Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

To get good, go after the metagame

commoncog.com

Post reply on HN
Print view View original

191–200 of 307 posts

Re: To get good, go after the metagame

#191

Post by roel_v »

city41 wrote: post #131
Proziam wrote: post #50

Earlier quoted context omitted.

In my experience it comes down to two things: 1. People think they already have the fundamentals 'down' 2. Actually mastering the fundamentals can be boring and make a game feel like work (which, if you don't truly enjoy pushing the limits in a game, it is.) A totally random example from League of Legends. 99% of players go into game after game and literally never review their previous games. They will spam games but…

I also think fighting games are a good example. Most beginners gravitate towards the special moves and supers. But basic moves and understanding positioning can elevate your game much more.

This is true of real fighting too...

Re: To get good, go after the metagame

#192

Post by Spellman »

_pastel wrote: post #174

Playing a game immediately after learning the rules is my favorite challenge - especially when the other players are all new too. At this stage, tactics are mostly about keeping all the rules in your head and thinking hard before each move. But strategy is really interesting. Everyone is guessing blindly, and you can often win by guessing slightly less blindly. I have two meta-strategies: (1) Game phases While learni…

I do a similar calculus when learning a new game.

However, I break things down a little differently.

1) How to win. What gets counted to win, usually VPs. If you understand that, and how to efficiently accrue them, then you're a few steps ahead. Then, how to end the game. Can you control how it ends?

You use this in engine vs endgame, but I prefer breaking it out because not all games have two phases. Some simply generate VPs as you go (no compounding engine mechanic), and what you really care about is turn efficiency.

2) My other rule is compare to similar mechanics from other games.

I have heavily studied other games, so spotting how a draft system works or engine-building and turn efficiency is easily recognizable and I can calculate sorta optimal moves. Similarly this gives you an easy framework to figure out how the specifics of this game compare to each other.

3) Table dynamics. Almost always you can get a huge edge just by knowing what you need to do NOW vs latter. This is mostly about efficiency. If you know player X is going to take Red next turn, then you can decide if denying them is worth it, or that means you're safe to take/ignore Blue. So sussing out what other players are doing, especially if they're cooperating, is especially helpful.

To accomplish this, I actually try to turn the table dynamics to friendly and talkative. Usually in the guise of it's a friendly game and look at my sharing my insight/knowledge. So others freely share what they're thinking. Means discovering strategies in the game happens faster, but I also glean important tips from others.

Re: To get good, go after the metagame

#193

Post by Schwolop »

richk449 wrote: post #7
shinryuu wrote: post #2

First time I ever understood what "the meta" refers to. Though I've definitely participated in the metagame without realizing it before.

The definition of meta used in the piece doesn't match my understanding. Take his final example where he says that adjusting tactics in response to your opponent in ultimate frisbee is meta. No, that just strategy. Same goes for most of the examples he gives. Looks like I'm getting old, and the word has been redefined from out under me: In video games, however, the metagame has its own meaning—and depending on which…

I think a good example that aids definition is "meta-optimisation" (aka "meta-computation).

Optimisation is a process that aims to find the best possible solution to a problem. E.g. let's use Dijkstra's algorithm to find the shortest path between two locations.

Meta-optimisation is finding the best possible process to do optimisation. E.g. let's use A* with a reasonable heuristic instead of Dijkstra's algorithm, because it'll be faster to run and give the same optimal result.

The meta-game here is that time-to-compute PLUS travel the path is the real thing that needs optimising - not just the path length.

Re: To get good, go after the metagame

#194

Post by roel_v »

EdwardCoffin wrote: post #42
blackandblue wrote: post #27

Earlier quoted context omitted.

would you guys mind giving examples of what you are calling fundamentals?

My biggest improvement in chess came after comprehending The Theory of Steinitz [1]: 1. At the beginning of the game the forces stand in equilibrium. 2. Correct play on both sides maintains this equilibrium and leads to a drawn game. 3. Therefore a player can win only as a consequence of an error made by the opponent. (There is no such thing as a winning move.) 4. As long as the equilibrium is maintained, an attack,…

This applies word for word to jiu jitsu, as well...

Re: To get good, go after the metagame

#195

Post by roel_v »

gorpomon wrote: post #72

I am the worst (maybe best) person to play board games with because 5 minutes after learning the rules I loudly proclaim what I perceive the meta to be, and unashamedly telegraph my moves in regards to it. I lose almost all of the time, but it makes the night much more memorable and I enjoy the mental exercise of trying to quickly grok a meta. Sometimes for fun I loudly proclaim "I'm going to Moneyball this!", and th…

In law school, I had a 'how to negotiate' workshop once. We were given a fictitious deal to negotiate (with some external circumstances hidden from us), were paired up with a random person and then had 15 minutes to execute in the hallway. The second I got my numbers, I started building some sort of game theory model around the circumstances in my head, filling in some numbers and scenarios, all while stalling the girl opposite to me so that I had more time to work on my 'optimal strategy'. So then in the last 3 minutes or so, I 'executed' on my theory and we both got our negotiated numbers. The girl looked rather bored and played a bit dumb, giving out the feeling of 'I'll let you have this one, what do I care.'

