games are mostly power fantasies, except when they are designed around clear challenges.
Perception is a wild thing. I remember a while back, Destiny 2 started to move away from Connection Based Matchmaking in casual PvP playlists to Skill Based Matchmaking, after being CBMM for like 2 or 3 years. A lot of higher-skill people were up in arms because they didn't want to play in sweatier lobbies, but nevertheless the update went through. And the socials were filled with people complaining that the new matchmaking was horrible, lobbies were awful now, and Crucible was ruined. But there's the kicker- we found out later that week that due to an oopsie, the change never went live- it was still the same CBMM it always was. Some PvP-focused content creators like Aztecross then went and tried to retroactively change the narrative to save face, which was equally hilarious and infuriating.
Another time some Exotic boots got nerfed shortly after it was released and was found to be super powerful, and the community got super upset cuz they misunderstood a poorly worded dev communication about why they were nerfed. A month or two later they not only reverted the nerf, they buffed them stronger than they first were, and NOBODY used them because the public perception was still "They were nerfed into the ground."
That matchmaking bit is crazy; didn't know about that. Thank you.
11:35 *ProtonJon laughs maniacally*
Back when I was into Magic the Gathering, I remember reading an article claiming some imbalance is a good thing. If everything is perfectly balanced then the game is too homogenized and lacks flare. It's also nearly impossible to avoid- some players are just too dedicated, devoting thousands of hours to finding that perfect unbeatable combination. Not sure what's changed since I stopped playing DOTA 2 but I remember feeling like Rikimaru was far too OP in expert hands.
Imbalance is always going to be a thing as long as skill expression is a thing. And the more types of skill your game allows, the more imbalance there is, but the more *game* there is to the game as well
I think balance feels much better if the way it will affect the game is communicated well to the player.
Let's say I'm fighting a character who could kill me in 6 hits before, but they got buffed. I've heard "They got a damage buff" and "They're A tier now", but I know I can win the fight before they can land 5 hits, so I take the fight, they kill me in 4 hits, and I lose the game.
That's frustrating, because I made the correct decision with the best knowledge I had, but the knowledge I had was too vague. And the average person who feels like they should've won, will feel like it's the game's fault when they don't win, so they'll blame the balance
If I had instead heard "They can kill you in 4 hits now, be careful", I would've known to not take the fight head-on, I shouldn't win that fight any more so it's my fault if I take it. If the opponent's character really needed that buff, then it's good balance, I just need to adapt to it
This is an extremely basic example, just to get the point across, and I know it's not that simple to do irl, but surely we can do better than tierlists
Hell yeah you actually got Mortdog on. Stick figure and everything.
One thing interesting about game balance is how the meta is not necessarily the best way to play, just the agreed best way to play.
In Counter-Strike, there was a gun that was so incredibly powerful for several years. This gun, the SG 553, was completely ignored by the majority of the player base and the entirety of the professional scene because the meta was its alternative, the AK47. And everyone knows how to use the AK.
It wasn't until the AK was nerfed for a week or so and people started using the SG that people found out it was really good. It was so good in fact that when the AK nerf was reverted, people would prefer using the SG. It was used so much that it ended up being nerfed.
It's also funny how, if you used the SG before it became meta, people would mock you for using it. Imagine that, players mocking players for using one of the most powerful gun in the game just because no one knew it was good.
I have noticed this a lot also, mostly with valve games as well over also overwatch when it was first released, that the meta absolutely dominates on what gets changed and 90% of the time the chance is just based on what some ivory tower e-sports players say, nerfing things that threaten their meta and buffing things that support their ideal of the game when most people do not play like that.
My favourite example of balancing by perception is that in the early days of Splash Damage's Enemy Territory multiplayer only expansion for Return to Castle Wolfenstein, the community was insistent that the Allies' Thompson machine gun was stronger than the Axis' MP40.
Except the weapons were identical in numerical balance for damage, fire rate, reload time, recoil handling etc.
The solution? They made the Thompson sound worse.
I'm not a game dev or even aspire to be one, but I absolutely love picking out trends and analyzing raw data points in pvp gaming spaces. In apex, they released Horizon and Valkyrie around the same time, both to relatively lukewarm reception. Both considered fun, but not overwhelmingly strong. I picked out immediately that both of them could simply leave a fight vertically, without the help of any structure to attach to and no vulnerability to shoot at them if done correctly. It was a whole year before the greater playerbase caught on to this and both characters rose to meta status before eventual nerfs despite there being almost no changes to either in the year leading up to it. In overwatch, I have for the last 2 years thought it absurd that almost every character's critical multiplier was 2x or greater as that meant that there is a huge disparity in the amount of damage you could be taking from the same number of shots making the single tank format feel awful becuase you could be doing just fine and then be dead almost instantly by taking a few headshots from the enemy in unison. Lo and behold, Overwatch has just announced a tank rework that decreases their headshot damage taken to level out time to death. There's a a few other examples, but mostly my point is that public perception of the power of something or public perception of what the problem is in the first place can really affect what is being played. Players who find effective strategies through personal data analysis rather than crowdsourcing the info from playrates and content creators can use said strategies for years being called "a specialist" becuase no one with a following has tried to replicate it out of a generally accepted feeling that it wouldn't be more broadly applicable.
