Models are sucking the fun out of fun
wordimagedesign.com
wordimagedesign.com
Once people figure out how to min/max for statistically optimal results, it becomes less about fun and novelty and more about how much time you have to study these strategies.
I completely understand how this "solving" of the game can be incredibly fun and rewarding to some people, but I personally don't have the time or motivation to dedicate. I like to explore and play by ear. As such, I have zero chance in anything competitive/multiplayer.
Not really complaining, just something I realized isn't up my alley a long time ago.
During the original March 2020 lockdowns I picked up Classic and had a great time enjoying a style of game that doesn’t seem to really exist anymore. Until I got to higher levels and started to realize everyone was just micro-optimizing the hell out of everything, and kind of aggressive about it (people actually private messaged me mocking the fact that I played a Human instead of Gnome because Gnomes were mathematically a tiny bit better).
There was none of the mystery or excitement from the first time around, just widespread obsession with getting “best in slot” gear and the perfect combination of other tricks to make everything easy.
I still had a lot of fun mostly staying away from this attitude where possible, but when a significant number of players in a multiplayer game are operating like that it bleeds into many areas of the game, even if you try to avoid it.
How long have you played? Ever since the Flea was introduced, there has been a consistent meta that largley ruins the game. That meta might change through various wipes, but there has always been a preferred way to play.
Flea and meta: the flea is a market between players instead of only having npc stores. It changed the way the game works, as people with a lot of money can always have meta loadouts. meta in this context refers to the best performing configuration of items, so the meta is x weapon with y ammo etc. (meta is a common term: for example in overwatch the various heroed are dubbed in the meta or out of the meta depending on the latest game mechanics updates)
Later I took a course with an ACBL life master, and learned the complex and completely artificial bidding systems and other ways to communicate every possible bit of information through quirky and limited channels. I guess some people find that fun, but to me it was mechanical and WORK. The game was less of a game.
because mastery and min maxing like this is at the basis of every form of competition.
even all the traditional physical sports are min maxed to the same extents. in fact the strategy and novelty aspect there is typically so small compared to video games (due to the inherent lower complexity of the sports) that in most sports you just start having a big focus on min maxing your physical capabilities much earlier than you would for example start to grind out and improve your macro speed and capabilities in a game like Starcraft 2.
But at that point they’re jobs or obsessions. They’re something you do because you must, not because you enjoy it.
Not in most rec leagues.
Nope -- it was speedrunshit. Basically, Guy A found a glitch that let him shave fractions of a second off, and then guy B found another glitch that let him shave a few more fractions of a second off, et weary cetera. The big deal about World 4-2 is that in Super Mario Bros., where you go when you go down a pipe or climb a vine is selected by the x-coordinate of the screen scroll position, so if there were a way to run past the vine that leads to the 6-7-8 warp zone and go down the pipe immediately following without scrolling the screen to the right far enough to change the pipe destination to the coin room, Mario will go to the warp zone area instead without the time-consuming vine animation, saving a few seconds. But this is only achievable by jumping backwards in a few frame-perfect jumps throughout the level leading up to that pipe, which in turn must be speed-optimized to shave a few milliseconds off so you can beat the other guy's Any% world record and... arrrrrgh. That's what World 4-2 was to these people. Not a little segment of a fantasy world. Not even a beautiful artifact of human art and design and how it communicates with the player, the messages the design itself shares with the person operating the game. Just a set of glitches and hacks to exploit to get through the game faster, and the history of the level was the history of finding these glitches to decrease the world-record time. How basic is it that I can remember discovering the 4-2 vine and my mind was blown at the revelation of this little secret!
Just today I saw another germane YouTube video called "Dear Developers, Stop Listening to Pros", ( https://www.youtube.com/watch?v=yFFpJW0j5z0 )which posits that professional players don't know how to make a fun game because they are concerned with minute details of mechanics and balance, and the vast majority of a game's potential audience want a fun experience in single- or multiplayer. So if you really laser in on the pro-gamer experience, you're creating a niche experience that won't be popular in the wider market -- and hardcore players will turn literally anything (Minecraft, Roblox, Excel) into an esport anyway.
That has less to do with anything the game/community has specifically done and more to do with people always looking for “easy” answers to the problems they see — blame the loss on balance, on team members, on lack of minmaxing, on luck, on people not doing their jobs, etc.
And they yell those answers at you because you’re playing a team game where they depend on your success for their success.
In the end, you need a thick skin to play effectively and actually get better, to pierce through the noise and rabble
And more importantly, at lower levels of play the meta is generally far less important than fundamental mechanics in terms of actually winning consistently.
