These forums have been archived and are now read-only.

The new forums are live and can be found at https://forums.eveonline.com/

EVE General Discussion

 
  • Topic is locked indefinitely.
 

The Great Ice Mining Interdiction: Not so Great

First post
Author
Previous Topic Next Topic
Tippia
Sunshine and Lollipops
#201 - 2013-08-21 07:30:25 UTC
S Byerley wrote:
You're somehow over simplifying the important bits and over complicating the math at the same time (I guess for the sake of homogeneity?) on top of autistic semantics.
How so? Again, the math is very very simple: probability × cost. The part you claim is complicated is applying a probability of 1 to a cost, rather than figuring out that the probability is 1, and filtering it out for later, rather than immediate addition.

Quote:
I've been over the figures with you
I think you're confusing me with someone else.

Quote:
If you're not treating it as a variable, you're using it as an offset/constant and simply calling it a risk for the sake of argument.
No, I'm calling it a risk for the reason that we're talking about risks, and the reason I'm treating it as a variable is because it is one. Again, very simple. All you're doing is hiding one of the variables in a filtering process that allows you to label them as fixed costs to be applied later — you still have to deal with the exact same variables, and the same number of them (two): probability and cost.

What I'm doing is making use of the strength and versatility of the risk concept to ensure that the total risk is calculated, rather than some half-arsed hybrid concept that tries to incorrectly label something that's unavoidable in order to to make it go away so it can be claimed that it doesn't exist, when it's still there and was there all along.

Quote:
No, just no.
Yeah, see, until you actually present an argument, you can stomp your feet and scream no as much as you like and still be wrong in every way imaginable. Which of these two options would you prefer to pursue?

Quote:
How bout I include it, but not as a risk
Then you're not including it, now are you? Rather, you're just arbitrarily leave it out. How about, instead, you include it as the risk it is, thus taking the full risk into account?
Tippia
Sunshine and Lollipops
#202 - 2013-08-21 07:38:59 UTC
S Byerley wrote:
Those points being?
The point being that all costs are risks and vice versa — that's the whole point of the risk concept.

So instead of going arbitrarily labelling something as “not a risk, but a cost” (which leads to there being no risks at all since you can put forth the same argument for all risks), we should just include all the costs and their corresponding probabilities, even if those probabilities happen to be 1.

That way, we can actually start with the proper discussion of whether the projected costs and the projected gains are out of whack, rather than this obfuscating and pointless “your gameplay is simplistic” line of argumentation mud slinging.
S Byerley
The Manhattan Engineer District
#203 - 2013-08-21 07:56:16 UTC
Quote:
I think you're confusing me with someone else.


I'm not.

Quote:
What I'm doing is making use of the strength and versatility of the risk concept to ensure that the total risk is calculated, rather than some half-arsed hybrid concept that tries to incorrectly label something that's unavoidable


Stop dancing and just give an example so I can show you why you're wrong please; "accidentally counting a 0.01% probability of incurring a given cost as a higher risk than a 100% probability of incurring the same cost" isn't detailed enough to do anything with - you have to show where you think they deviate.

Quote:
in order to to make it go away so it can be claimed that it doesn't exist, when it's still there and was there all along.


I can't be factoring it in later and claiming it doesn't exist at the same time, which is it?

Though in actuality it's neither; the constants get simplified before you do anything else.

Quote:
Which of these two options would you prefer to pursue?


With you? The latter; when in Rome.
S Byerley
The Manhattan Engineer District
#204 - 2013-08-21 08:00:26 UTC
Tippia wrote:
The point being that all costs are risks and vice versa — that's the whole point of the risk concept.


That's not a point, it's a semantic definition (and one I disagree with). Semantic definitions are axioms; you don't "poke holes" in them. You can show axioms are logically inconsistent, but you haven't.
Malcanis
Vanishing Point.
The Initiative.
#205 - 2013-08-21 08:11:28 UTC
Tippia wrote:
S Byerley wrote:
Those points being?
The point being that all costs are risks and vice versa — that's the whole point of the risk concept.

So instead of going arbitrarily labelling something as “not a risk, but a cost” (which leads to there being no risks at all since you can put forth the same argument for all risks), we should just include all the costs and their corresponding probabilities, even if those probabilities happen to be 1.

That way, we can actually start with the proper discussion of whether the projected costs and the projected gains are out of whack, rather than this obfuscating and pointless “your gameplay is simplistic” line of argumentation mud slinging.


A simple solution might be to have a chance that CONCORD doesn't turn up at all. The most obvious way of doing this is to make the chance that CONCORD appears equal the trusec of the system.

