Great Weapon Aptitude: The conclusive Tests

[Asketes]

There you have it, our results. I'm aware that this isn't a perfect lab-quality test. I'm aware that there's a margin of error with only 300 testable swings, and I'm aware that there are other possible variables that could be in effect here but i think we have a pretty good layout. These tests are not perfect and we wont argue that they are, but we are confident enough to definitively argue that Great Weapon Aptitude is a completely broken Enhancement line and in fact has always been broken..

Unfortunately a 300 sample size is too small when you're working with something that has a low chance of occurring (i.e. <20% chance or so). The 95% confidence interval (the interval at which you can be 95% sure that the true rate is within, given the data you've collected) for 30 successes over 300 trials is 7.1% to 13.9% (using the Wilson score method), so it's just as possible that with GTWF the true proc rate was 8% (and you got lucky) and with GTWF + WF it was 12% (and you got unlucky). Simply put, there's too few trials to be any conclusive proof. The better way to state it is that the measurements showed the WF enhancement didn't give any appreciable difference, basically, that (for example) you're 95% certain based on the data that the enhancement, if it does work, gives < 6% or so (the confidence interval's range) of an increase in the proc rate. But if the proc rate was originally around 10%, then a 6% increase is pretty big (relative to the 10%).

If there were (for example) 2000 trials, and with it staying at around 10%, the confidence interval would narrow down to 8.8% to 11.4%, so the difference it makes would be be around 2.6%, which is a lot smaller.

there ya go

it's obviously not perfect, but why would they spend more than like 5 hours on this?

it's well enough to at least approach a pattern.

sure given millions of swings, their data may and probably would differ.

BTW, it seems you are referring to swings, as they referred to glancing blows and the procs. They swung more than 300 times each.

So what if the margin of error isn't 5%. The mean of the data is still probably close to what they found. Even if it's a bigger confidence interval, there is a big possibility that it's only a difference of a few procs.

I doubt they really factored in for confidence and error.

Yeah that's right, I know just a teeny bit about stats, admittedly not very much, but enough to Ace my class. The point is: who cares what the interval is, this is more data than we've had before; and yeah bad data might as well be no data. They took about as many precautions as they should have. This isn't their job.

+1 for trolling though

oh btw. you're math speak doesn't allow for many to follow your logic (it's not flawed, just tough to follow)

[DagazUlf]

I've suspected this for quite some time as well. Glad to see someone putting in the effort to test it out.

[PopeJual]

BTW, it seems you are referring to swings, as they referred to glancing blows and the procs. They swung more than 300 times each.

This is an important point.

oh btw. you're math speak doesn't allow for many to follow your logic (it's not flawed, just tough to follow)

This is not an important point. If you're going to discuss someone's math, then using math speak is kind of required.

[Asketes]

This is not an important point. If you're going to discuss someone's math, then using math speak is kind of required.

Maybe I should rephrase:

If they could elaborate how they worded some of the things they said, it might make it easier for some people to follow their logic path. (how they got to their conclusion)

Keep the math speak, but also put in dumb speak

or in DDO terms:
d20 and 1-20 damage

good note Pope thank you!

[tarikator]

+1 rep for you, and good answers all in all too.

So my WF Barbarian is gimped. Nice!! Will the developers do anything about these things?? It is not fun for me to play a gimp toon. Simple. And if they dont' fix this and all the other miriad of issues, then what is my incentive to keep playing this game when so many other mmorpgs out there take their player base seriously??

I wonder.....

[PopeJual]

+1 rep for you, and good answers all in all too.

So my WF Barbarian is gimped. Nice!! Will the developers do anything about these things?? It is not fun for me to play a gimp toon. Simple. And if they dont' fix this and all the other miriad of issues, then what is my incentive to keep playing this game when so many other mmorpgs out there take their player base seriously??

I wonder.....

This analysis helps you to decide what feats and enhancements are worth taking. This does not say whether your WF Barbarian is a gimp.

Your WF Barbarian did exactly as much damage per second last week as it does today. Did you think that it was a gimp before? This understanding does not make WF Barbarians better or worse. It just means that we understand better how a WF Barbarian works (or doesn't work in this case).

