How do Mathamancy?

[Ultramouse]

So.. obviously there are some parts of mathamancy that just cannot be recreated in a world without magic, but do you think a level 10 actuary holding a Ph.D. in math with a specialization in statistics qualify as a low level mathamancer? I think yes, but it really comes down to how important the magic portion of Erfworld is.. Someone with those qualifications could probably tell you the chances of your winning a battle with all the information given, but is that enough to be counted a mathamancer? Idk. Feel free to reply with your opinion on it, or give ideas for what sort of degrees/diplomas you feel might correlate to a type of caster.

[Balerion]

My bet is that the mathamancer can get access to perfect information much more easily than a real life mathmetician; look at Parson's bracer as a good example. A lot of the questions asked of it require information it essentially doesn't know (is it worth spending all the questions to get a single answer, and coming up with a solid % is the best example of this). The amount of data you would require to get an accurate number like that in the real world is such that it basically couldn't ever be answered to a single % chance (at least with any reasonable error bounds).

My bet would be (given the turn based nature of things) that a calculator on its own would do incredibly well for figuring out battle odds, as that is relatively straight forward. It might take a while, but its just a matter of plugging in a bunch of known information. My bet would be that the higher level mathamancy just gets to levels of perfection that are impossible with the methods we have available.

[BLANDCorporatio]

Baaawww, and here I was hoping someone was asking for help with their homework on Abstract Algebra or something.

But on the subject of the topic ...

My bet is that the mathamancer can get access to perfect information much more easily than a real life mathmetician; look at Parson's bracer as a good example. A lot of the questions asked of it require information it essentially doesn't know (is it worth spending all the questions to get a single answer, and coming up with a solid % is the best example of this). The amount of data you would require to get an accurate number like that in the real world is such that it basically couldn't ever be answered to a single % chance (at least with any reasonable error bounds).

The best thing to do is to ask an actuary about just how accurate their predictions can get. Real businesses, with real loads of money, rely on the poor ol' downtrodden actuary to provide advice (whether that advice is actually followed is a different matter). The thing is, a good knowledge of a wealth of similar cases will, and does, offer an actuary a pretty good idea of what would happen in one particular case.

My bet would be that the higher level mathamancy just gets to levels of perfection that are impossible with the methods we have available.

Maybe misleading; it appears that, despite its sometimes patchy nature, the Erf system is "known" to artifacts at least, and easy to compute (which is not the same as "simple"; systems with simple rules but many parts may turn out to be very tough to compute as opposed to systems with complex rules but few parts). It appears, for example, that Erfworld would be amenable to implementation on Stupidworld's MMO scene. In fact, its physics is almost geared to that purpose.

So.. obviously there are some parts of mathamancy that just cannot be recreated in a world without magic,

Which ones? I'm so far not seeing any.

[Balerion]

The best thing to do is to ask an actuary about just how accurate their predictions can get. Real businesses, with real loads of money, rely on the poor ol' downtrodden actuary to provide advice (whether that advice is actually followed is a different matter). The thing is, a good knowledge of a wealth of similar cases will, and does, offer an actuary a pretty good idea of what would happen in one particular case.

I agree, we can do amazing things with statistics. But there is a point when the model just gets too complicated; assigning a value to one question is tricky, but probably doable, so we can calculate the total value of the questions. Figuring out a base value of that information for charlie is a lot harder though; we can't simply say "he could sell it for this much", because the information is worth the expected value of him discovering it on his own, balanced out some by the value of the information to him for re-sale/planning. So the bracer, from Parson just asking it to, took in hundreds upon hundreds of possible configurations of Charlie's information gathering network, balanced out the odds of each path finding the information and at what cost. Then assigns a % chance to charlie's network being set up that way. Then takes all of that to figure out the value of the information, and decide that in x% of cases. All from ordering it to do so; the equivalent could maybe, maybe, be done with a few thousand man hours here on earth.

Maybe misleading; it appears that, despite its sometimes patchy nature, the Erf system is "known" to artifacts at least, and easy to compute (which is not the same as "simple"; systems with simple rules but many parts may turn out to be very tough to compute as opposed to systems with complex rules but few parts). It appears, for example, that Erfworld would be amenable to implementation on Stupidworld's MMO scene. In fact, its physics is almost geared to that purpose.

My biggest thing is that on earth, any calculation will have error bars; because the types of things its doing are stat based, we generally at least end up with =/-3%. With regards to simple combat, I agree. The physics is such that there wouldn't be any error, because its a well understood system to the point that its physics, not stats. But for figuring out if Charlie is responsible for goblins vanishing? to give a % chance on that without an error value says volumes (same for the above calculation). And that is where I think we have evidence of perfection beyond what we can accomplish.

[Thunder]

My biggest thing is that on earth, any calculation will have error bars; because the types of things its doing are stat based, we generally at least end up with =/-3%. With regards to simple combat, I agree. The physics is such that there wouldn't be any error, because its a well understood system to the point that its physics, not stats. But for figuring out if Charlie is responsible for goblins vanishing? to give a % chance on that without an error value says volumes (same for the above calculation). And that is where I think we have evidence of perfection beyond what we can accomplish.

that or its just lazy, i mean its actually easier for me to say this has a 13% chance rather than a 13 +/-5 % chance. also were asuming the calculator is actually correct

[drachefly]

Baaawww, and here I was hoping someone was asking for help with their homework on Abstract Algebra or something.

Awww, it would have been so appropriate - Wanda just had to deal with Abelian gropes.

[BLANDCorporatio]

She wasn't feeling particularly associative though.

[drachefly]

Yeah, but they definitely commuted - after all, she found their inverse and applied it. No more gropes.

[BLANDCorporatio]

Quite so, that grope action wasn't right.

[Kaed]

Enough with the goddamn math puns oh my god. xD

I'd say this actually might be a thread to do it in but they show up in EVERY thread D:<

[drachefly]

Quite so, that grope action wasn't right.

I had thought there was a problem because the grope would fail closure - no grope is not, after all, a grope. But the fellows, being field units, formed a ring around her. Then she's the additive identity and the grope action is an attempt to multiply. A lousy attempt, that violates Erf biology, but no more so than the homologies formed by considering the gropers in respect to the fellows (cf. Wanda).

[BLANDCorporatio]

Funny you should mention homologies; maybe that's how Wanda developped an interest in chains. Indeed, she seems quite adept at that, as Jillian has commented that, compared to Wanda, Vanna is not capable of a proper sub grope.

Enough with the goddamn math puns oh my god. xD

We haven't managed to get "symplectic" in yet. The only one I can think of now is that symplectic puns are, well, symplectic, but it only works in Greek. If at all.