Inner Peace (Through Superior Firepower) – Episode 008

[Ilari]

And what balances out a thinkamancer?

Stanley?

Do the arkentools create huge imbalances that have to be paid back, or do they literally break the balance of the game? If life is mostly stolen from the same side when a croakamancer uses her power, it probably creates a net loss, as the uncroaked decay away anyway. Or maybe that itself is the balancing act, but I think not. Either way, the pliers creates units that dont decay, Im going to lean towards the theory that they break the balance. But its even scarier if they dont and the life the pliers create has to be balanced somewhere.

I, myself am leaning towards the "break the balance" theory. The Arkentools are dev tools. They are supposedly what the Titans used to create Erf, and you can't create a world simply by rearranging that same world's resources; they don't exist yet! Combine that with their 3D CG art style which marks them as blatantly out-of-place (Moreso than Parson, who gets the same hand-drawn art style as the rest of Erf despite being from Stupidworld) and I get the sense that they came from elsewhere. If there is a cost for using an Arkentool, it probably gets paid by Arkworld.

[GrandReaper]

Why are people so concerned with Luckamancy infringing upon Mathemancy's territory? Let's see if I can counterpoint this in a more organized fashion than I tend to.

Firstly, since a universe with perfect probability (infinite outcomes, drawing with replacement, etc) is always uniform regardless, let us reject this as the law for Erfworld for discussions sake. This is reasonable as, with Erfworld being a world with "game physics" it could just as simply work as a deck of cards without replacement, then reshuffled when done, or a computer's "random" number generator which, at some level, is always deterministic.

Now, working under the assumption that actions come from a finite "pool" of possibility does not undermine Mathemancy, even if we believe that the "Deck Size" or the periodicity of the RNG is known to a Mathemancer. In those cases simply not knowing the number of cards left in the "deck" currently, or the algorithm behind the RNG would still leave them in the dark, prediction wise.

Thus, a Luckamancer "drawing" cards out of order, or swapping values in the RNG table would not change a Mathemancer's approach to determining probabilities unless:

1) The Mathemancer was aware of ALL rolls used thus far in this "cycle" of probability
2) The size of the "deck" or periodicity of the RNG is small enough that the changing of several rolls significantly effect long term probabilities.

So, the only time 2) would likely apply would be near the end of the "cycle" of probability, before the deck is reshuffled or the RNG starts to repeat. However, to disallow Mathemancy perfectly predicting the future we already had to reject the possibility that a Mathemancer could know when the cycle was ending. Furthermore, 1) is only possible at the expediture of what I would imagine an extremely high amount of juice, turning the Mathemancer into little more than a life-to-number translation machine, not to mention likely monopolizing a Thinkamancer as well.

I don't think I was able to make that last point as clearly as I wanted, so I'll just leave my understanding and see what people think: Even if the world has finite outcomes (that must be recycled by some mechanism or else the game has an end by "running out of cards", an entirely different problem I won't address here) simply not knowing how many are left in the cycle means that all probabilities must be calculated based on the initial condition/case - the number of outcomes that succeed initially / the total number of outcomes initially.

In other words, even though conditional probability WOULD apply, so much information is hidden that a Mathemancer will always have to use simple probability regardless. If the pool is "large enough" then this would still be a highly reliable art.

[Kizmet]

tinfoil hat time...

1) Clay was told specifically to NOT boost Wanda by order of Delphie (Nov 14th update), but he boosting other warlords.
2) Luckamancy takes luck/good rolls/crits from non-boosted and redistributes it to the boosted.
3) Clay said that Delphie said... "your the only unit that does not need [boosting], you have some kind of fate crap on you" (Nov 14th update)
4) Perhaps Delphie has it in for Wanda? Would a Predictamancer resent fate? Could you use Luckamancy to drain a specific allied unit's luck by boosting the rest of the side?

PS
5) Wanda is a have-not. Therefore, Wanda is the 99%

[Kreistor]

The only problem I have with this is that it more or less runs counter to the information that we got at the beginning of these updates. The balance has to be paid, and I don't think one's enemies count quite so much as one's self or side when it comes time to pay.

Does a Shockmancer get randomly hit by lightning every once in a while to balance his power? Sizemore gets crushed by rocks for all his destruction? The volcano? What random event happened to balance that account?

No, balance is certainly not a common thread in Erfworld. Some people will always find a way to see balance, but if you take everything and toss it on the pile, you won't find it. Where's the balance to violence? Should be Peace, but that's nowhere. Where's the balance to death? It should be birth, but we've got "Pop!" instead.