There is also this perception that, because games are usually more fun if the player wins, and ranking is more fun if you are making gains, people just want to win more than they lose. And that is something that just doesn't happen - you either get to become the best and have nowhere else to overcome, or you aren't the best and will stay behind someone better than you. If a game ranking system is actually balanced, every player will rise up to a point where you will win exactly 50% of the matches. However, if every other game one loses it, the person will feel bad, stuck and miserable. And even then, if you try and make stuff like faking games to give a confidence shot and make everyone get 50%+ games won by getting bots to fail against the players, if they ever find out (and they WILL find out), players will feel outraged by it.
Loving the interviews with actual devs. You get some good context and knowledge. Keep it up.
Missed opportunity to tell another interesting balance story that partly involved player perception and partly game theory mishaps.
In set 9, they introduced a mechanic called "Legends", which was a character you could pick before going into the game that would guarantee you to see a certain set of augments themed around a playstyle. The results were kind of mixed. It helped newer/more casual players get to narrow what they needed to learn to play and have fun, but it reduced variety for people who wanted that. Ultimately they removed it after the set.
This story is about one of the legends: Draven. His augment set was all about being rewarded for playing aggressively, with the first augment he always offered being "Spoils of War." This game some % chance on kill to drop a loot orb. You play this, you spend your money aggressively in the early stages to win-streak and earn money back from the augment.
This was viewed as underpowered for most of the set, so it saw little play. So one patch, they buffed it to drop the orb slightly more often. Basically right away, streamers and others were saying it was OP, so a lot of people started playing it. What made this worse was the fact that it turns out that the more people in the lobby who played Draven, the better it was for you to play Draven. That's because:
a) If the lobby is aggressive and you're not, you'll lose too much HP and die.
b) If people are playing stronger boards, that often means they're running more units. Since you have the chance to get loot orbs on unit kills, the more units the enemy has, the more potential money you get by winning.
c) With all this money running around, a lot of the lobbies would end with people getting 3* 4 costs, which are ordinarily supposed to be super rare and give you an overwhelming likelihood of winning if hit. Because units get taken out of a pool of units of the same cost, the more people taking 4 costs out of the pool, the easier it becomes to hit other 4 costs. This particular problem is a more general problem TFT faces that Mort's talked about.
So because of perception and game theory, within a few hours a lot of lobbies had 8 Draven players. People freaked the fuck out, because they always do, and the team fixed it literally the next day, like they usually do.
But what's the funniest part of it? Just like with the Vlad story, the change didn't actually properly go through. Only the text was different. This entire catastrophe of a day was essentially entirely created by player perception, amplified by the mob mentality on streams and reddit.
I think an interesting other aspect to balance is unfun METAs and the difference between this is FUN and this is STRONG. I don't have any idea how to balance for video games, but I've been playing D&D for sixteen years and writing my own content for it. The 5e ranger is the quintessential example here IMO because if you play a very specific optimized build it is very strong, but its then one very specific course of action from there out. You summon animals and use a fully optimized set of hand crossbows. It's both unfun because you never make any other choices and the DM has to deal with 10 badgers eating up the turn order. Making ranger strong hasn't fixed many players problems with ranger because they want the class to be fun and engaging.
Martials (classes with limited magic or no magic at all) in D&D overall have had this issue in 5e where there simply aren't many choices to be made and at its worst its really split the community even if they're "balanced" around certain numbers like damage dealt and how much damage they can take (which even then those numbers are very easy to skew).
Of course all of this comes with that huge grain of salt called personal taste, but its still interesting simply because of how long its held on in 5e.
Not a multiplayer example, but my personal example is the Mage and Duelist in 7th Dragon III Code VFD. The Mage is way better than the Duelist, not even counting the support and healing, it does tons of Elemental damage, and status effects reliably, while the Duelist, using 3 elemental cards is more random. But that's what makes it more fun too. It makes battle more exciting, and, once in a Blue Moon, when you can set up the special combo for Massive Damage, it's absolutely amazing. It's all about the Risk/Reward, even if it may not always be worth it. I kept a Duelist in my team until the end of the game, because it just felt better to play.
This was my issue when I first played 4th I think. Every class got at-will skills, basically making "regular" attacks pointless. You had to get creative with a skill shot or something left field to have any fun, otherwise it was always just use the at-will because it's based damage or base att +2 or whatever.
Ruined a few years of DND for me.
Ranger is also interesting because it's state in 5e is a response to it's performance in 4e aka it was absurdly busted. 4e strikers (damage focused classes) were separated in power based on how well they used their minor actions (5e bonus actions) for damage. Ranger was by far and away the best striker at turning minor actions into damage on top of just having really good standard actions (hi Blade Cascade aka 10d8+5*Strength+15+5*what ever else you have on hit). This led to it's release state in 5e where it had been nerfed to the ground despite being in an edition that didn't remotely resemble the edition before it.
