Is too!

You've omitted some important variables in your statistical analysis. All I'm saying.

If you swing one sword that does y damage and have a crit rate of x, you end up a damage rate of z. If you have two swords that do the same y damage, and have the same crit rate of x, you end up with a total damage rate of 2z, though your crit rate for both swords averaged together is still x. If you have two swords that do the same y damage and have a crit rate of x, and deal a total damage of 2z, plus a beast that does f damage, and has the same crit rate of x and ends up with a total damage of g, you end up with a damage rate of 2z + g.

The crit rate for all three attacks averaged together is still the same x. However the damage is higher because you're dealing the SAME AVERAGE NUMBER OF CRITS, BUT FROM MORE SOURCES.

I can't really make it any simpler than that.

Try it with actual numbers in place of the letters. Account for the randomness. I'll stand on my head. Maybe I'd see it then.

That's not how probability works, Zarquan.

Let's say your swings do 100 damage, your beast attack does 10 damage, and you crit 50% of the time for double damage.

Without beast attacks:

1.5(100) + 1.5(100) = 300 avg damage

With beast attacks:

1.5(100) + 1.5(100) + 1.5(10) = 315 avg damage

The presence of the beast attack doesn't reduce the damage of the other two swings, even if it doesn't do much to help.

???
Profit.

Provide data to prove the statistics are skewed from random.

Standing on my head isn't helping. Where in that formula do the levels of the critical hits come in? What about the variances in balance times? I will admit that I'm wrong... once I understand exactly why. That said, I don't really care because I am better served by my beast healing me, anyway.

x average damage of a single bashing attack
x any other boosts to damage (e.g. DMP)
x number of attacks you can make per second
= average DPS of a single bashing attack

A beast attack would have the exact same formula, just replace 'bashing attack' with 'beast bashing attack'.

So you average total DPS would be 2 x average DPS of a single swing + average DPS of a beast attack

This is what theory says. My anecdotal experiences seems to contradict it, which was the reason why I wanted to collect data and then determine if there is any statistically significant difference with my critical rates and if it was affecting my damage output.

Do you by wasted crit mean that if one more of your crits would take out the mob, but your beast hits it and takes off most hp, and then you crit, that the last crit is mostly wasted? Because even if that last bit of hp is shaved off by a crit that you do, that does not reduce your chance for the first hit on the next mob to be a crit at all.

A massive critical hit on a mostly dead thing is one form of a wasted critical hit, but in this case it's more about the beast getting a massive critical hit and I get bupkis. Statistics will show that your probability of getting that same massive critical hit multiple times in a row decreases with each hit, and if it doesn't, then anecdotal evidence should surely be enough.

I've gotten the same critical hit on my beast attack and the subsequent standard cudgel hit I deal out, even on the higher end of critical hits. Fighting alongside another player that gets a lot of crits while I don't would make it seem like they are stealing my crits, but that's just not the case. You can apply that to all sorts of things (read being blown out of trees, I swear it happens to me more then anyone else).

I do agree that getting a beast crit then a massive cudgel crit is really lame, because you feel gypped out of a lot of damage (see report 436 and hope). However, unless you have some overwhelming evidence that you actually have some sort of crit cooldown or ration (which all evidence points against), I don't understand why you refuse to believe that (mostly) random crits are just that, completly (mostly) random. The mostly only referring to the inability for an algorithm to be 100% random.

I don't know. I've gotten double WSCs and triple WSCs often enough that I don't think your chance for a WSC on the second hit is any worse than on the first.

This is like arguing that if you flip a coin twice, the chance it will be heads on the second flip will be smaller if it was heads on the first.

You'd think that, but a coin is a little different from a computer RNG, especially considering different algorithms or seeds can be used.

Eh, you have a point there, but I'd imagine that the differences aren't large enough to tell the difference, and wouldn't in the long run matter much. For the amount of rng stuff that floats around Lusternia, I can't see them using something with such a transparent trend that is consistent enough to change any random results.

Yes.

Seriously, what? Your pet getting (or not getting) a critical has no bearing on you getting it or not, there is no algorithm that sadistically states "Hah you just got a crit so we won't let you get another" (or we dual-wielders would never get double crits, which we do all the blessed time-- and those poor Monks would be pathetic bashers!).

The math is simple. Even with streaky RNGs, over a long period of time it is more damage than not using it. If you don't want to do it because it's annoying to make it work through two weapon attacks that's one thing, but claiming that it eats critical hits is kind of... uh. Well, just look at Monks.

Its like this:

You have a bag with 100 marbles in it. There are, lets say, 85 white marbles, and 15 black marbles. You reach in to the bag, and have a 15% chance of producing a black marble.

If you do not remove the marble from the bag and keep reaching in, you will have a 15% chance with every reach to pull out a black marble.

Unless there is some mystery code in the works, this shouldn't change whether you're a mage, or some sort of crazy octopus monk.

Now, I can see why, and have emotionally FELT like my beast getting a critical "screwed me out of a real critical hit". But, lets create a hypothetical world where that particular draw of the statistical lottery was fixed in order, and had the beast NOT been attacking, your regular weapon would have received that number, and thus gotten the hit. This is what we FEEL like when the beast gets the annihilating hit.

However, even in this hypothetical, what you aren't then seeing is the critical hit that YOU got because your beast ate the "normal" roll of the dice before you swung. That's just human nature.

So even if you don't buy the statistical argument that the beast attack doesn't matter for your crit rate on normal hits, even within a scenario where it does "eat" the critical for your regular hits, by the same logical structure, the beast attack can ALSO eat the regular attacks, letting you get crits with your regular weapons you wouldn't normally have gotten.

And I can't think of any real situation where your beast makes you take an extra hit on a target you wouldn't otherwise have needed. It can make you waste a bit more a crit you'd already have overkilled with, but at the same time I've had Laurelin kill things so that my normal attack hit the next guy in line, which is nice.

Allow me to run another proposal that might be easier to understand.

See, Z's argument that you get less damage using a beast when you tally up is correct, but ONLY if you measure by the number of attacks.

Assuming 1000 attacks with beast, and 1000 without, then the 1000 attacks without the beast will have a higher damage total than the set with beast attacks, because you count the beast attacks as equal to the regular attacks, despite their far lower damage potential.

What should be done is to look at the damage over time instead, and check the eventual damage result after 100 minutes with beast attack, and 100 minutes without. The damage result with a beast attacking should be higher.

Or to put it like this, if the beast "steals" your criticals, it must also steal your "non-criticals", thus cancelling it out in the long run.

Now if beast attack is worth it compared to other beastmastery skills, that is an entirely different question.

For further reading. The goat problem

I'll buy that for a dollar.