[gobe]

ninja'd regarding the RNG comment above. I generally agree with all of GrandReaper's comments and I was going to post about the RNG hypothesis myself as that was my first thought after reading the comic.

I think that Erfworld is meant to represent the life of units inside a video game, as though it was going on in a real world. But with all of the characteristics of most video games intact, including actions being determined by two things: unit stats (including all the modifiers of course..) and randomness. And randomness in video games comes from a (semi-)random number generator.

To be consistent with Clay's explanation, I think that it is possible that the RNG uses a different seed (a different pool of numbers with its own ordering) for each side (so that they all get the same amount of good and bad rolls overall, perhaps). And Luckamancers would only move the numbers that are drawn from the RNG around, perhaps simply change their ordering locally, to ensure that specific actions of their choosing receive a boost, but of course some other actions OF THE SAME SIDE will get de-boosted somewhere down the line. Perhaps the Luckamancer might not be able to choose who gets deboosted, since it's only future actions that will draw the misplaced low scores.

That is, unless you also use a very smart predictamancer, or even link them together! A mathamancer would also tremendously increase the effectiveness of a luckamancer, telling him where more luck is needed and by what amount, so that he would never overdo it and so reduce other units' luck much less.

P.S. First post, long-time reader of this forum. Hello everyone.

[BLANDCorporatio]

In fact, the 'not if it is hopeless' clause may exempt predictamancers from duty altogether. Huh.

Huh indeed. Nice catch.

Now, working under the assumption that actions come from a finite "pool" of possibility does not undermine Mathemancy, even if we believe that the "Deck Size" or the periodicity of the RNG is known to a Mathemancer. In those cases simply not knowing the number of cards left in the "deck" currently, or the algorithm behind the RNG would still leave them in the dark, prediction wise.

Nope, not if you don't require predictions to be individually correct ("this exact piker will survive this battle"), but instead allow for a more lumped, statistical prediction ("70% of our pikers will survive").

The issue of initial condition of the random generator is also irelevant. Either the random generator is, or statistically convincingly impersonates, an ergodic process (parameters like average, spread, probability distribution are constant for all time), or some statistic test will discern variations in its parameters and allow a Mathamancer to detect where in time the process is. The first case is typical for (pseudo-)RNGs, and makes the initial condition irelevant. The second case makes the initial condition something that can be estimated and accounted for in the prediction. So either a Mathamancer can find out what the initial condition is, or will not care because it will not affect (statistical) predictions.

The bottom line in this is, if Mathamancy is able to deliver "predictions" like "70% of our Pikers will survive", then apparently even a Luckamancer expects about 70% of Pikers to survive. No Luckamancer influence is going to nudge that too far in one direction or the other. And one very elegant way to improve one of the Piker's odds is to shuffle rolls around.

[CNagy]

Does a Shockmancer get randomly hit by lightning every once in a while to balance his power? Sizemore gets crushed by rocks for all his destruction? The volcano? What random event happened to balance that account?

No, balance is certainly not a common thread in Erfworld. Some people will always find a way to see balance, but if you take everything and toss it on the pile, you won't find it. Where's the balance to violence? Should be Peace, but that's nowhere. Where's the balance to death? It should be birth, but we've got "Pop!" instead.

I can't answer your question about Shockamancers and Sizemore, except to say that if there is a balance waiting on Sizemore, it could still be waiting in the wings. There is quite simply not enough information to know what, if any, consequences await the users of other magic specialties. The point I'm focusing on is: no violation of Numbers. Life is all Numbers, remember, and the equation of life balances out at 0. If you don't discount what Clay is revealing out of hand, then there is also a balance to Luckamancy. Your speculation that a Master class Luckamancer can boost an ally and pay for that boost by jinxing an enemy doesn't seem to reflect any of the balance we've been told exists unless Erf doesn't care about sides. From Clay's explanation of how their side has been faring, luck-wise, and his realization as to why Luckamancers weren't supposed to give that little secret away, it seems Erf does care about sides.

Speculation time: the first update says you can travel to the flat country that is equilibrium at any time, at the cost of your life. It also says that Wanda carries a balance that will have to be paid. This, I think, is where Fate steps in--it will not let Wanda die before a certain point. If her death alone were enough to address the balance, then there would be no difference between her and any other unit. So she cannot die before Fate allows, because dying would leave the equation of her life at 1 = 0 (not necessarily 1, but it illustrates the point.) Some understanding of this would explain her fatalistic view towards outcomes.

