monk ratios, primary and offhand choices

[bcbrown]

Originally Posted by DeathsSilkyMist [You must be logged in to view images. Log in or Register.]

It is true. If you fight 2 mobs, you'll probably get 1 below average fight, and 1 above average fight with 2h.

If you fight 2 mobs with 1h weapons, you'll probably get 1 below average fight and 1 above average fight. Both weapon sets will have the same probability of below-average or above-average damage: 50%.

But I think I understand the point you're trying to get at, however imperfectly your argument is articulated. I think you're trying to say that a string of bad luck can result in a time-to-kill that is longer than the CH chain can sustain the raid will wipe, and that kind of "string of bad luck" is more likely with a higher-variance weapon. Is that right?

[Troxx]

Originally Posted by DeathsSilkyMist [You must be logged in to view images. Log in or Register.]

When you are fighting 1 mob that spawns every 7 days, you don't want to get the below average fight. Unless your guild wipes and is able to try again, you won't get a second fight to average it out.

Or join a more skilled guild that won’t wipe because one monk chose to use a 2 hander and got “unlucky”. Does this happen to kittens much? Or are you just blowing smoke out your ass like usual?

[DeathsSilkyMist]

Originally Posted by bcbrown [You must be logged in to view images. Log in or Register.]

If you fight 2 mobs with 1h weapons, you'll probably get 1 below average fight and 1 above average fight. Both weapon sets will have the same probability of below-average or above-average damage: 50%.

But I think I understand the point you're trying to get at, however imperfectly your argument is articulated. I think you're trying to say that a string of bad luck can result in a time-to-kill that is longer than the CH chain can sustain the raid will wipe, and that kind of "string of bad luck" is more likely with a higher-variance weapon. Is that right?

Think of a set of weapon swings as a coin toss. I'll go back to the original example. For every 150 swings the 1h 11/18 weapons dish out (this includes dual wield), you'll get 60 swings from a 29/30 weapon in the same time interval. This means every 60 swings is 1 coin toss. Heads is above average, tails is below average.

If you only toss the coin once, you have a 50% chance that you'll get below average. However, you only have a 25% chance to roll two tails in a row. So 1h is more likely to give you a head + tails, resulting in getting the average.

I'll gladly take a 75% chance of getting the average over a 50% chance of getting below average when looking at a single fight that cannot be retried for 7 days. On a normal distribution this will even out eventually, but most people don't fight that many AoWs per year, unless you are extremely active.

[bcbrown]

Originally Posted by DeathsSilkyMist [You must be logged in to view images. Log in or Register.]

Think of a set of weapon swings as a coin toss. I'll go back to the original example. For every 150 swings the 1h 11/18 weapons dish out (this includes dual wield), you'll get 60 swings from a 29/30 weapon in the same time interval. This means every 60 swings is 1 coin toss. Heads is above average, tails is below average.

If you only toss the coin once, you have a 50% chance that you'll get below average. However, you only have a 25% chance to roll two tails in a row. So 1h is more likely to give you a head + tails, resulting in getting the average.

I'll gladly take a 75% chance of getting the average over a 50% chance of getting below average when looking at a single fight that cannot be retried for 7 days. On a normal distribution this will even out eventually, but most people don't fight that many AoWs per year, unless you are extremely active.

A couple of notes:

• If you flip a single coin, you can never get the average/expected value
• If you flip two coins the probability of getting the expected outcome is 50%, not 75%
• If you're reducing 60 swings, each with 21 possible outcomes, to a single coin flip you're already misusing statistics and probability and any further analysis will be tainted.

I'd really like to hear Ripqozko, 7thGate, Snaggles, or Solist give their reasons why they think (if they do) that 1h is better than 2h on high-ac raid targets. I understand why using low-delay 1h weapons is better than high-delay 1h weapons (damage bonus is the same in both cases and is applied after AC is taken into account), but since 2h damage bonus scales with delay I share Vivitron's conclusion/confusion.

[DeathsSilkyMist]

Originally Posted by bcbrown [You must be logged in to view images. Log in or Register.]

A couple of notes:

• If you flip a single coin, you can never get the average/expected value
• If you flip two coins the probability of getting the expected outcome is 50%, not 75%
• If you're reducing 60 swings, each with 21 possible outcomes, to a single coin flip you're already misusing statistics and probability and any further analysis will be tainted.

I'd really like to hear Ripqozko, 7thGate, Snaggles, or Solist give their reasons why they think (if they do) that 1h is better than 2h on high-ac raid targets. I understand why using low-delay 1h weapons is better than high-delay 1h weapons (damage bonus is the same in both cases and is applied after AC is taken into account), but since 2h damage bonus scales with delay I share Vivitron's conclusion/confusion.

I see you misunderstood basic statistics again. You've had this issue before in previous threads.

It is highly improbable to get the exact average in any small data set. It will be at least slighty over or slightly under most of the time. Using a coin flip as an analogy is perfectly valid in this scenario. Remember a coin can land on it's edge, but people generally just don't take that into consideration for these kinds of thought experiments.

