PSA: no public MathWing / ship evaluation for X-wing 2.0

By MajorJuggler, in X-Wing

I've gotta chime in and say it's pretty dismaying how much crap people here are giving MJ. This guy has done a lot over the years that has informed the way the game is played and the way that we think about it, and he sure as **** hasn't turned a profit on that investment. The fact that he believes that his IP is worth something, and is taking steps to protect it is not a reason to give him ****. He's spent years doing rigorous analysis of the bones of the game we all play, and has shared that work quite freely. Has anyone on this thread giving him grief done half as much as he has?

I for one salute him, and am disappointed that the game runners did not avail themselves of his talent. It's too bad that we won't be getting that analysis either, as a consequence.

Edited by Babaganoosh
42 minutes ago, GreenDragoon said:

I see your points. There's nothing to add to 1 and 3 (except maybe 3, that this does make Mathwing valuable for a huge amount of people. Knowing when to avoid the joust could be another step to getting better).

As for point 2: we've discussed the shortcomings (and huge advantages!) of metawing before. The reason I brought it up was simply as an example for another tool where the constraints need to be understood. Another example could have been dice calculators which can inform your choice of action and token spending, or to gauge matchups to some degree. Calculators can tell you what damage you can expect. Mathwing can tell you whether the ideal jousting situation is worth it or not, which actions to take, and whether range boni change the result. Such jousting situations can occur during a game, too, and inform your decision. That's it. And for that it's awesome.

And I've always prefaced anytime I've talked at length about Metwing with all the provisos for how you look at it. TBH I've also stopped talking about Metawing pretty much entirely in the last year or so.

I think dice calc & probabilities are a bit different - those are hugely useful to know and have to hand. I guess my principle issue with Mathwing is because I understand how important the dice calc & probabilities are during a game, and how much they change in a real time way, that I know how pointless crunching all that extremely valuable nuance into a single amorphous 'jousting value' blob actually is. The numbers you need to know don't exist at a squad level, they exist at an action level.

3 minutes ago, Babaganoosh said:

Has anyone on this thread giving him grief done half as much as he has?

Yes, multiple people. Quietly.

Edited by SOTL
6 minutes ago, SOTL said:

I guess my principle issue with Mathwing is because I understand how important the dice calc & probabilities are during a game, and how much they change in a real time way, that I know how pointless crunching all that extremely valuable nuance into a single amorphous 'jousting value' blob actually is.

A large enough number of perfect simulations would result in the calculated expected values though, and that makes them sufficient.

15 minutes ago, SOTL said:

The numbers you need to know don't exist at a squad level, they exist at an action level.

Which is part of version 3.0 and which we‘ll never see, unfortunately.

Anyway, it is clear that you don‘t value Mathwing. I think I‘ve made my case that it is useful for a large part of the playerbase.

1 hour ago, Azrapse said:

The mathematical fundaments behind the game are obscure to a majority of the player base.
What you are saying there is that it is better to keep those fundamentals obscure in the name of fun or variety.

That is as flawed as the Security through obscurity anti-principle in the world of cryptography.

That leads to those with the knowledge of those fundamentals to play at advantage against those without it.
At the beginning Expose was a super popular card. Many equipped it because it let you roll one more die!
Then someone explained how bad it was mathematically, and everyone stopped using it.
If that person (or some other) had not explained why those 4 points were such a waste, we would have had lots of people playing at a disadvantage for years against those with the knowledge that it was worthless.

It's like encouraging a hermetic game, where those who know win, and those who don't lose.

Fair enough. You've given it far more thought than I did. I was just trying to reasonably explain why my gut feel leads to me thinking a certain way.

Just to further clarify, I've no problem with someone doing this work. I guess I prefer to be the hermit in this case. Rule of cool etc.

If you do the math to a level that you see the pros/cons, good on you. But if efficiency doesn't suit my play, or doesn't look good thematically, I'll pass.

And that is one reason I'll never win a tournament! :)

I didn't pickup x-Wing to do math.

7 minutes ago, Wiredin said:

I didn't pickup x-Wing to do math.

So you don't use range bonuses, or... ?

Just now, ObiWonka said:

So you don't use range bonuses, or... ?

okay. I didn't pickup X-wing to do anything above grade 2 math?

@MajorJuggler, can I hire you as a consultant? Cuz I can't ever seem to win anything. :unsure:

3 hours ago, ViscerothSWG said:

The b-wing to X-Wing was a 0.04% difference. The b-wing was acceptable but the X-Wing was unplayable. It's a old thread, happy forum searching.