Fun Speculation time: Weirdomancy is also on the Numbers axis. When a Weirdomancer removes a special from unit temporarily, does some random unit elsewhere gain it temporarily? While the image of a wild critter somewhere suddenly gaining the ability to soar only to splat when the weirdomancy wore off is an amusing one, barbarian cities (frozen until an attacker enters) would be a convenient place to temporarily store specials.

[rlc]

I, myself am leaning towards the "break the balance" theory. The Arkentools are dev tools. They are supposedly what the Titans used to create Erf, and you can't create a world simply by rearranging that same world's resources; they don't exist yet!

This is my thought also, but I can also see the possibility that its a debt that has to be repaid. Which would mean Charlie might have a huge balance against him.

But, yeah, I think you have it right.

[gameboy1234]

I would disagree with that being where the luck came from; recall that same battle: Artemis was shocked at how effectively her knights were killing the dragons."Did any of them even miss?" and she was criting all over the bloody place; except when it came to killing Sylvia and Captain archer. I think the price for Sylvia's life was the 6 dwagons who got slaughtered, doing far less damage than they would normally have been expected to.

Oh, good point. I did not go back and re-read that update, so you are probably correct. Slyvia seems to have stolen her luck from her dragons. And I guess possibly, Artemis herself a bit, there at the end.

[GrandReaper]

Now, working under the assumption that actions come from a finite "pool" of possibility does not undermine Mathemancy, even if we believe that the "Deck Size" or the periodicity of the RNG is known to a Mathemancer. In those cases simply not knowing the number of cards left in the "deck" currently, or the algorithm behind the RNG would still leave them in the dark, prediction wise.

Nope, not if you don't require predictions to be individually correct ("this exact piker will survive this battle"), but instead allow for a more lumped, statistical prediction ("70% of our pikers will survive").

The issue of initial condition of the random generator is also irelevant. Either the random generator is, or statistically convincingly impersonates, an ergodic process (parameters like average, spread, probability distribution are constant for all time), or some statistic test will discern variations in its parameters and allow a Mathamancer to detect where in time the process is. The first case is typical for (pseudo-)RNGs, and makes the initial condition irelevant. The second case makes the initial condition something that can be estimated and accounted for in the prediction. So either a Mathamancer can find out what the initial condition is, or will not care because it will not affect (statistical) predictions.

The bottom line in this is, if Mathamancy is able to deliver "predictions" like "70% of our Pikers will survive", then apparently even a Luckamancer expects about 70% of Pikers to survive. No Luckamancer influence is going to nudge that too far in one direction or the other. And one very elegant way to improve one of the Piker's odds is to shuffle rolls around.

Well, I'm not really here to argue, since I don't actually think we disagree, but some niggling details are standing out. Fundamentally we seem to agree that the Mathemancy will hold, regardless of any tampering from Luckamancy. However, remember that even lumped predictions like the aforementioned "70% of our pikers will survive" technically has a probability approaching 0 as the number of pikers increases (individual points in a continuous spectrum and all that). On the other end of the spectrum, when numbers are low enough we are dealing with discreet cases the variation even in a (near) ergodic process will have unpredictable pockets of probability. Thus, something along the lines of "on average" or "we should expect losses around" or even a confidence interval are all reasonable information a Mathemancer could provide (I personally love the idea of hearing "Losses should be less than 15%, 19 times out of 20" from one). Even that's not quite proper, but I still like the sound of it.

[BLANDCorporatio]

However, remember that even lumped predictions like the aforementioned "70% of our pikers will survive" technically has a probability approaching 0 as the number of pikers increases (individual points in a continuous spectrum and all that).

No, technically it doesn't. However many the pikers, they are not a continuum of points. There's more than one kind of infinity. What happens is that as the number of Pikers increases, possible (discrete) outcomes become more numerous. 2 Pikers: either both die, both survive, or the first dies the last survives or viceversa. Four pikers, well, there's now the possibility that one dies whereas three survive etc. This new possibility "eats from" the odds of there being exactly two pikers that survive.

But, as you yourself notice, statistical predictions come with confidence intervals.

On the other end of the spectrum, when numbers are low enough we are dealing with discreet cases the variation even in a (near) ergodic process will have unpredictable pockets of probability.

Aaaargh. What does that mean?

Thus, something along the lines of "on average" or "we should expect losses around" or even a confidence interval are all reasonable information a Mathemancer could provide (I personally love the idea of hearing "Losses should be less than 15%, 19 times out of 20" from one). Even that's not quite proper, but I still like the sound of it.

For once, me too. I suspect that is what Mathamancy provides. But I'd say this works better for larger numbers.