[eulogy098]

[QUOTE=Asketes;3156769
If they could elaborate how they worded some of the things they said, it might make it easier for some people to follow their logic path. (how they got to their conclusion)
[/QUOTE]

What part of the experiments wording do you find difficult to understand? I did my best to express our procedure in a clear enough way that anyone reading it could easily duplicate the tests. So if people do not understand how I reached my conclusions, or what exactly I did to get them- please tell me what you would like me to elaborate on.

[Lorien_the_First_One]

While not surprising, its very disappointing. Every time I miss this place and log onto the forums I see some bug that makes me sad

Nice work.

Devs... either make it work, or stop people from being able to waste APs on it.

[sirdanile]

Good job... +1...Time to respec my barbarian... i've got 6 extra ap to blow

[Asketes]

Maybe I should rephrase:

If they could elaborate how they worded some of the things they said, it might make it easier for some people to follow their logic path. (how they got to their conclusion)

Keep the math speak, but also put in dumb speak

or in DDO terms:
d20 and 1-20 damage

What part of the experiments wording do you find difficult to understand? I did my best to express our procedure in a clear enough way that anyone reading it could easily duplicate the tests. So if people do not understand how I reached my conclusions, or what exactly I did to get them- please tell me what you would like me to elaborate on.

- notice I say some people
- notice no one has responded to your post yet

I was merely implying there are a wide range of people that frequent, or infrequent, this forum.
It might make what you said easier to understand if you laid it out with less numbers and gave more examples .

I'm not trying to troll on your post; I understood it pretty well during the 2nd read through.

It's tough to get that type of info out to the crowds. I was trying to impress that while your original post was correct by it's own means; it really only succeeded in saying that all OP's work was in vain. It doesn't seem as such though.

I've stated that even if they'd worked through more glancing blow data, it's likely they would have gotten similar numbers.

As the number of tests rise, it will get closer and closer to the true number. 300 glancing blows probably means they did at least 2000+ swings (i'm not really sure how many). Even if it's not close to a 95% CL, the data they came up with is still probable. Even if it varies by 15-20%, statistically anything is possible yes. but under those 300 swings for each test, for not one single Glancing Blow proc to increase, is very unlikely. You know what unlikely means in stats.

sure, i can probably roll 300 crits in a row, but it's about as likely as I'm going to have 7 experiments result in almost identical data.

Anyways I'm done, not arguing any more.

happy day

-Perma

[FluffyCalico]

This is not an important point. If you're going to discuss someone's math, then using math speak is kind of required.

There is also something called writing for your audience. If you write something in a way that 99% of your intended audience doesn't understand then you might have passed math but you failed in both english and common sence.

[PopeJual]

There is also something called writing for your audience. If you write something in a way that 99% of your intended audience doesn't understand then you might have passed math but you failed in both english and common sence.

I'm pretty sure that the audience in that post was the original poster.

Other people were certainly free to listen in and respond if they chose, but that looked like a non-private comment that was meant for one person.

Also, "English" should be capitalized. I mention this not as a dig on your grammar, syntax, spelling or semantics, but just because I think that the rule is funny. None of the subjects taught in school are capitalized except for English (and other languages). It almost makes me think that English teachers made up that rule to make people think that languages are more important than history, mathematics, sciences or physical education.

[Asketes]

Also, "English" should be capitalized. I mention this not as a dig on your grammar, syntax, spelling or semantics, but just because I think that the rule is funny. None of the subjects taught in school are capitalized except for English (and other languages). It almost makes me think that English teachers made up that rule to make people think that languages are more important than history, mathematics, sciences or physical education.

well if it was language class it wouldn't be capitalized, but English denotes the name of the language class. Much like Statistics.

Well I can at least dream of the ability to mold grammar the way I want, right?

[PopeJual]

well if it was language class it wouldn't be capitalized, but English denotes the name of the language class. Much like Statistics.

Well I can at least dream of the ability to mold grammar the way I want, right?

I meant languages as in English, Spanish, German, French, Latin, Russian, Chinese, Dragon, Svirfneblin, etc.

[Asketes]

I meant languages as in English, Spanish, German, French, Latin, Russian, Chinese, Dragon, Svirfneblin, etc.

ya

[Vanshilar]

What i find to be most odd regarding the weapon effects is the fact that there are different quantities of Stunning blow procs and holy procs. Prior to testing i assumed that if a glance special was triggered, all weapon effects would go off on that glance , but this turns out to be untrue.

