Jump to content
Sign in to follow this  
FTS Gecko

MathWing: Killing Diversity In The Game Since 2014

Recommended Posts

It's less the model that's to blame and more the way we use it. We look at its predictions at treat them as infallible gospel, and when we fail we turn to the model to take the blame off of us, to show us the flaws in our tools.

 

And if you don't believe that we treat the model with far, far too much blind reverence, consider this: if it were slightly to moderately wrong we'd never know.

Share this post


Link to post
Share on other sites

and yet the dogged pursuit for perceived optimum "efficiency" appears to be stifling those options and killing off that variety.

Competitive games like you see at regionals, national, worlds, or even just the Store Championships have always been about optimum efficiency. That's true of every game out there, the higher the level of competition the less room for 'fun' builds you have.

You can't blame Mathwing for pointing out what the most effective ships are, people would of figured it out on their own without any help.

As long as people consider winning to be more important the playing a "fun" list, then some ships won't be options because they don't aid in that goal.

Anyone who thinks this is about WAAC or that people aren't playing for fun, simply don't get the point...

Share this post


Link to post
Share on other sites

I've got nothing against mathwing as long as people approach it with an understanding of the caveats.  I just wish I saw more "X is overcosted by about Y points, so here's how you need to try and use it's advantages make up that difference" instead of "X is overcosted by about Y points, so you're playing at a disadvantage if you use it".  The second statement is not necessarily untrue in a given instance, it's just lazy, because there's a lot to discuss about how to use a particular ship with its particular set of strengths to make it work in a squad context instead of a theoretical context.  

Share this post


Link to post
Share on other sites

In any competitive match there will always be this min maxing as competitive players learn which builds are better than other builds. From Hearthstone to MOBAs to Magicthe Gathering to Netrunner and even X-wing. I'm sure Armada will end up with its math wing when it comes out.

 

Only partially relevant to OP but again it talks about how high level play attracts audiences.

Edited by Marinealver

Share this post


Link to post
Share on other sites

 

The TIE Interceptor's stat line cost efficiency is actually a hair higher than the X-wing.

 

And the TIE Interceptor has boost and barrel roll, so if the PS1 Interceptor isn't viable, then the PS2 X-wing certainly is not either.

 

So, it seems as though (inherently flawed and limited in scope) mathematical models are once again declaring ships to be entirely "not viable".

 

It's this kind of cynical, statistic-driven analysis that has led - over the last 12 months - to a self-fulfilling prophecy of a meta. Entire ranges of ships, pilots and upgrades are ruled out wholesale because they're considered (under the said inherently flawed and limited in scope model) to be marginally less "efficient" than their cousins.

 

New players pick up the game, gaze in awe at the wealth of options available, join a forum to find out more and are told "yeah, don't use that unless you want to lose."

 

TIE swarms became the norm because we're told they're "efficient".  BBBBZ becomes the norm because we're told it's "efficient".  At the expense of everything which is apparently not.

 

This (incredibly well designed and balanced) game is packed to the brim with an absolute wealth of options and variety, and yet the dogged pursuit for perceived optimum "efficiency" appears to be stifling those options and killing off that variety.  So here's a novel approach: to hell with so-called mathematical efficiency; whatever happened to fun?

 

 

I don't have a problem with it -- or rather, I don't blame math if a game lacks diversity. 

 

On the other hand, I do think people put a bit too much stock in mathwing.  It's still a random game, and a slight efficiency improvement isn't going to be noticed in a single, or even ten, games. 

 

Basically, if there's some inefficiency that you weren't aware of before mathwing, it's unlikely to be severe enough for you to lose your next game.  It's much more likely you will lose because of a bad maneuver or dice roll than any "inefficiency".   There are, of course, more glaring overcosted upgrades and ships that might well cause you to lose, but these are mostly obvious even without mathwing.

Share this post


Link to post
Share on other sites

Given that most players are playing a game rooted in the most efficient 100-point builds, it's fabulous that some of the more mathematically-inclined members can help shed light on which combinations are efficient, and why that's the case.

 

Which would be great were it not for the nature of the metagame: it likes to self-perpetuate. A list at the top has a tendency to stay there because people copy it: the more copies of a list are entered into a tournament the more likely it is to win. In a tournament where every list is a Fat Han, Fat Han will come first. It'll also come last, but we don't have the data on what fails spectacularly. We only see the winner.

This also applies to MathWing. Mathematical predictions would be great for confirming why things are winning if the people playing in the tournaments couldn't see MathWing. But they can. MathWing predicts a ship as good, it gets used. It gets used more, it turns up at the top more. It turns up at the top more, that's taken as a validation of MathWing's prediction.

 

On the other hand, I do think people put a bit too much stock in mathwing.  It's still a random game, and a slight efficiency improvement isn't going to be noticed in a single, or even ten, games. 

 

And MathWing also has a margin of error: it's not a perfect model. It may not even be right.