@@Micky96 as a 4e fan, I can acknowledge ranger as op, while also adoring that they made ranger actually live up to the fantasy and gave players enough options to be the ranger they wanted to be.
The addition of the seeker class also let ranger stick to a more martial focus while seeker mixed option. Variety is the spice of life.
@@Jasonwolf1495 Yeah, I mostly just brought it up as an example of how balancing is informed by previous versions. I'm a 4e fan myself (playing a lvl 24 Swordmage rn actually) and I'm actually eyeing playing a Ranger next campaign because I secretly still kind of like Drizzit. Other options I'm eyeing are Warlord and Sorcerer.
Still don't know what "TFT" stands for, at first I thought it was Team Fortress Two.
The Helldivers community has been very critical of the balancing happening in that game currently. Hopefully this sheds some light into what the some devs might be thinking.
Frost did excellent video on this topic. It is mostly players not understanding initial balance and being annoyed that weapon being OP before is just normal now. Again, perception is harsh mistress.
One thing not in this video - loud voices vs silent majorities in community, or plane with holes for player feedback - be sure to know that there are people who not gonna provide feedback because they either drop off or are totally fine with game.
Helldivers certainly came to mind for me too. It feels like the community goes into overdrive about any minor change or (in the case of their recent content release) if things aren’t necessarily as strong as they wanted.
The behaviour of some of their PR isn’t helping and their overall QA is a bit lacking (I feel terrible for their CEO, who seems to be a community firefighter alongside his normal work), but it certainly is important to remember how complex the internal structure of a PvE shooter is.
A similar game, Deep Rock Galactic, has similar balancing issues but tempers it by letting players express themselves via weapon modification (upgrades and “overclocks” which can radically alter weapon function). It makes the loadout more personal and a mark of self-expression in the game. Thus, players are often more interested in finding out which combinations work “for them” than finding the meta defining option.
I used to help run a LARP that was also constantly under design as we tried to balance it and also broaden the kinds of fantasies it was possible to play.
One of our biggest issues with game balance is that we were trying to balance two games: PvE was an entirely different game from PvP. The other issue with trying to balance the game is that instead of giving the designers unilateral control over the game, any changes had to be voted in by the players (an idea I had introduced very early on because I had just gotten out of three different larps where the people in charge just gave themselves arbitrarily large amounts of power).
I eventually quit the game (there was a faction of players that were trying to take over and deliberately making things awful for me to drive me out and we for some reason couldn't ban them), and a few years after the other founding members quit the whole thing collapsed, which made me sad.
Been really enjoying the Design Delve series, thanks much for doing it.
Small nitpick: 3:35 What you're describing is called the Nocebo effect - It's the opposite of the Placebo effect.
Came here to check if someone else picked up on that too. Gratz!
this is like trying to apply the phrase "reverse racism" ... it's a distinction with no actual difference what so ever
Possibly the biggest illusion of balance I've seen is in the game Heroes of Might and Magic 3, and it's extremely well executed. In that game, you control heroes that lead armies, gather resources and capture towns. There are 8 town types that produce different types of army - the Tower faction produces Magi, golems and other wizardly adjacent creatures, Necropolis summons the undead, Inferno summon demons and devils, etc. Some people claim the game is well balanced, that a good player with any of the 8 factions (or at least, a good subset thereof) will be equally capable of taking on any given challenge. They are wrong. This belief comes from the campaign, which expertly balances the difficulty of campaigns against the power of the factions such that you are presented with a suitable difficulty curve.
Several factions are close to parity, but one stands head and shoulders above the rest: Necropolis. Their heroes have an ability called Necromancy that allows them to convert 10% of the entire enemy army to skeletons upon victory (or skeleton warriors, the upgraded version, if there is no slot in the army for the regular kind). As they develop, they can all push this to 30% and even beyond with certain powerups. Skeletons are one of the better bottom tier units in the game, and the game is structured such that a swarm of low tier units can easily defeat a more modest set of high tier units. What makes this game breaking is that every map has most of its resources guarded by unaffiliated armies. These are meant to be a weak but beatable challenge that slows the rate at which you gain power...but if you play Necropolis well, most of these fights will add more to your army than you lose. So while everyone else is slowly burning through their army, Necropolis players are enhancing it beyond what their production normally alllows. And this is only half of the game breaking - the other half is the Vampire Lords. Their 4th tier unit, Vampires and the upgraded Vampire Lords are powerful and generally become available around the start of what I'd call midgame. Vampire Lords are a particularly potent 4th tier unit with the ability to fly, excellent movement range, high attack power, and do not provoke retaliation attacks (most creatures, when they attack, get attacked back by the survivors on their turn; Vampire Lords don't trigger that, so they take about half as many attacks as they dish out in a one-on-one fight.) Then, this already high power unit gets this absolutely absurd ability where for every point of damage it deals, it heals. If a vampire lord from that stack has been killed, healing past full health for the surviving stack will resurrect those dead vampires. And these things deal a LOT of damage, so they get a LOT of healing. Once a person has over 40 Vampire Lords or so, they are going to tear through most battles with minimal effort. And for every other unit in the game, Resurrection, particularly the kind of Resurrection that persists after one battle and into the next, is a late game spell cast by heroes, which takes a good chunk of their spell points and can only be done once per cycle through all the army's units - thus making it an opportunity to do a great deal more than just bring back a few dead units. Between these two units, Necropolis is already extremely potent. Add in the fact that their tier 3, 5 and 6 units are all average or above average, and you've got a force that's as unstoppable as death itself.