[Kreistor]

Thus, something along the lines of "on average" or "we should expect losses around" or even a confidence interval are all reasonable information a Mathemancer could provide (I personally love the idea of hearing "Losses should be less than 15%, 19 times out of 20" from one). Even that's not quite proper, but I still like the sound of it.

Noooo... confidence intervals are for estimates involving unknown populations. When you know the population (ie. you can count your troops and the enemy's), standard deviation applies, but confidence intervals are irrelevant. Confidence intervals are used when sampling from a population, and so apply to exit polls, but not to a vote once it has been counted.

[SandroTheMaster]

This is one of the most important pieces of information we got so far about Erfworld mechanics. This also explains why the unled diggers with siege towers received such a beating.

I wonder what higher level and higher aptitude Luckamancers can do to offset this, though, because if they can only increase the scale of their luckamancy, then this drawback only snowballs more and more, making Luckamancers extremely dangerous casters to a side. Though it seems they can boost a whole side... so what does THIS does to them to offset? They pay in bad luck in subsequent turns?

[Lamech]

Hmm... idea. Jinx hunters, or some other job where failing a few times doesn't really matter; if the bats can't get their bite to land the first 27 times, who cares, the deer can't escape or fight back. You don't control where the boosted luck goes, but you actually end up with a net gain* of "luck". Another trick would be if your about to score a slaughter to bless the enemy. You slaughter them, keeping only fresh guys on the front. They take a few more hits than they should, but it doesn't benefit the bad guys any. Boom! Now their luck is drained.

*Technically not, but luck spent on deer is worthless so who cares.

Also another idea: Luckamancer-ditto link. Don't steal good luck, copy it.

[SandroTheMaster]

I'm sure that if Parson is lectured by Wanda on Luckamancy, he can find a way to use it for his advantage.

[Dark Matter]

Interesting... luck which doesn't change the over all odds, and in such a manner that he can't really predict where the bad luck will go.

That's not "useless", but it is a lot less useful than it first seemed. The mind boggles at the implications. The idea that the warlord gets the bonus and the unlead get the penalty, is that just what normally happens?

If you're using and abusing luck a lot, giving yourself great luck all over the place, maybe eventually you get allies like "Sir Stab-you-in-the-back"... or if you're seriously "lucky", someone like Wanda.

[name lips]

Wow.

Looks like when a Luckamancer casts a spell, he's "stealing" those numbers from... somewhere. But he doesn't get to pick where. They tend to be stolen from his own side. But not always. It's totally unpredictable.

So you could have 2 high value warlords, and he "boosts" one of them, and the other one dies. OR a lowly stabber might die. You can't tell, because you can never tell for sure what would have happened if he hadn't boosted the warlord. Maybe that other warlord would have rolled a 1 anyway. Maybe the boosted warlord was going to roll a 4 anyway. You never know beforehand.

Unless of course you're a Predictamancer... I'm beginning to see why she keeps Clay Dice under such close watch. The Luckamancy/Predictamancy combo might be kind of scary...

[Rizban]

I have to admit that the usual story has me checking the site once or twice a week to look for updates, but this current text story about Wanda has me checking two or three times a day. I can't wait to see what happens next and find this story so very compelling.

[Sushi Database]

Okay, I've been thinking, and it's led me to sign up (finally, I've been reading since early in Book One). I think what we're hearing here is not just how Luckamancy, the magical discipline, works. This is how Natural Luckamancy works, too. And while us learning that is important, WANDA learning it is more important... because I think what Clay is describing to her is a natural, passive buff that Wanda was popped with. This is the manner in which "Fate" favors Wanda: it steals numbers from others to allow her and those that would further her cause to come out ahead.

Wanda is going to remember this, this lesson on how Luckamancy works. And she's going to go through her life noticing the little coincidences that favor her, and the simultaneous drawbacks her side suffers. She's going to realize that whenever she pushes Fate, someone else will pay the debt. And that's what's going to shape her into the Wanda we know, present day... confident, self-assured, outwardly fearless, and utterly convinced that the Fate is on her side.

[King Mir]

Some one made the analogy of luckamancy as rearranging cards, and I think that fits pretty well.

Probability in erfword is rather strange it seems. In our world, previous events have no bearing on future probability. There is no balance sheet. If you flip a coin and get heads 10 times in a row, there is a %50 chance of rolling heads again. But in Erfworld, it seems the chances would be lower. Like drawing cards form a deck, each good card drawn reduces the number of good cards left.

Did you really just use two real-world examples to make your point that Erfworld is different from the real world?

http://en.wikipedia.org/wiki/Conditional_probability

I don't get what your objection is. The rules of mathematics and probability are necessarily the same in Erfworld and our world. What's different is when they apply.