Edited by TIE Pilot

Share this post


Link to post
Share on other sites

    Even when playing the game for fun and enjoyment we all have the desire to win. The information in Mathwing was never intended to shut down creative squad building, nor to push a meta a certain direction. These things would work them selves out even without the math. Players would soon realize the points spent of certain upgrades and pilots don't pan out and move onto other things. We are creatures of habit and we do the things that reward us.

    If a certain build wins more we tend to use it more. This goes right to the heart of a term "net-decking". Net-decking, is copying what others have had success with. We see this often in many games not just X-Wing. Those players who find a way to see the trends in squads being played can and will find a way to defeat the current popular squads. This reset the meta and the cycle continues.

    This is what the term "meta-game" is all about, the game inside the game. The best players adjust quickly to growing trends while the rest catch  up.  Keep in mind this only pertains to the tournament scene and nothing else. If the meta is creeping into your kitchen table games then it's time for some house rules. If the meta bothers you so much, Store Championships and Regionals are not for you, go play to have fun.

    Playing for fun does not mean, not competitive. The fun of the game for me is winning...  :o! That's right I play to WIN! This always seems to shock some players in X-Wing when a player declares they are trying to win. The difference between trying to win and WAAC (Win at all Cost) is vast. I can be cordial and respectful to my opponent while trying to win. There is never a reason to be a **** to your opponent or bend the rules to gain an advantage. 

    What does all this mean? If the Mathwing is what you find disturbing then you are not cut out for higher level play. Stick to your casual games at the kitchen table and by all means have fun..

Share this post


Link to post
Share on other sites

 

Given that most players are playing a game rooted in the most efficient 100-point builds, it's fabulous that some of the more mathematically-inclined members can help shed light on which combinations are efficient, and why that's the case.

 

Which would be great were it not for the nature of the metagame: it likes to self-perpetuate. A list at the top has a tendency to stay there because people copy it: the more copies of a list are entered into a tournament the more likely it is to win. In a tournament where every list is a Fat Han, Fat Han will come first. It'll also come last, but we don't have the data on what fails spectacularly. We only see the winner.

This also applies to MathWing. Mathematical predictions would be great for confirming why things are winning if the people playing in the tournaments couldn't see MathWing. But they can. MathWing predicts a ship as good, it gets used. It gets used more, it turns up at the top more. It turns up at the top more, that's taken as a validation of MathWing's prediction.

 

On the other hand, I do think people put a bit too much stock in mathwing.  It's still a random game, and a slight efficiency improvement isn't going to be noticed in a single, or even ten, games. 

 

And MathWing also has a margin of error: it's not a perfect model. It may not even be right.

 

 

Right, there may be some aspect of confirmation bias going on, but these things are played over dozens of iterations.  If Mathwing predicted that the most efficient ship combination was something like 4 pre-fix TIE Advanceds, that might get lots of players trying it out.  I doubt very much if that meta would survive, however, because it really isn't an efficient combination.  The fact is, while Mathwing has a margin of error and the game has a degree of randomness to it, the margin of error has limits.  Knowing those limits helps to improve the balance of the game over time.

Share this post


Link to post
Share on other sites

I've got nothing against mathwing as long as people approach it with an understanding of the caveats.  I just wish I saw more "X is overcosted by about Y points, so here's how you need to try and use it's advantages make up that difference" instead of "X is overcosted by about Y points, so you're playing at a disadvantage if you use it".  The second statement is not necessarily untrue in a given instance, it's just lazy, because there's a lot to discuss about how to use a particular ship with its particular set of strengths to make it work in a squad context instead of a theoretical context.  

An excellent point that needs repeating. Strategies on how to deal with inefficiencies or solutions to problems are significantly more helpful than listing problems. A naked HWK is absolutely inefficient but right upgrades can make it a force to be reckoned with.

 

That being said....

If a ship is point-for-point unable to fulfill a particular role better than any other ship out there, then that needs to be known as well. If there's nothing I can add to a rookie X-wing to make it as efficient of a jouster as a blue squadron b-wing, I need to know that so i don't keep wondering why BBBBZ tears through my XXXXZ every time I face it on the field. 

I can't blame mathwingers for not providing tactical advice along with awesome math because I would look at that matchup and just assume either A) my dice are cursed or B) my opponents dice are blessed because humans are wonderfully ignorant creatures when it comes to on-the-fly averages and predictions.

I like to think Mathwing is a jumping off point for tactical discussion and (on occasion) awareness of less than optimal builds :)

Share this post


Link to post
Share on other sites

New players pick up the game, gaze in awe at the wealth of options available, join a forum to find out more and are told "yeah, don't use that unless you want to lose."

That's a bold statement. I'd guess the vast majority of new players just play the game instead of checking the forums for what they should play.

People give MathWing way too much credit for shaping the meta. If there were some list archetype that could be consistently successful that wasn't predicted by MathWing, I suspect we'd have seen it by now given the number of Store Championships played. (There are nearly 3,000 lists entered in List Juggler, and that's only from voluntary reporting.) I highly doubt every competitive player is a mindless slave to MathWing.

Share this post


Link to post
Share on other sites

I personally don't really care about it anymore. I learn from it, but that doesn't stop me from playing what I want. I make my own experience. I love to experiment, but I'm not stupid either. I would not bring 5 HWK without turrets to a tournament.