The way the campaign balances this is by making the Necropolis campaign absolutely brutal. Every level gives the enemy a major advantage in number of towns (and thus army production capacity, which is a fixed amount per town at the start of every week) and several levels put strong roadblocks in place to stop you from making an early rush - as rush tactics are an extremely good way to counter that sort of disadvantage as the army size discrepancy compounds every week of production. This results in memorable levels that pit you against extremely powerful enemy armies....and if you're a player who knows how to take advantage of Necropolis's advantages, you absolutely steamroll them. And if you're not...well, you might be able to pull off the Tom and Jerry strategy (bait them into attacking in a stupid way and use that to make them hurt).
Great game, though! And super cheap, because it's some old 1999 game on GoG that the world has forgotten.
I think Limbus Company had a pretty big surprise recently with it's latest update.
So, they just released two new identities, whose gimmick is "I can inflict one of the five main status ailments, but it's randomly picked which one it is."
This didn't seem very strong on release. However, it turned out that this meant that their skills counted as all status type inflictions, and their IDs counted as five of the seven status effects at the same time.
So, all the mirror dungeon gifts that rely on skills or identities that inflict an ailment? Yeah, turns out those were triggering far more often than intended. If every attack counts as burn, bleed, tremor, rupture, and sinking all at once, then they could inflict way more of it intended with the right gifts, and also count for those super "Have 5 of an Ailment on your team for a big bonus" gift.
They've been rebalanced to count as only Bleed IDs for fusion gifts, and their skills will only trigger gift effects if it specifically triggers the ailment by random choice, or would have counted anyway (For example, being an Orange attack for a gift that triggers on Orange or Burn attacks).
I personally didn't see any problem with them until I read the patch notes to fix them up.
There is currently some drama in the Helldivers community about balance, regarding recent nerfs to popular weapons, new weapons being underpowered, and how much balancing is really needed in a cooperative PVE game like Helldivers.
Main problem for.helldivers is that the patches almost always contain nerfs with small buffs occasionally sprinkled in. Feels like all the fun is being taken away
I fully agree with the Developers ideals. Unfortunately, people are jerks. One of the things I learned from LoL when I was still playing it is this. Find something neutral and learn to play the hell out of it. Skill with your kit is more important then what is Meta. On a personal level I have never liked the Eruptor, my bot weapons have been Plasma Punisher, then crossbow, while using the Senator as my secondary. While I find blitzer just DEVISTATES bugs even though it is useless killing objective items
@@klyxes It may seem boring, but balancing down has usually seen good success. Balancing up makes things more explosive and harder to compare, it also can lead to power creep.
i am of the mind set of; if there is only a single loadout players are using then it removes the point to have other options. why have 5 other guns if no one is ever gonna use them? that to me is why Balance is needed in PvE; to create more variety, and prevent a single meta from dominating.
why develop new things if no one will ever bother changing from what they have? if the new things start out Over powered AF, then people will use them but then you start to create a new problem of Power Creep.
@klyxes Every time I helldived, no matter the difficulty, everyone would bring the same loadout and if you didn't use that load out you were kicked. Granted on higher difficulty there is no need for 3 bile titans ever at the same time.
Something that's actually really interesting is competitive multiplayer games designed to be unbalanced.
While such games are unsurprisingly rare to find, a game I myself personally enjoy quite a lot is called Blood Bowl, Imagine chess with dice and punching.
In Blood Bowl there are many different factions, each essentially a different fantasy archetype, who you pick from in order to build your team. This team will generally play ~1 hour long games within the competition its participating in, out of which it can play anything from ten to hundreds of matches depending on the competition format and length, which while progressing, it can skill up its players, buy new ones, or even see them die.
However where the balance hits an interesting snag is how many ludicrous ways the very different teams are built and interact with each other. Some teams are designed in such a way that they're solid in their early stages, but lack the potential to hold up at higher stages, some teams lack many basic skills early on, but develop into terrifying menaces within time. Some teams do pretty good at all stages, and a few teams are designed to be catastrophically deficient.
However despite there being a clear meta orientation, there are so many factors, such as matchups and how a team develops over time, but most critical remains the dice factor. The top coaches (players) who will put dozens of hours per week into a completely open ladder system where you can easily get matched up against complete newbies, only have a win rate of ~70% - 75%.
The whole balance aspect and team interactions are really worth a closer look than what I can fit in here, there are many eternal lengthy discussions about many of the balance factors of the game, if you look at different tournament rules you'll see extreme differences in what different factions get or don't get depending on their tier. Even the weakest factions have their own ways of fighting back, and there have been many cases where severely deficient factions managed to outperform the better ones.
This is all exponentially increased by whatever the hell happens behind the developer's doors, since it exists in both table top and virtual format, in the latest big rule change the worst faction by far instantly became the best faction by far due a large amount of highly unconventional strategies, all mainly because the parent company changed its roster to better fit the factory miniature sprues.