The attacker will then be taking a risk and everyone will be happy!

"Just remember later that I warned against any change to jump ranges or fatigue. You earned whats coming."

Grath Telkin, 11.10.2016
Tippia
Sunshine and Lollipops
#206 - 2013-08-21 08:24:52 UTC  |  Edited by: Tippia
S Byerley wrote:
Stop dancing and just give an example so I can show you why you're wrong please; "accidentally counting a 0.01% probability of incurring a given cost as a higher risk than a 100% probability of incurring the same cost" isn't detailed enough to do anything with - you have to show where you think they deviate.
It has all the details it needs, but if you want an example, how about this one? Which scenario offers the higher risk:

· Shooting freighters on the Jita gate in Perimeter in a pimp Machariel (p = 1, c = 2bn ISK), or
· Farming missions in Irjunen in a pimp Machariel (p = 0.0001, c = 2bn ISK)?

Apparently, if the gank victims are to be believed, it's the latter with its projected loss of 200k ISK, rather than the former with its projected loss of 2bn ISK.

Quote:
I can't be factoring it in later and claiming it doesn't exist at the same time
…and yet, that's exactly what people are trying to do, which is why it's such a nonsensical claim.

Quote:
The latter; when in Rome.
So, the former then. Ok, so where's your argument?

Quote:
That's not a point, it's a semantic definition
No, it's actually the entire point of the concept and the point of what I've written so far. The risk concept lets us use all kinds of costs regardless of where, when, how, and how often they occur. It means we can express costs as risks and risks at costs, depending on which word (and perspective or mindset) you prefer to use, but either way, we have to actually include all the costs and probabilities to get the right number at the end. Labelling something as a cost does not mean it's not a risk, and trying to claim that something is risk-free because it has a component that has a probability of 1 is both incorrect and dishonest.

You can disagree with it as much as you like, but if you do, you need to stay out of discussions of risk.

Quote:
Semantic definitions are axioms; you don't "poke holes" in them. You can show axioms are logically inconsistent, but you haven't.
…because that would be your job, since apparently you think that my point is a semantic definition. I presume you're interested in poking holes in it since you disagree with it.
S Byerley
The Manhattan Engineer District
#207 - 2013-08-21 08:25:05 UTC
Malcanis wrote:
A simple solution might be to have a chance that CONCORD doesn't turn up at all. The most obvious way of doing this is to make the chance that CONCORD appears equal the trusec of the system.

The attacker will then be taking a risk and everyone will be happy!


S Byerley wrote:
The problem is that suicide ganking (or at least certain forms) is risk-free, highly profitable, and requires minimal investment. It's the combination that makes it unbalanced - because Eve generally balances activities around the three.

Your example generates risk, but ignores the associated increase in profit - thus the nonsensical conclusion.


Why you always so derivative Malcanis? Straight
Lady Areola Fappington
#208 - 2013-08-21 08:33:55 UTC
Malcanis wrote:
Tippia wrote:
S Byerley wrote:
Those points being?
The point being that all costs are risks and vice versa — that's the whole point of the risk concept.

So instead of going arbitrarily labelling something as “not a risk, but a cost” (which leads to there being no risks at all since you can put forth the same argument for all risks), we should just include all the costs and their corresponding probabilities, even if those probabilities happen to be 1.

That way, we can actually start with the proper discussion of whether the projected costs and the projected gains are out of whack, rather than this obfuscating and pointless “your gameplay is simplistic” line of argumentation mud slinging.


A simple solution might be to have a chance that CONCORD doesn't turn up at all. The most obvious way of doing this is to make the chance that CONCORD appears equal the trusec of the system.

The attacker will then be taking a risk and everyone will be happy!



I've always advocated that the sec status of a system should also be the percentage chance that Concord shows up. If you advocate for that, Mal, I'll have ALL your babies.

Won't even ask for child support.

7.2 CAN I AVOID PVP COMPLETELY? No; there are no systems or locations in New Eden where PvP may be completely avoided. --Eve New Player Guide
S Byerley
The Manhattan Engineer District
#209 - 2013-08-21 08:43:25 UTC
Tippia wrote:
Which scenario offers the higher risk:

· Shooting freighters on the Jita gate in Perimeter in a pimp Machariel (p = 1, c = 2bn ISK), or
· Farming missions in Irjunen in a pimp Machariel (p = 0.0001, c = 2bn ISK)?

Apparently, if the gank victims are to be believed, it's the latter with its projected loss of 200k ISK, rather than the former with its projected loss of 2bn ISK.