 

But I won't blame someone for listening to it or making his lists exclusively based on it. The math is not lying. If you want to be efficient, you should at least recognize it.

 

What I have a problem with, is people that don't use it well or will blame their defeat on it (except in rare cases when it is really too much). It's not because statistics says that you should do on average 2.25 hit that you will during this particular game. It's not because statistics says that 1 attack out of 3 should at least hit that it will, or that it won't hit more. The game is still at the mercy of randomness: **** happens, deal with it.

Share this post


Link to post
Share on other sites

I believe it's more a rant designed at new players going online to get "good" advice or lists.  They read the hyper competitive environment and follow that.  The local game becomes very similar to the high level meta in variety (not much variety).  The local meta becomes boring. 

 

I can agree with this and do lament it, but I don't think there is anything you can do about it.  The forums are places that hyper competitive players go to and chat about the game.  They exchange ideas and improve their play.  That's not a bad thing, even if it's not what everyone wants.  Those that are new want to play a good game and follow the discussions going on instead of discovering the game for themselves.  That's where we kind of enter the echo chamber.  Maybe we should tell more new players to test things out themselves?  I don't really see any "fix" for it, even though I lament it happening. 

 

I'm really lucky that the guys in my local game shop are really cool and we play casual games each week.  That is, unless we are trying to play the meta for practice at tournaments. 

Edited by heychadwick

Share this post


Link to post
Share on other sites

Given that most players are playing a game rooted in the most efficient 100-point builds, it's fabulous that some of the more mathematically-inclined members can help shed light on which combinations are efficient, and why that's the case.

Which would be great were it not for the nature of the metagame: it likes to self-perpetuate. A list at the top has a tendency to stay there because people copy it: the more copies of a list are entered into a tournament the more likely it is to win. In a tournament where every list is a Fat Han, Fat Han will come first. It'll also come last, but we don't have the data on what fails spectacularly. We only see the winner.

This also applies to MathWing. Mathematical predictions would be great for confirming why things are winning if the people playing in the tournaments couldn't see MathWing. But they can. MathWing predicts a ship as good, it gets used. It gets used more, it turns up at the top more. It turns up at the top more, that's taken as a validation of MathWing's prediction.

Except numbers exist on how often ships win in comparison to the number of those ships entered into tournaments. So we do know roughly how a ship is performing given the number of entries of that ship, kind of like a batting average. So you can look at that ship's "batting average", and if MathWing provides reasonable estimations, it should also predict the ships with a high "batting average". Generally it does. Edited by GiraffeandZebra

Share this post


Link to post
Share on other sites

What I feel people miss about Mathwing - it's a model. I know that models can be used to predict things, but I don't think the models designed thus far are perfect. It's hard to say whether people are all doing X because X is best, or if it's because such-and-such said that X is best, so everyone is doing X.

I wonder what the lists would be like if the game was played out at the various locations its played at, and no information was shared and nothing was up on Youtube and each individual meta existed in its own unique environment. Certainly X might win the Nationals, but the diversity in the remaining group might be greater than it might be this year.

 

Jacob

Share this post


Link to post
Share on other sites

It's this kind of cynical, statistic-driven analysis that has led - over the last 12 months - to a self-fulfilling prophecy of a meta. Entire ranges of ships, pilots and upgrades are ruled out wholesale because they're considered (under the said inherently flawed and limited in scope model) to be marginally less "efficient" than their cousins.

I disagree.

For every thread discussing why a given ship is inefficient, there are always a handful people who chime in with the "you're not utilizing the ships strengths; stop trying to fly it like an interceptor/jouster/flanker/whatever." Then those proponents take that ship to regionals and proceed to not make the top 8 cut.

Funny how that works out.

Don't get me wrong, I fly unique snowflake lists too. I just do it knowing that I'm trading a decrease in tournament performance for an increase in personal enjoyment. What I don't like is when people get all judgmental about other people picking a different balance point on the tournament performance/personal enjoyment seesaw. If it's a legal list and they're flying it casual-like, game on.

The way to address stagnant metas is to give the devs accurate, relevant feedback so that they can incorporate that when putting balance tweaks into the next wave. Some of the most relevant feedback is statistical analysis of the ships, combined with tournament results and trends.

Oh, that's MathWing.

Share this post


Link to post
Share on other sites

Math is language to describe things, it doesn't create them. 

 

The "inefficiency" of things like unnamed Defender or single Alpha Squad Interceptor is clear to anyone who has experience with game no matter if he's aware of MathWing or not.

 

Did you really need some kind of complicate calculation to get that Delta Squad Defenders aren't worth their points? Did you try to fly a ps1 30 points ship which stress itself when doing hard turn and has only straights green (without even have the 1!)?

 

Having said that, MathWing never claimed to be anything more than a mathematical model create on an arbitrary and subjective base which take into account just simple jousting values, and we all know that Xwing is much more than just that, so if people are taking MathWing as their Bible it's their problem, not its creator fault

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Sign in to follow this  

×
×
  • Create New...