Give Ludo her own side-series and call it The Ludo Narrative
And if the space between the microphone keeps changing it would be "Ludo Narrative Distances"
That's why I never play multiplayer games anymore, every time I did it was a bad experience.
My biggest frustration with perceived game design is when the devs don't communicate intent to the players. Here is a character and huge selection of equipment go have fun. Then I pick what looks fun and then find out that character was not intended to be used that way and now your stuck with a useless setup that is not fun. If you want players to perceive balance I feel you need to give at least a communicate a modicum of intent. I find when you communicate why things are designed in a way it gives players a foundation to work on. Some will go off that path if given the option and any success or fail is on the players. But not communicating makes the players perceive intent and then that focus goes on the developers and as failure is more likely when experimenting and this negativity impacts perception.
Some terms I'm surprised never come came up in this video are symmetrical and asymmetrical gameplay, because those are kind of an important covariate hanging over this whole discussion. I feel like everything being discussed here is specifically applicable to a certain window within the Symmetrical-Asymmetrical window where you're mostly talking about character classes and is going to vary wildly as you move further to one side or the other. I think that multiplayer, gameplay symmetry could easily be an episode of Design Delve.
Honestly one of the advantages of rogue-lite design is that it generally doesn't have to be finely balanced to be fun.
If I discover a serious exploit, but can only take advantage of it because the combo isn't likely to happen in more than 10% of games... it's not that bad. 10% of the time I get this badass killer combo out, and the rest I don't - and in trying to build into my killer combo, I may be passing up other good-but-not-as-crazy options to try to get my broken combo out, and then I fail because the RNG tells me to go #%@# myself and I have to make do.
Now, if I discover an exploit that lets me deterministically ACHIEVE a killer combo in a rogue-lite or similar RNG drive game, then you've got a real problem.
This guy said in a better way something that, as a fighting game player, has been tugging at my mind.
If your goal is balance for balancing sake, you might as well simply have one character. The point of balancing is to balance for fun, or the way he puts it, a desired experience. By balancing towards that goal, you can create an unbalance that many people actually love.
In my own time, I'll tinker with ideas for card games which I'll sometimes test with friends. One thing I've found over time is that it's not just *how many* levers you have at your disposal for balancing cards, but the range you can move those levers in. A game like Magic where 1 power is small and 5 power is huge has less range on that lever than say Digimon, where 1000 is tiny and 15000 is massive. Thing is, after you've locked in the scale of those levers, it's really hard to change them because it will also alter the rest of the levers you have in the game. If Magic wanted to expand the power/toughness scale, not only would it invalidate older cards, but it would also mean the starting total of 20 life would be outdated as well.
Tale as old as time. Even when a new Quake-style oldschool arena shooter comes out today (only to die in two weeks, because they're fucking brutal), you have people ingame flame others after dying the same way they did 25 years ago. "Rocket Launcher is a shit noob gun, splash damage", "Lightning Gun is a shit noob gun, too much DPS and knockback", "Railgun is a shit noob gun, you can snipe and hide".
Like... Rocket Launcher's projectiles are slow, you need to predict enemy's movement more so than with any other gun to land direct hits, splash damage is too weak compared to what other guns can do consistently and, finally, there's self-damage! Lightning Gun is mid-range, the beam is short so it doesn't hit at long range at all, and at short range aiming it is impossible, avoid that range! Railgun has the lowest DPS because it takes so long to reload after every shot (as in, it reloads instantly after you fire it and you can't even switch to another gun), chip damage can be compensated for with good positioning, movement and item pickups, and players who are too cocky with the Railgun can get punished because they can't defend themselves while it reloads. Every option has strengths and weaknesses, there's so much depth to it that people are still playing decades later with minimal tweaks.
The most important part is, there's no builds, there's no characters, no loadouts, there's so few variables to balance and they pretty much nailed it 25 years ago already. Since it's multiplayer, players can adapt and evolve. Everyone starts equal, everyone has access to the same 9 (nine!) guns, you have the freedom to *position yourself to render the enemy's preferred gun ineffective*, you just need to be better at exploiting the other guy's weaknesses than he is at exploiting yours. But new people don't want that, they'd rather resort to throwing out aforementioned absolutes or, better yet, calling players who are better at the game than them "the real noobs" because they not only don't see the whole picture, they've decided that they shouldn't have to
While talking during the end credits does mean hearing the calming music less, if the talking is as charming as this I am all for it😄💖😊
I think that what makes balancing so hard isn't so much the math associated with it, but how the "fun" aspect reacts to the balance. There was this one point in the speedrunning history of Goldeneye64 that a player accidentally discovered that you could shave a half second off your run by continuously looking down rather than forward and this nearly broke the community bc it was just not fun to play like that, but you'd never keep the first place spot without it. Fun is not just reliant on success, if winning is less fun than playing unbalanced then you have to reorganize whatever balance you imposed on the game.