By the time Integrated Astromech came out, the Bwing was already dead. The Xwing fix was DoA because it was balanced with another dead card.

6 hours ago, SOTL said:

The game is better off without you.

There are some people like that around here. :wink wink, nudge nudge:

I'm the turd that won't flush.

Through the saviour complex, I think I see what you're trying to get at.


You're trying to say FFG follows the inverse square law.

nrlmN4.gif

2 hours ago, Babaganoosh said:

The fact that he believes that his IP is worth something, and is taking steps to protect it is not a reason to give him ****.

Sure, but then why is he coming here to moan about it to us as if we can do anything? What is the purpose of his post outside of just ranting about how FFG won't give him the attention he feels he deserves?

21 hours ago, MajorJuggler said:

Greetings all,

A couple months back I had asked Alex, Frank, and Max if FFG would be interested in hiring me as a consultant to do some technical balance analysis for X-wing 2.0. The upshot of the conversation was that FFG management would only be willing to let me work on pre-release material if I joined the general playtest group and signed the standard playtester NDA. The playtester NDA includes an Intellectual Property paragraph which is very broad, and essentially states that any mathematical process I reveal to them becomes their permanent and exclusive property. This would have multiple implications:

  1. As a general playtester I would be unpaid.
  2. I would be unable to later publish the general mathematical theory of how point values can be approximated for any wargame, including posting on forums here, or publishing in academic literature, or using these formulas to playtest for another company, or as a designer for another company. If I signed the NDA and later did any of these at any point in my life, FFG could sue and hold me personally liable.

Neither of these implications were acceptable to me, so I declined. The developers can now change costs post-release, so as a corollary of this decision I have also decided to not make any of my X-wing 2.0 analysis public. FFG has invested many man-years of designer/developer labor into X-wing design and playtesting, but their technical analysis capabilities are significantly less than the infrastructure I have developed. Succinctly stated, I don't feel like giving away that information for free.

There is of course considerable debate as to how useful such an analysis toolkit will be now that the developers can change point costs dynamically. I estimate that it would save them 2-6 months of calendar time on getting point costs settled to their "final form". Certain ships and pilots will be easier for them to price correctly on initial launch than others. This is however, largely a theoretical discussion at this point, since I won't be performing this service for them anyway.

I know that many of you look forward to my analysis of various pilots and ships as they are previewed, so wanted to make this PSA. I will still be performing the analysis privately for myself, but I will not be sharing the results. Lets hope that the FFG developers can get 2.0's point costs and balance worked out in a timely manner, so we don't end up with the poor balance and powercreep that defined 1.0.

Cheers,

-- Bob

I tried to skim through the thread to see if I missed anyone bringing this up, so forgive me if it was already addressed: Looking at the general parameters of the NDA/IP agreement as you described them, it looks like a boilerplate contract taken from just about any standard contracting/hiring of a developer that I've ever seen in my 19 years of experience as a software engineer. Thing is, it should be possible to re-negotiate the terms of that agreement. See if you can strike that paragraph, or at least strictly narrow the language so that they get a for-pay license with a limited duration. You produced this IP prior to any work with them, so this should be eminently reasonable.

If they aren't willing to accept those terms, then walk away. It's your IP, and you have the right to determine if and how it distributed. You developed it without being paid by them, so they should not get ownership of it. If you choose to make it available for free or not, it's still your IP and your right to do with it as you please.

18 hours ago, RookiePilot said:

@MajorJuggler - I do hope that you can use your work to win something.


But I mean, he hasn't yet, right? Five years of 1.0, and MathWing's staunchest experts didn't have a terribly impressive competitive pedigree, despite claiming to have unlocked the power-level curve of the games components on a fundamentally objective level.

It's not like, for all the bravado of MathWing's crystal ball, did MajorJuggler discover Dengaroo or anything comparable. For all of it's trumpet-blaring, MathWing never even predicted anything like Dengaroo (or the double-droid Fat Han, or that Control-Bots would win Worlds this year, or that... ). MathWing can sometimes use its modeling to explain why a list did well, but usually only after the fact. It can fit reality to fit its explanations, but it usually wasn't great at using its analytic power to make predictions. ****, Creationism can explain everything and anything after the fact (and that doesn't make it a powerful theory, that makes it a problematic theory, in part, because it can't ever be falsified or submitted to rigorous tests)... but real analytic power comes in making bold predictions about the future.

It has always seemed like competitive X-Wing is/was far too complex of a game to be reduced down to an abstracted brute jousting value.

MathWing is neat, and in some regards it's pretty impressive. But I think its value and utility have been greatly exaggerated.

3 minutes ago, Freeptop said:

If they aren't willing to accept those terms, then walk away.