An easier way to think about it (since coins confuse you) is to simply look at how many swings you'll get per fight. The 11/18 weapons are hitting more than twice as much as the 2h 29/30 weapon. You get two datasets worth of hit data during one fight with the 1h weapon when compared to the 2h weapon. One dataset will probably be over the average, and one will probably be under the average. The 2h weapon will only get one of those.

And yes, the chance of getting two heads or two tails in a row is 25%. Just google it. The idea is you don't want two tails. It's better to get any other combination. So if you have 3/4 combinations you desire, what percentage is that?

[bcbrown]

Originally Posted by DeathsSilkyMist [You must be logged in to view images. Log in or Register.]

I'll gladly take a 75% chance of getting the average

Originally Posted by DeathsSilkyMist [You must be logged in to view images. Log in or Register.]

And yes, the chance of getting two heads or two tails in a row is 25%. Just google it. The idea is you don't want two tails. It's better to get any other combination. So if you have 3/4 combinations you desire, what percentage is that?

If you get two heads, you did not get the average.

Originally Posted by DeathsSilkyMist [You must be logged in to view images. Log in or Register.]

It is highly improbable to get the exact average in any small data set.

It is in fact more likely to get the exact average in a small data set than a large data set. If you flip two coins there is a 50% chance of the exact average. If you flip four coins there is a 25% chance of the exact average. If you flip 10 coins there is a 0.9% chance of the exact average.

[DeathsSilkyMist]

Originally Posted by bcbrown [You must be logged in to view images. Log in or Register.]

If you get two heads, you did not get the average.

Yeah you still don't get it. I am not sure how much simpler I can make it. You need to listen please. I can only hope you are just messing with me like you have done in the past when you can't admit you are wrong.

Average Dataset = 1 full set of weapon swings in a single fight.

Dual wielding 1h 11/18 weapons gives you twice as much weapon swing data in one fight compared to a 29/30 2h weapon. That is the equivalent of two fights worth of 29/30 2h swing data.

Tails = under average dataset
Heads = over average dataset

Tails + Tails = under average dataset
Tail + heads = average dataset
Heads + Tails = average dataset
Heads + Heads = above average dataset

You have a 75% chance to get average or above average damage when you have 2 average datasets. Dual wielding 11/18 weapons gives you the equivalent of two 29/30 2h average datasets in a single fight.

Originally Posted by bcbrown [You must be logged in to view images. Log in or Register.]

It is in fact more likely to get the exact average in a small data set than a large data set. If you flip two coins there is a 50% chance of the exact average. If you flip four coins there is a 25% chance of the exact average. If you flip 10 coins there is a 0.9% chance of the exact average.

When I said small dataset I didn't mean 2 coin flips. I was talking about 1 full fight worth of swings, which can be hundreds of swings. Tens of thousands of swings is going to be closer to the average than 100 swings. This is really simple stuff.

[Duik]

OMFG coins? You are an insufferable twat that will just never. Ever, ever give up even when the war is won or lost.

Just play ffs and enjoy the game. If you can afford a great 2HB AND 2 great 1HB weapons. JUST BUY BOTH AND FIGHT SHIT WITH THEM.

If it truely is "simple stuff" you should have no trouble explaining it. Simply.


Jesus Fucking Christ.

[DeathsSilkyMist]

Originally Posted by Duik [You must be logged in to view images. Log in or Register.]

OMFG coins? You are an insufferable twat that will just never. Ever, ever give up even when the war is won or lost.

Just play ffs and enjoy the game. If you can afford a great 2HB AND 2 great 1HB weapons. JUST BUY BOTH AND FIGHT SHIT WITH THEM.

If it truely is "simple stuff" you should have no trouble explaining it. Simply.


Jesus Fucking Christ.

If you insist on trolling and spamming threads, at least try to get better at it. You've been doing this for years, and people still don't care about your nonsense.

[Ripqozko]

Originally Posted by bcbrown [You must be logged in to view images. Log in or Register.]

A couple of notes:

• If you flip a single coin, you can never get the average/expected value
• If you flip two coins the probability of getting the expected outcome is 50%, not 75%
• If you're reducing 60 swings, each with 21 possible outcomes, to a single coin flip you're already misusing statistics and probability and any further analysis will be tainted.

I'd really like to hear Ripqozko, 7thGate, Snaggles, or Solist give their reasons why they think (if they do) that 1h is better than 2h on high-ac raid targets. I understand why using low-delay 1h weapons is better than high-delay 1h weapons (damage bonus is the same in both cases and is applied after AC is taken into account), but since 2h damage bonus scales with delay I share Vivitron's conclusion/confusion.

theres a few exceptions , abashi just wins no matter what even on high ac, for ranger ive just parsed over the years better with 1h on like 3 targets, otherwise cek sword destroys 1h badly.