I think one of the trickiest things with balance is that people use it to refer to different phenomenon. Some people use it to mean that all strategies or characters or weapons (depending on the game) are equally viable. Other people mean that they want all options to be equally enjoyable. Sometimes those are similar goals, but not always, especially since preference and skill level are going to affect both.
I've been playing the Roguelite _Astral Ascent_ a lot lately (2D platformer with a lot of Hades in its veins), and the randomness distribution thing is super well balanced there. In that game you get 4 spells (can be cast from mana which is replenished by landing hits with basic attacks), and can have up to 4 Auras (always active, may spawn helpful things on a timer and/or specific action, add passive buffs, etc).
Both of these have an elemental type (or at least affinity): Fire, Electric, Ice, Poison, or Neutral and both can be upgraded or replaced through shops, random drops, or events. You always start off a run with no auras (until you unlock the ability to choose one at the start) and 3 cost 1 level 1 neutral spells + 1 spell you can choose from.
Early in the run your choices are pretty randomly distributed (and low level) but as you get past the second area the game starts to look at what you've selected so far and nudges the RNG to give more synergies. One of the most important things is "Gambits" these are upgrades to your spells that may spawn additional damage causing objects, add debuffs, or even heal you. Selecting gambits that are of the same type as your spell improves the damage output of the spell directly. An example of a good synergy might be that your poison spell that hits 3-5 times has a gambit that has a 40% chance to spawn a poisonous slime on hit and an aura that increases the damage that a slime will do to enemies, or how easily it applies the poisoned debuff to them.
This system came about because the early patches didn't do this and many runs died because you weren't able to synergize in any meaningful way except through sheer random chance. It's kinda the same way as the Shen augment discussed in the video, If I get lucky in the starting room and get a level 3 Aura that has a 15% chance of healing 1HP on missile hit, but can't find anything that spawns missiles I wasted an aura slot, but even finding one gambit somewhere in the run suddenly makes it a lot more viable.
Astral Ascent is fantastic. I'm sure it took them a lot of effort to get to where they are now, but I've been playing it recently and it feels like a dream. I hope more people find out about it!
Hi Mort! Pre-alpha tester for Riot here! You don't remember me!
Really cool getting your input here. Second Wind is popping off.
Helldivers 2 comes to mind. Balancing wise the game is incredibly delicate as its attrition-style gameplay loop amplifies imbalance in a "Win/Lose-more" way, its like L4D on that. You are to have a fighting chance, not be a horde slaying god. Which means that the game is more likely to nerf something than buff and boy do players complain about that!
Some in the community are still demanding their enemy trivializing Godgun is returned to the exact state it was before it got nerfed and that instead they should just "Buff everything else", oblivious to how much work that would be and what kind of problems it generates in the long run. So now they call everything in game "Mediocre" repeatedly while the odds are very much that if they never had a Railgun power-trip. They wouldn't have been spoiled and be able to enjoy the rest moreso..
As a fighting game player, one thing that is super clear is that the initial impressions of players are very likely to be wrong, so balance should probably just sit for a while. It has happened many times that a poorly rated character has turned out to just be the best or other way around. Also frequent balance changes reward players for jumping into the the thing that is hard to deal with right now, while less frequent changes reward players for investing more deeply into more complex parts of the game. You probably don't want to make changes too infrequent, but the latter way should be possible imo.
Absolutely. The recent Path of Exile patch had this issue, where the devs didn't wait even a single week for their new content patch to sit before they buffed it. The result was one of the craziest imbalanced leagues I've ever been a part of. The playerbase likely would have found good uses for what was deemed as 'underpowered' given enough time. But they just reactionarily buffed everything past the point of no return.
About the balance perception misconception I'm reminded of Killer Instinct and how the devs ended up balancing a character(I think it was Jago) in a different way the community was suggesting, and everyone got mad at then. But then it turned out the nerfs were really good and they understood what made the character bettter or worse more than the community did, so the devs sticking to their guns like Mortdog said is very true
The only kind of content that can make starting the weekend even better
A perfectly balanced game with no exploits is a rare thing indeed.
If you want to see a ongoing non-video game version of this: Yu-Gi-Oh. Lack of resource management and stupidly complex interaction mechanics means a trash card from a decade ago can suddenly become game-breaking because a new support card for a completely different archetype allows the older one to infinite loop into a FTK e.g. Frog Burn decks
I would stand by the idea that no game is balanced, if by 'balanced' one means perfectly balanced. The trick is that you never really need perfect balance either, you just need enough balance. The first priority I think is making sure there are no dominant strategies, things that must be taken for winning to be possible (because your opponents will be taking them even if you don't). Make sure everything has a counter, make those counters reasonably accessible, and you don't need perfectly matching numbers, players will sort out the rest.
From there it's worth looking at some second order goals, like making sure no option is too aggravating to play against or only has counters that are annoying to use, trying to make all (or at least as many as you can) of options a player can take become either one of the strong options or a useful counter against them, and how differing skill levels affect balance considerations (and what to do about them).
Perception is also important to consider though, and I actually think it's the tougher nut to crack. Not only can it come from outside the mechanics, things like visual effects, sound design and UI elements can affect it, but it can come from outside the game entirely.
