EVE General Discussion

risk is difficult do define in eve terms.

if you do exploration in low with a t3 you have the risk to die in a catecamp
if you plex in FW with a 1mil frig and an alpha clone you have the risk to get cought when you fall in sleep but you don't care at all

so lets assume both have the same risk to die its still a huge difference in potential isk loss/reward

I measure risk in Eve in terms of how much it costs me to play versus the risk of not enjoying my Eve playtime.

Thus far, risk is zero.

This would change by a factor of 100% should any form of 'gold ammo' ever be introduced.

So... still zero?

Then you cannot measure risk vs reward, as each person quotient of what they consider fun factors into both the abstract terms of risk and reward.

The solo plexer in FW has a very nice risk to reward ratio on account of the spectacular rewards and trivial-in-comparison risks. Moreover, as mentioned, risk is about what you stand to lose and how likely that is to happen. The solo plexer's risk is very low simply because he puts very little on the line to get his payout, even if he loses every ship he sends out. You could lose a dozen t1 plexing frigates in an hour and still come out far ahead in the black.
-
On a slightly different tack, I suppose you could argue that every activity in EVE should have the roughly same risk:reward ratio, since a fixed amount of risk should yield a proportional amount of reward (this discounts things like effort and scalability). The difference would lie in the absolute numbers and the precise nature of the risks.

I don't suppose Mal Ishos could provide us with a reason for why risk and reward are subjective beyond "different strokes for different folks"?

Seems unlikely.
On an unrelated note, Lucifer recently invited me to a skiing trip on his backyard slopes.

Now, the fun part about risk perception is how it varies not just with the subject but with the context, and how much of that poor judgement is externally constructed. People are appallingly bad at estimating risk before-hand and even worse after something has happened. The former, for instance, makes people panic far less than Hollywood would have you believe and far less than they really should be doing, but the latter makes the exact same people assume that their entirely rational response to extreme situations is “panic” because that's what you're supposed to have been doing in such a situation.

…and at no point in all of that is the actual risk a factor in determining the behaviour.

The subject of risk assessment is apparently more complicated than I first thought. Fascinating, though.

As someone who has lots of spaceship likes, I am offended that you would diminish their importance.

Attempting to quantify risk vs. reward isn't some quack idea OP came up with. The finance and insurance industries are constantly trying to develop more accurate mathematical models for "risk management." Zoologists and ethologists are applying game theory (the branch of economics/mathematics) to analyze animals' decision making. And of course many books have been written on risk vs. reward balancing for casino gambling. Current risk vs. reward analysis tools are starting to incorporate fractal geometry ("chaos theory"). And also irrational behavior (where the decision maker chooses to do something that does not maximize his material gain).

Anyway, all of these people would say that OP is making a viable proposition. He defined Reward as isk/hour. What's left now is to define all the possible forms of Risk, and quantify them. A tall order, but it sounds like a great topic for somebody's thesis. "Risk vs. Reward Analysis in an Internet Spaceship Game."

(Personally I think this whole quasi-science of game theory/risk management/predictive analysis is just barking up a rope. It works in a highly controlled environment with narrow parameters, such as a roulette table or a blackjack game. Or even EVE maybe. But for a wide-open system like the world economy, there are just too many unknowns. There's no way a model can incorporate every possible risk, or accurately assign a value to it. That's one reason for the financial collapse of 2007/2008-- excessive reliance on proprietary (and little-understood) market prediction tools.

(Source on all this: The Economist magazine, over a couple of years of reading. They seem to be obsessed about this quantifying and predicting trend).

In short, you don't. It's all statistics at this point. It answers “what's the average risk for a highsec Hulk?” On the other hand, the same goes for the reward numbers that get tossed around, which are usually based on excel or IPH formulas rather than actually timing and cataloguing the time and earnings.

Arguably, there's some kind of normal distribution of behaviours, with heavy mitigation going on at one end of the spectrum and repeatedly undocking in a hostile system while wardecced at the other, and with the bulk of them being careless and/or semi-AFK. Then you can try to (probably incorrectly) place yourself on that scale and guesstimate that your risk is probably only a tenth (or whatever) of that of the AFKer and ε of the serial-undocker.