The victims (and everyone else) are correct assuming sensible definitions. However, you'd be ill-advised to pursue an activity based on risk without considering cost and profit.

Quote:
…and yet, that's exactly what people are trying to do, which is why it's such a nonsensical claim.


Can't blame me for what other people do.

Quote:
So, the former then. Ok, so where's your argument?


Nope. The latter.

Quote:
No, it's actually the entire point of the concept and the point of what I've written so far.


Then you're wasting your bits trying to prove it; axioms can't be proven - if Math and Philosophy are to be believed.

Quote:
…because that would be your job, since apparently you think that my point is a semantic definition. I presume you're interested in poking holes in it since you disagree with it.


You're arguing a definition of "risk", which is a semantic argument by definition - unless you'd like to contest those definitions as well. I accept no responsibility for changing how you choose to define things.
Ludi Burek
The Player Haters Corp
#210 - 2013-08-21 08:52:32 UTC
There is always the risk that the ganker(s) may not receive any tears, rage and logical fallacies thrown at them as reward for their efforts, although if one is to pay close attention to this thread this risk appears to be low.
Tippia
Sunshine and Lollipops
#211 - 2013-08-21 08:53:25 UTC  |  Edited by: Tippia
S Byerley wrote:
The victims (and everyone else) are correct assuming sensible definitions. However, you'd be ill-advised to pursue an activity based on risk without considering cost and profit.
So a risk of 200k ISK is greater than a risk of 2bn ISK according to this “sensible definition”? Yeah, no. That's why that particular definition is rejected as nonsensical and replaced with the actual definition of risk.

Quote:
Can't blame me for what other people do.
I'm not. I'm explaining to you where the opposition to their claims come from.

Quote:
Then you're wasting your bits trying to prove it
Good news: I'm not trying to prove it. I'm presenting it as the way risk is calculated, and then it's up to you to prove that it doesn't work. So far, you haven't.

Quote:
You're arguing a definition of "risk", which is a semantic argument by definition - unless you'd like to contest those definitions as well.
…and you have not been able to show that it is logically inconsistent, whereas the notion that costs are not risks immediately leads to the paradoxical conclusion that an near-infinitely smaller probability can lead to an infinitely higher risk even when the cost is left constant, which is… well, I don't know what that is but it sure isn't a useful definition of risk. Maybe it could be used to measure fear or anxiety, since the mind has this funny habit of normalising common risks.
Malcanis
Vanishing Point.
The Initiative.
#212 - 2013-08-21 08:55:39 UTC
S Byerley wrote:
Malcanis wrote:
A simple solution might be to have a chance that CONCORD doesn't turn up at all. The most obvious way of doing this is to make the chance that CONCORD appears equal the trusec of the system.

The attacker will then be taking a risk and everyone will be happy!


S Byerley wrote:
The problem is that suicide ganking (or at least certain forms) is risk-free, highly profitable, and requires minimal investment. It's the combination that makes it unbalanced - because Eve generally balances activities around the three.

Your example generates risk, but ignores the associated increase in profit - thus the nonsensical conclusion.


Why you always so derivative Malcanis? Straight


It's integral to my character.

"Just remember later that I warned against any change to jump ranges or fatigue. You earned whats coming."

Grath Telkin, 11.10.2016
Tauranon
Weeesearch
CAStabouts
#213 - 2013-08-21 08:58:15 UTC
Infinity Ziona wrote:
S Byerley wrote:
Tippia wrote:
As always, risk = cost × probability.
Just because the probability is 1 doesn't mean it's not a risk — it just means that the risk is so hight that it has the same value as the cost.


Too lazy to read the entirety; are you still arguing that suicide cats (which work out to about the isk/hour of BS ammo) are inherently risky?

Like most people, I tend not to factor ammo costs into my risk assessment.

You are correct. When suicide ganking the ship is the expendable ammo. A risk is not a risk if the outcome is certain. A risk requires an element of chance.

So there is no risk in ganking. Just expenditure.


The outcome isn't certain. One thing that does happen is the target goes to ground when you are setting up, even if its not because they saw the impending gank. Mack filling is a trigger for people to make decisions like log off after all. In which case you risk time passing without fun and without making isk (either from bits of dead exhumer or by opportunity cost of not doing other activities).
Malcanis
Vanishing Point.
The Initiative.
#214 - 2013-08-21 09:02:28 UTC
When will people learn not to argue with Tippia?

Just because there's a 100% chance you'll lose doesn't mean you're not wasting your time because there's no risk she'll lose.

"Just remember later that I warned against any change to jump ranges or fatigue. You earned whats coming."