Really great interview! It's interesting how a lot balance really just comes down to perception.
Cool addition to the format. You have made another great video.
It's always interesting to me to find points where the meta is both un-fun and incorrect. Friend of mine used to be very into the great-weapon-master in Neverwinter. Because he could see that this "garbage" class was just specifically great at destroying adds in certain boss fights. So he was topping DPS charts, as rogues are single-target. Didn't even replace rogues as highly necessary boss-shankers. But it really made them mad because they were very used to measuring self-worth by pure damage numbers.
MtG, when I had a program to help me play it on Arena (I need something to beep at me when it's my turn, because I can't both relax and pay attention, it's just where my ADHD is.) I loved building decks that were not optimal, but were fun and wrecked metas. Like the days of the cat-oven deck. Cheap little 2-card combo, just involved an etb trigger and a cat jumping in and out of the graveyard. And the main counter-play was a -2/-2 to all creatures instant that, at end-of-turn, deleted whatever it put in the graveyard if it was still there. And on Arena, you had to sit through every slow step of activating the cat, paying its price, activating the oven, selecting the cat, repeat... and it did *one* point of damage per full sequence. No fun to play, no fun to play against. So I built a deck with a lot of cards that people had dismissed from the then-recent sets. I wanted exile effects and things that capitalize on it. The idea was to remove a lot of threats and then ult a planeswalker about it. The big scary threat let you plop a couple of choice things you'd exiled onto the board under your control. But the secret powerhouse was a 3-mana planeswalker that could ult for the # of cards in exile, and +1 & lifegain off removing creatures from the opponent's graveyard... and -1 to exile a zero or one-cost non-land permanent. The cat was what kills you in the cat-oven deck, but the *oven* is the combo piece that can't change zones.
Another fun time was in Brawl mode when the rage was, weirdly, Field of the Dead. You could spot the decks on sight because they'd always have this robot commander that let you put any land from your deck on the board when you played him. Which was silly, because his other ability was the game-winner. These guys weren't even building real landfall decks, and they badly wanted one to three 2/2 zombie tokens per turn. So I showed 'em Assassin's Trophy, a very cheap removal card with the unusual text "destroy target permanent", with the downside that the opponent gets to replace it with a basic land from their deck. Field of the dead requires 7 lands with different names. These decks tended to have allllll the 2-color lands and triomes and whatnot... and some forgot that basic lands still count, so they didn't have any. (Sometimes, the popular meta is just... bad.)
And now? Helldivers 2 is the land of buffs and nerfs and... I'm weird, I remember stuff (not always). So I'll see a weapon that got nerfed and go "okay, 350 to 325, still up from the 250 it was before. Probably takes those guys I needed 5 shots to drop up to 6. No problem." And... that the railgun got armor penetration buffed again, and I think "great! people should be happy about that." but no, they were not, they'd already forgotten railguns ever existed. And... like... this isn't a game about balance. It's about running for your life across a warzone where you are extremely outnumbered except for your orbital strikes. (And satire of the 'murrican military/patriotism propaganda machines.) You've got 5 backup clones like it's Paranoia. Honestly, if they wanted to randomize the weapon stats every round, I think that'd be pretty fitting. Balance is so weird a concept.
If I recall correctly, back during the Halo Bungie days, there were a lot of players who flamed forums and raked Bungie over the coals for balance issues. And yeah, Reach especially had a lot of them (most of the Halo games did in spades).
But talking about "perceived balance" and the placebo effect, it was on display there hard. One incident I recall was players demanding a change. Bungie released a patch saying they had. Players praised it, saying it was perfect.
A week or two later, Bungie revealed that they'd actually done the *opposite* of what players had asked for, and made a comment to the effect of 'players know something is wrong, but actually *why* is generally completely wrong.'
Stuck with me.
"Perception is reality Ludo" I find that fascinating too JM8 because that's something Dr. Maruki's therapy sessions in Persona 5 talk about
I've never personally had a good sense of balance when it comes to games, but witnessing conversations about balance has spawned this observation: "Ubalanced" tends to mean "doesn't play the way I'm used to" or "...the way I want it to".
Thanks for the discussion.
Anytime your game isn't diametrically equal, it will always be a game of subtle advantage and disadvantage; and advantage and disadvantage will EVENTUALLY be solved in layers. It's the exploration phase - the breadth of that phase - that really make for especially good PvP games.
That was one heck of meaty episode 😊 Also I love this topic because it is combination of how you want your game to be played, players playing it and their perception aaaand actual balance, pseido random, rule set based or not
Crazy how Super Smash Bros Melee kind of works despite having no intention to balance and no patches.
no items, fox only, final destination.
If you ignore the majority of the roster I suppose...
Good balancing and good matchmaking are the heart and soul of a good multiplayer game. And thus they are the hardest bits to do. Scissors will think rock is OP, every time. And some fans will shout their support for scissors from the mountaintop. Balancing isn't just a matter of whats in the game, but managing the frothing rage of toxic players who WANT a game to be unbalanced, too.
In order to balance a game correctly, you have to choose WHO to piss off and when. You can't please everyone and shouldn't try to.