If people want to do that, that’s fine, but I haven’t found many games that haven’t been stripped down to the purely technical, almost “speedrun-esque” competition. Really no fun, because treating the game like some sort of equation to uncover and master totally removes any sense of reality and the ability to immerse yourself.
A lot of the current players are studying Game-Theoretic-Optimal (GTO) models of gameplay, and memorising range tables and bet sizes based not just on your cards, but also on where you are sat at the table in relation to the dealer, and the size of your chip stack.[0]
I know that Poker is a game of skill, and the more you put into understanding the game, the more you're going to get out - but this just seems like... a lot of work.
Is it really making the game more fun?
It's kind of weird, because watching Rounders from back in the day, you'd see a lot of people express respect for Knish (John Turturro) and his attitude towards grinding it out, playing careful odds, treating Poker as a job. And if the poker scene is filled with people "playing the man, not the cards" and learning how to make good reads, and, well, dreamers, then having someone like Knish around makes for an interesting contrast.
But if everyone plays like Knish? Where's the fun in that?
Sedol commented at the time that although he appreciated AlphaGo's strength, he never wanted to play it again and I agree. It's not even really that it sucks the fun out of it, it's that the decisions the AI makes are often inscrutable and counterintuitive. It removes the traditional feel of Go being a negotiation.
At that point it's not about work or fun. It's about winning. Same thing is happening at companies - it's not about great product, it's about maximizing profit. Winning and money are obvious reasonable things to strive for, so I'm not sure how we can refocus either of these areas on QoL.
Its fun when you are the only one with an edge and winning. Its no longer fun when everyone has the same edge.
Goes for startups, college admissions, SEO/marketing, etc.
Or, it's possible that a lot of current poker players are the kind of nerds for whom figuring out GTO strategies and memorising probability tables for games is intrinsically fun, and they'd do it even if they had no intention of ever actually playing the game. In the same way that some people memorise everything about different types of train, even if they never intend on working in the railway industry, but because it tickles the right part of their brain. But I doubt it.
I much prefer the US fantasy sports format with a draft and only a dozen of players in one league, such that there is only one team which has a Mohamed Salah.
I have read up on draft systems, and while I basically get the mechanics of how they work, I don't have a good feel for how getting (or missing) your pick at a particular stage of the draft plays out.
I personally hate it. Though I can see many do not mind.
The first time I experienced this was playing the original Demon's Souls, and then later playing Dark Souls. I wasn't actually any good at playing Demon's Souls, and during my first playthrough I stumbled onto all sorts now-well-known ways of abusing the game's mechanics:
- Baiting individual enemies with arrows so you can fight them one at a time.
- Killing enemies from afar with arrows by standing outside of their agro range.
- Physically turning off the console to avoid a death being recorded. (at the time I didn't know exiting the game would work better)
There are more examples than just this, but I distinctly remember when first playing these games that I often didn't feel like I was playing the game "correctly." By 2009, most games had pretty lengthy tutorials, and would often have a "live" tutorial every time you encountered some new type of enemy. They'd tell you "proper" way to find the enemy's weak point, often of a voice-over. Demon's Souls obviously did none of this, and you were left to figure out how to make it past each enemy. Often the best solution (from a difficulty perspective) was the jankiest. Take the Ornstein and Smough fight in Dark Souls. Is there anyone who doesn't use the pillars to get cover from Smough's hammer attack?
In any case, the point I'm making is that the less optimized games can be much more rewarding, and often demand much more player creativity.
I don't think there's anything about the game that is giving you that joy. It's the fact that you are discovering strategies yourself. When games have large online communities, it's like setting off an AI. You have so many people trying different strategies and sharing them with others that you don't get the joy of solving it yourself.
That said, I have a weekly group that plays pretty low level paper commander, and it's a blast. We live in different states now but play on Spelltable and get together every few months to do a draft. It's the only way to play, IMO. Plenty of room for creativity within the confines of our personal meta.
I hope future games will use more randomization to keep players guessing and building community.
This is an interesting post to read through: https://www.ichess.net/blog/why-has-professional-chess-scene...
It's amazing how often this significantly upsets other players. It's not real money (or even significant if they went bankrupt and couldn't wait 24 hours to get the next handout) at the tables I plan (small amounts). I've seen serious meltdowns over pennies.
Long ago I played in cheap & FTP tourneys. Multiple people would go all in on the first hand or relatively soon. Whoever 3-4xed their chips in the first few hands had an enormous advantage for the rest of the game. It was the single largest reason I stopped playing.
I want to clarify, my bets were not random and they are not even close to the order of magnitude as the all-in bets. Poker isn't just "mathematical purity" it's also psychology and that's part of the skill. A player losing their cool because I kept throwing 30 into the pot instead of 10 when we all have 1000 on hand is losing the psychology game big time.