Well the result is easy to predict - after the hidden circumstances and our respective numbers were revealed, it turned out she had basically steamrolled me and any smugness I might have had about my rational approach and winning strategy was crushed. I stopped doing any formal analysis of real world decision making after that.

Re: To get good, go after the metagame

#196

Post by Cogito »

richk449 wrote: post #55
gowld wrote: post #39

Earlier quoted context omitted.

The Splendor part was a but confusing. The author's point is that the simple strategy has no meta, because players can't interfere with each other. It's just a race. But the strategy of saving up points opens up a metagame because players can now interfere with each other, so the game becomes about guessing opponent strategies and counterstrategies. Another analogy is a bicycle race. You could have everyone run separ…

That’s exactly my point. Nobody calls a peloton meta racing. A peloton is just strategy. Always has been.

The peloton strategy is the current meta.

Re: To get good, go after the metagame

#197

Post by Schwolop »

aliswe wrote: post #120

Is management perhaps the "meta" of software development?

There are many games and many meta-games in this field. Here are two.

For SWEs, one base game is to earn the most money. The play-space appears to be one's current company, so working hard on company-beneficial projects and doing well should lead to promotions and pay-rises.

A meta-game for SWEs is realising that there are many companies that might hire you, so working on projects that benefit your current company may not be the best strategy. Instead, some SWEs attempt to work on projects that make their CV look more attractive to other companies.

A base game for Managers is trying to get the most output out of the team of SWEs available, by shuffling work around, training people, putting experienced and inexperience people on the same project to spread knowledge, etc. There's a lot of strategy and tactics in this; trading-off short term losses for long term gains, and the like.

A meta-game for Managers is being aware that some of those SWEs may be playing their own meta-game. Do you help that SWE to go up and out to replace them with a different SWE, perhaps with a more useful skillset? Do you cynically restrict them from working on CV-positive projects in an attempt (likely to fail and definitely unethical!) to keep them around? Do you find ways to pay or reward them more pre-emptively so that they're less interested in their meta-game?

Sometimes these meta-games are "too clever by half" and you may well be better off ignoring them and focussing on the base game itself. The important point is to be aware that such meta-games exist, and that even when people appear to be playing the same game, they may in fact be playing to lose in order to win a meta-game instead.

Re: To get good, go after the metagame

#198

Post by fwip »

Thriptic wrote: post #147
dfxm12 wrote: post #65

Earlier quoted context omitted.

It's fairly common not only in sports but source of many regulations and laws in any sector. Regulation is more of a pay to win game.

It's more complicated than that. Largely, it just seems that way because most people never bother to understand the rules past a cursory Google whereas big companies have teams of very experienced regulatory professionals from the agencies themselves who are all about it and therefore know every way to tweak inputs so that they trigger certain regulations rather than others. I was working on a regulatory compliance p…

If you don't have money, you can spend a lot of time becoming proficient in the field, and maybe find a loophole or two, and assure yourself you are in compliance.

If you have money, you can afford the compliance experts to find you every loophole.

If you have even more money, you can get new holes written into regulation, tailored specifically for you.

Re: To get good, go after the metagame

#199

Post by Crazometer »

thereare5lights wrote: post #172
novok wrote: post #161

Earlier quoted context omitted.

I think it depends on the game too. In Dota 2 (and probably LoL) the balance between heros can change significantly per patch, and a lot of winning is having good hero picks compared to the other team. If you don't keep up with the meta, then your win rate will go down, and this is true for amateurs and the pros.

I don't agree with that. It goes back to > It's not until you reach a level where everyone around you has a mastery of the fundamentals, that the meta comes into play. At any given MMR bracket, everyone around you has a similar mastery of the fundamentals so the meta is relevant. However, put someone with the fundamentals of a 6k+ MMR player into a 3 or even 4k match and they'll demolish everyone else even with a dis…

The assumption I work with is that mmr is a function of both fundamentals and meta. Fundamentals act as the floor, and meta as the ceiling. Team interactions make this all the more interesting - some examples:

A player with no understanding of the meta but good fundamentals can easily be directed by their team and have the same output as a high(mmr, meta) player. However, direction will not always be available and this will result in some losses.

A team full of players with high meta knowledge but low fundamentals will lose the fundamental game too hard and lose most matches. Also experience the "too many chefs in the kitchen" effect.

A team full of players with high fundamentals but poor meta knowledge will perform reasonably well, but be outshined by a more balanced team.

The team that won the largest DOTA championship 2 years in a row was clearly one that on balance had medium fundamentals (at a pro level), and outstanding meta.

Re: To get good, go after the metagame

#200

Post by danem »

Fnoord wrote: post #171
mercer wrote: post #121

Earlier quoted context omitted.

I've noticed this with chess too. As a kid, I briefly developed an obsession with chess. The book I read through argued that if I simply followed a few basic rules religiously, I'd instantly be a 'better' player than many who maybe played a lot more, but didn't know about these rules. And sure enough, when my obsession quickly waned, and without having played much chess at all, I found myself consistently winning fro…

Do you recall these rules? I'm rather curious about them.

I found John Bartholowmew's[1] videos extremely helpful when I was playing a lot of chess. Definitely saw a huge improvement.

https://www.youtube.com/watch?v=U2huVf1l4UE

Post reply on HN