Sometimes, the best games are the ones which piss off a lot of people, but they piss off the RIGHT people at the right times. And that can improve a game.
In my experience; a dev needs to have a solid grip on exactly what decisions they want players to be able to make, and the consequences of every possible decisions they can make. If you know exactly what a player has to do, think, and be in order to succeed, then you can balance your game around those expectations.... Just be ready to see players do something you never would have guessed in a million years. Players are good at that.
Thus, balancing will give you a freaking brain cramp.
as someone who has been thinking of making game (far more thinking then learning), this has kind of reinforced an idea I've had for a while, that being, the feel of a game is more important then parts of the game itself, it doesn't matter if "this item is absolutely broken" if it feels like absolutely nothing, or more importantly, "this enemy feels broken/annoying to fight" even if the enemy's abilities are on par, if not worse than yours, and I feel as though these have been ignored more often.
I have played several games as of recent where the opposition feels, annoying. Yes, their hard to defeat. Yes, Its intended. Yes, what I do against them has counters their using. But none of these feel like "a hard challange to overcome" or "something that needs a new idea" It feels like a slog, a slog designed to wear you down, and wait, for every move you do, and to stop it in the least gratifying way possible, because its the best option.
Ludo is a great DAWG love her to bits!
Another approach to balance is to make everything OP, that way players all get to do some game breaking thing. The end result can be quite chaotic. It's a technique used in modern boardgame design, I'm thinking particularly about a designer called Carl Chudyk, most famous for "Glory to Rome"
This reminds me of all the gamers on Reddit or whatever that are always saying that balancing things or making good games is so simple, devs should just listen to the fans, and them in particular.
Thank you for this. It was actually insightful and lightly technical. Really appreciate the deeper content! Glad to support this kind of video!
right now design delve is the best show on second wind, the production value man
This is a good format. Every gamedesign channel eventually turn into soapboxing, the writer endlessly repeating their own biases. Getting people in from different parts of the field to talk about their experience would completely subvert that!
It's fascinating to hear another POV on design and I hope you're able to bring in many more. Good luck on the new 'cast.
Mortdog is honestly a GOAT. TFT is the best auto battler if you have the time for it. Used to play it nonstop during the pandemic and it's generally a balanced game with tons of highs and lows. It can't be understated how important his (and his team's) transparency about the game have helped the community. Want to pick it back up, but to do so actually takes a bit of time to learn anything, and I am short on that free time.
This series is so criminally under viewed. I don't know how channels like GMTK and Razbuten pull in so many more views than design delve. Tell your friends about this show!
The most important part of balance that rarely gets seen when developers collect balance data is how fun it feels to lose. I feel that's the most important part of balance in any game. People will forgive a completely imbalanced game if they enjoyed playing even if they lost, and people will hyperinflate even minor imbalances if losing feels bad. The Hearthstone devs were notoriously bad at this. Over and over they'd trot out how their "internal data" showed win rates were balanced and they were happy with the state of the game while playing certain decks and playing against certain decks felt like utter trash.
Game length and game length after victory has essentially been decided (how long does the death spiral take) both play into this.
Love the Ludo thinking Mortdog was gonna be a dog. I'm a sucker for that cute dog humor.
Would love to see if a balancer for games like DOTA, OW2, LOL, and other team basedgames would chat. They must be much harder to balance and its be really interesting to see how well they can actually anticipate stuff.
Also games like SC2 or the upcoming Frost Giant game (forgot the name) I think either of these games balance people would love to talk. Especially Frost Giant since they're trying to launch a new rts game.
I love watching SC2 pro play and find the balance interesting, even if playing the game is easy beyond me.
I really enjoy these thought-provoking episodes! Keep up the great work!
It's funny this comes up, just today I was praising the balance of Eve Online. My buddy just laughed at me and said "What do you expect? They've had 20 years to perfect it." That may be the case, but it is still quite the feat and I remain impressed.
One thing I hate is how "high risk, high reward" is balanced. My personality gravitates towards strategy that involves high risk high reward and they are usually very fun to play. For example, the AWP in CS cost almost double the AK, but gives you a chance to kill from long range. Or invoker in Dota 2 requires you to manage 10 spells, as opposed to the average 4 spells, but it also offers much more versatility. However, when subset of players who are dedicated towards the game gain mastery over them, these people basically skilled away the risk, and makes it a low risk high reward, and it can make the experience unfun for other players. That is when balancing comes and nerfing it to the ground, but now it makes the people playing more casually realizing that it is now "high risk, same reward" instead, which takes away a lot of the fun from the game, because people are penalized for being riskier. It can be even worse if they are power-crept, since it makes them "high risk, low reward" (elementalist weaver from GW2 comes to mind for me).
Everyone thinks balance is perfect equilibrium. Its not. Balance in multiplayer game design is like a see-saw; it fluctuates up and down but never falls completely and never stops
I still have no idea what this game was supposed to be.
Ah, posted one minute ago. New JM8 and Ludo content on a Friday, yes please
Edit: Guest appearance in this video is beautiful!
Felt slightly like we where jumping between topics without linking. So having the podcast explains it.