There's fundamental flaw in simulation.

[tobindax]

I just realized there's a fundamental flaw in simulation compared to formulation. Simulation does some stuff that at least partly go full circle to fundamental knowledge that is de facto given!

For example, you're given a certain crit chance. Why the f simulate it! You know it!

Anyway. flaw.

[Vlad]

Wait, what?

Usually you use simulation to accomplish more than just calculate how often you crit. :P

[haylie]

Because, for instance, even though the tooltip says a spell has a 20% crit chance, the simulation might reveal that it's actually a 15% chance. Not everything is fool-proof. In real life, there are people whose job is to constantly run tests on stuff like food and tap water, to make sure it's healthy and not E. coli infected, even when they know that in theory, it should be perfectly fine.

[Dawn]

Originally Posted by Vladinator Usually you use simulation to accomplish more than just calculate how often you crit. :P

This. ^^


The message you have entered is too short. Please lengthen your message to at least 10 characters.

[Vlad]

Just to expand my post slightly; simulation is most useful when you have several sources of data that have no constant value, hence it would be beneficial to simulate all the possible outcomes. In the end you can statistically estimate a min-max range of what ever you are after.

Indeed there is no point counting all the times you crit and when you didn't crit to calculate your crit % -it's already stated in-game. But if you let's say have a item that has a proc that triggers randomly on a spell crit and you would like to figure out how much crit % you passively have (at average that is) you'd need to simulate enough times in order to get a proper min-max range.

[Dawn]

Most simulations are done to value at least one thing against another. Or how something performs in a specific situation. Or to simply proof some fool!

[Xubera]

and also remember that statistics are just that.

20% crit chance doesnt mean every 100 hits, you have 20 crits.

Just because flipping a coin gives you a 50% chance to get heads and not tails, you could still get 10 heads in a row after 10 flips without a single tail. it doesnt change the odds, you cant say this coin has a 100% chance to get heads and a 0% to get tails.

[Crissa]

Actually, that's what it's supposed to mean.

But the method for creating hits and crits has changed over time. Simulations are of various values, many of them taking into account the quirks, and some are flat simple, not taking into account the hidden game mechanics.

At one time, 20% crit meant that 20% of your attacks would crit. This had the perverse result that you'd actually crit more often on things that you hit less frequently.

Then at another, they made it so you could have a better or worse crit. The better crits would get cut off as you missed more often - at least, that was the theory - but again, there were edge cases where it was more important to crit than hit.

But a simulation is only as good as the data it's built upon.

-Crissa