Grath Telkin, 11.10.2016
S Byerley
The Manhattan Engineer District
#215 - 2013-08-21 09:13:10 UTC
Tippia wrote:
So a risk of 200k ISK is greater than a risk of 2bn ISK according to this “sensible definition”? Yeah, no. That's why that particular definition is rejected as nonsensical and replaced with the actual definition of risk.


I'm separating cost vs. risk and you're not; when you correctly factor that difference in: 200k < 0 + 2b, or 200k > 2b -2b

Whichever you prefer, it's all perfectly consistent.

Quote:
Good news: I'm not trying to prove it. I'm presenting it as the way risk is calculated, and then it's up to you to prove that it doesn't work. So far, you haven't.


You're presenting the way you want to define/calculate risk. If you want to argue that's the commonly accepted way, it's another matter entirely; you've offered no evidence to that effect.

Quote:
…and you have not been able to show that it is logically inconsistent,


I've said several times now, I have no desire to.

Quote:
whereas the notion that costs are not risks immediately leads to the paradoxical conclusion that an near-infinitely smaller probability can lead to an infinitely higher risk even when the cost is left constant


You seem confused. If you make the outcome uncertain, that cost becomes a risk with associated probability - the cost+risk term changes continuously in response. If you're bothered by the discontinuity between probable and definite outcomes, that's a separate philosophical and subjective issue.
baltec1
Bat Country
Pandemic Horde
#216 - 2013-08-21 09:13:50 UTC
Malcanis wrote:
When will people learn not to argue with Tippia?



Same reason why I **** into the wind. One day I might win.
S Byerley
The Manhattan Engineer District
#217 - 2013-08-21 09:17:18 UTC  |  Edited by: S Byerley
Malcanis wrote:
When will people learn not to argue with Tippia?

Just because there's a 100% chance you'll lose doesn't mean you're not wasting your time because there's no risk she'll lose.


I find talking to walls therapeutic in moderation, personally.

The time spent is a certainty, a cost if you will, but the gain is subject to chance.
baltec1
Bat Country
Pandemic Horde
#218 - 2013-08-21 09:22:19 UTC
So I guess we shouldn't point out the risks of the gank failing or the loot fairy giving the middle finger.
S Byerley
The Manhattan Engineer District
#219 - 2013-08-21 09:27:14 UTC
baltec1 wrote:
So I guess we shouldn't point out the risks of the gank failing or the loot fairy giving the middle finger.


You can if you want; I'd throw it under profit though - deviation maybe under investment (as in, you have to do it for longer to see the same consistency in payout)

How often do your ganks fail?
Tippia
Sunshine and Lollipops
#220 - 2013-08-21 09:34:35 UTC  |  Edited by: Tippia
S Byerley wrote:
I'm separating cost vs. risk and you're not; when you correctly factor that difference in: 200k < 0 + 2b, or 200k > 2b -2b
…and the point is that separating the two is not correct, because the only difference is how high the probability is. In one case, the risk is 200k, all included; in the other case, the risk is 2bn, all included. Any claim that the latter is lower is nonsensical.

Quote:
Whichever you prefer, it's all perfectly consistent.
I prefer the risk to be reported in full and without parts of it arbitrarily removed for no good reason, at which point it's not particularly consistent to claim that 200k > 2bn.

Quote:
You're presenting the way you want to define/calculate risk. If you want to argue that's the commonly accepted way, it's another matter entirely; you've offered no evidence to that effect.
Look up ISO 31000.

Quote:
You seem confused. If you make the outcome uncertain, that cost becomes a risk with associated probability - the cost+risk term changes continuously in response.
No. It's not me being confused — it's you mixing up the terms and confusing yourself. If the outcome is uncertain, the cost becomes a projection based on the risk (equal to cost × probability). A risk, meanwhile, is a cost with an associated probability. Once you start talking about risks, you can (and should) include all costs with those associated probabilities, even in the cases where p=1.

If there are no uncertainties at all, you can take the shortcut (or, pehaps more accurately, not take the longer route) and just skip the conversion-to-risk-via-probability-association stop entirely. However, it is still entirely correct to talk about it in terms of risk. If there are any uncertainties, you should stop talking about costs entirely and only really discuss risks (or, at most, talk about projected costs, which is just syntactic sugar for risk).

Oh, and since costs are risks and vice versa, there are no discontinuities (or at least no more than there are in the measure of probability — I haven't kept up enough to know if anyone is arguing that probability is a only finitely divisible).

Quote:
You can if you want; I'd throw it under profit though
I suppose it would annoy you immensely if I started talking about profits as risks as well?