Any game that involves dice or random cards has at least some luck involved.
A while back in Destiny there was some sniper that required getting PvP kills with it in order to get an upgrade (or something like that). I suck at PvP but was able to do it and you know what - I got decent at sniping in PvP simply because I kept doing it over and over and over. I believe its the grinding of the content that makes the player better. Those pro gamers that play 8-12 hours per day are really good. A lot of people would be really good at some of these games if they put that time in. The changed stats on gear is an illusion to keep players engaged (not a fake illusion but I think a facade with marginal difference).
-GK Chesterton
An example from today is in Elden Ring - largely a single player game, but with multiplayer elements. When I started playing, there was a huge variety in player stats, weapon/armor selection, and people were even encouraged to try out different things by the in-game "rebirthing" mechanic. Then someone found out that a single sword used a specific way vastly overpowers everything else, especially in multiplayer, and now everybody who plays player-versus-player is locked in to using that same sword or else it's not even a fair fight. Players have been begging the developers to change this, and they only recently (as of this week) did something about it.
For as uniform as, say, the standard meta is, it's still based on player consensus and behavior, not computer analysis of game mechanics or simulations of games. Turns out player consensus is still pretty good. I don't know if you could moneyball Magic.
I happen to play both FPL and HS. The difference is that for Hearthstone since you play against others deck the meta can change, hence a win% is only relevant in the current context. That doesn't really work like that with FPL.
I agree this is taking the fun out of the game, but really only if you focus on winning rather than building yourself your own deck or team. So I'd let other copy paste the optimal team every weeek, I'm good thank you. Ironicaly if you want to win a league, most likely you'll have to make a couple of bold choice during the season, gambling against a model and being right.
Anyway, I've started coding my own bot to help me pick FPL players, it's never going to be as good as any online model and probably even myself but in the journey the fun will be :)
I see data-driven approaches are being used more and more everywhere, not just in games, and I can see how that can be seriously problematic in some cases.
But, for example, my 14yo son plays on a popular minecraft server, with his friends, and he's been writing python scripts to download in-game item sales/pricing data, and make graphs/do analytics on it, so they can get an edge in online trades. I think he's having more fun doing that than actually playing, tbh. But the point is there is no 'optimal' strategy or configuration in that game, it's still 'open'. It just now requires more sophisticated play rather than a naive casual approach.
Many people in quant trading industry learn it every year the hard way.
I don't know about fantasy football, but in the video-game speed-running community people often come up with new categories to maintain the fun. Sometimes a game can be broken to the point that it can be finished in mere minutes, but people still play the 100% category as well.
How about restricting what kinds of data and models are allowed? Another option is to straight up ban the model team and force people to get creative.
It sounds like you're basically already doing this, but maybe you should lobby your preferred platform to add explicit support for this kind of restricted play.
I think the obvious one (to me, anyhow) would work: Penalize a person relative to the number of people who take him. I'd have to sit down with some pencil & paper to work out the best way to do so, my gut says it'll be something a bit non-obvious to make the math work out the best (possibly involving something like the logarithm of the number of people who selected a given player), but off hand it's not immediately obvious to me what that is. (I suspect just dividing the value of a player by the number of people selecting them is a bit too direct and has some pathologies in it. It certainly does around low numbers of selections, e.g., if you selected a person first you take a heck of a hit when the second person selects them.)
I'd motivate the change with a large player base by saying this is an attempt to simulate at a large scale the fact that in the real world, teams can't pick the same person more than once.
Even if this didn't solve the problem, I am fairly certain it would destabilize the meta and at least turn it from something static to something dynamic, which is still an improvement over there being one single true solution.
I suspect that with computers getting better at so much of what we took pride in, we'll have to start looking very deeply at what makes a human a human. A human still has to aim the computer at something the human finds meaningful - hopefully, we'll have a lot more time to think about that question.
The stuff we got to do to keep the lights on prevents us from doing that fulltime. For now.
Edit: Although in the West, if you have some cash, even work-life seems more and more like a pretend play. We all dress up and act professional so we can weave a nice narrative about being productive when that clearly is very questionable. There's a truckload of gaming going on on just about every level. OK, I might be having a bad day..
Alternatively, in competetive environments, things that have stock trading or market mechanics mean that as soon as something becomes “common sense” it’s probably not the correct move anymore.
> You can find people who could barely add and subtract yet have the ability to recite various player and team statistics.
because it calls to mind a scene from 'the wire', i think a lot about: https://www.youtube.com/watch?v=O1mmePD549o