He won't admit it but he really doesn't want to walk away.. that's why he's trying to extort the community or whatever this post is supposed to accomplish.

52 minutes ago, Rakaydos said:

By the time Integrated Astromech came out, the Bwing was already dead. The Xwing fix was DoA because it was balanced with another dead card.

And we knew they were dead in each case because of mathwing. Thereby suppressing then from the field of play which then skewed metawing further against them.

Back when the .04% was relevant data, nobody was playing the same list 100+ times to perceive the difference, or the thousands of times needed for that difference to make a mark on the win/loss total. My point in that being:

If you aren't running to the store during a 10% off sale, you shouldn't care about a 10% difference in mathwing values that are themselves an incomplete picture of the overall effectiveness of a ship. That's all I ever argued.

I still appreciated the free data when given. And hope MJ stays active on the forums.

Ultumately I hope to someday have complete balance on release because

2dgq9y.jpg

I'm sorry you didn't get the opportunity you were seeking, but I'll be honest - there is a limit to the involvement of math.

So much of this game is overlapping abilities that can't be applied a value.

I'll be honest, if they are bringing in someone new for the team, I'd rather see a technical writer to vet all the rules and card wording before someone to "balance" the math values of the game.

10 minutes ago, AllWingsStandyingBy said:


But I mean, he hasn't yet, right? Five years of 1.0, and MathWing's staunchest experts didn't have a terribly impressive competitive pedigree, despite claiming to have unlocked the power-level curve of the games components on a fundamentally objective level.

It's not like, for all the bravado of MathWing's crystal ball, did MajorJuggler discover Dengaroo or anything comparable. For all of it's trumpet-blaring, MathWing never even predicted anything like Dengaroo (or the double-droid Fat Han, or that Control-Bots would win Worlds this year, or that... ). MathWing can sometimes use its modeling to explain why a list did well, but usually only after the fact. It can fit reality to fit its explanations, but it usually wasn't great at using its analytic power to make predictions. ****, Creationism can explain everything and anything after the fact (and that doesn't make it a powerful theory, that makes it a problematic theory, in part, because it can't ever be falsified or submitted to rigorous tests)... but real analytic power comes in making bold predictions about the future.

There's something in modeling called "overfitting" where your model has so many variables that you can basically make it say whatever you want it to say. Predictive ability is problematic for this kind of model.

26 minutes ago, justaslug said:

He won't admit it but he really doesn't want to walk away.. that's why he's trying to extort the community or whatever this post is supposed to accomplish.

Or he’s just pre-emptively letting us know that he’s not going to be posting the same stuff that he was before once 2.0 hits, and this is why. But whatever salts your coffee.

4 hours ago, Azrapse said:

That leads to those with the knowledge of those fundamentals to play at advantage against those without it.
At the beginning Expose was a super popular card. Many equipped it because it let you roll one more die!
Then someone explained how bad it was mathematically, and everyone stopped using it.
If that person (or some other) had not explained why those 4 points were such a waste, we would have had lots of people playing at a disadvantage for years against those with the knowledge that it was worthless.

And, IMHO, this is exactly what wrecks the game. Everyone will one play the subset of cards that have been "scientifically proven" to be worthwhile. This has 2 long-term effects:

1. No one will experiment. The game quickly becomes a limited collection of math-whizzed choices to the exclusion of all other options.

2. It puts the game in the 99th percentile. If a large segment of the population was NOT hyper-optimized, a lot more people could play the upper 5-10% of the game (thus having far more options) and have a chance at winning because people were playing what they wanted, not what they were told to play to WAAC.

27 minutes ago, ViscerothSWG said:

Ultumately I hope to someday have complete balance on release because

2dgq9y.jpg

In the nicest possible way, this is completely delusional.

A. Perfection is impossible.

B. The required use of the Tweek-o-Matic X-Wing App just proves that even the devs know that nothing will ever be well-balanced.

I'm going to put very strong odds on the notion that 2.0 is going to see the same rarified playing field that 1.0 did. It may take longer, and it will likely be a different collection of ships, but if people are going to netlist, metawing, and/or mathwing, the game will condense down to it's new soul-crushing singularity that represents tournament play.

Edited by Darth Meanie
1 minute ago, PhantomFO said:

Or he’s just pre-emptively letting us know that he’s not going to be posting the same stuff that he was before once 2.0 hits, and this is why. But whatever salts your coffee.

He could do that with much less of the saviour complex however.

What is wrong with you people?!

4 minutes ago, GreenDragoon said:

What is wrong with you people?!

Dang it. I guessed page 6 for this response