That's certainly a bit odd. I imagine it's possible that it got its own individual roll if you were activating the attack using the (active) skill, i.e. from the hotbar. I know that weapon procs (holy, of greater bane, etc.) all either occur together or none occur; however, other things (such as armor destruction for example) occur separately from those. On the other hand the developers haven't discussed much about glancing blow procs so a lot of it is speculation as to how things work.

Two *different* effects per glance were counted and averaged, and every single time they calculated to within 1.5% of each other, that being the case it's clearly much more accurate than your presumptions indicate.

My point is that 300 trials doesn't justify the use of superlatives like definitively, absolutely, and so forth:

...that we can definitively say that Warforged Great Weapon Aptitude is 100% non-functional with any rank.

...

Test 2-3 Conclusion- Great Weapon Aptitude rank 1 has absolutely no effect

...

Test 4-5 Conclusion- Great Weapon Aptitude Rank 2 is also completely non-functional

...

Test 6-7 Conclusion- Great Weapon Aptitude Rank 3 is also absolutely non-functional.

...

These tests are not perfect and we wont argue that they are, but we are confident enough to definitively argue that Great Weapon Aptitude is a completely broken Enhancement line and in fact has always been broken.

If you will be making superlative statements like that, you better have enough data to back it up. With a relatively low-probability event (~10%), you'd need quite a bit of data to be making those kinds of statements. Having a disclaimer at the end saying I know the testing wasn't perfect flies in the face of the rest of the superlatives presented and doesn't excuse their being used.

It's certainly possible that in fact the enhancements don't work; this is around #4738592 on my to-do list to test when I get enough time for it. Actually I've done some testing on glancing blow proc rates but haven't actually posted them because I felt ~2000 trials of each scenario isn't enough yet. That's why I mentioned confidence intervals because they're sort of the standard metric used to determine if a set of data is sufficient. Unfortunately, the farther away from 50% the chance of something occurring is, the more trials you have to do to get a small enough confidence interval.

The real kicker, for me at least, is that even if this enhancement line tripled your chances for special weapon effects it would still only average maybe 2 more damage per glancing swing with a min2 or equal item. Six AP and feat requirements is a heavy price for so little effect.

Well, its even heavier rite now since the GWA does nothing, but my point is that even if it worked its not very good.

That's more the point I think. The testing showed that in all likelihood the enhancements have a pretty negligible effect, basically 5-6% or less proc rate change (possibly 0%) which is going to be far less than 1% of the total DPS I would guess. Basically not worth taking.

[Do'Urden]

So disappointing on so many levels. This is a real kick in the crotch to all WF THF toons in addition to the glancing blows while moving nerf

[eulogy098]

I would recommend that this ability be redone and worked out to be a flat %increase to the damage of glancing blows rather than a boost to the proc rate of weapon effects.

This is my cognitive opinion for a few reasons- First, in order for an increase of special weapon procs to be of any DPS value at all it would have to quadruple the odds them triggering and that simply too much in regards to the odds of a Stunning blow or trip aoe. Even if rank 3 GWA gave an increase of 40% special weapon procs, it would still only be an average of maybe 3 damage per hit, and less so considering the recent Greater THF nerf.

So as it stands, even if it worked (and it doesnt) you either have an ability that is totally underpowered in regards to both combat feat procs and DPS, or you boost it so that it gives moderate-to-low dps but overpowered combat feat procs.

The only lgoical fix is to make it a flat damage increase to the glancing blows. 10%, 15%, and 20% would be reasonable. Turning that 31 damage glancing blow into 37 averages only 2-3 or so more damage averaged per maintarget swing, but it keeps in line with superior AOE damage concept and is a clear change to the mechanic. Players enjoy seeing the fruit of their investments, and seeing a flat damage increase, even though small, would be fun and appreciated by most players (i think).

in short-


I recommend the following fix for Great Weapon Aptitude:

Rank one, Increases the damage of Glancing blows by 10% (DOES NOT REQUIRE THF)

Rank two, Increases the damage of Glancing blows by 15% (DOES NOT REQUIRE IMPTHF)

Rank three, Increases the damage of Glancing blows by 20% (DOES NOT REQUIRE GTHF)

[eulogy098]

Remember, dont just read this post, open a Bug-Report ticket and report the bug! the developers have stated that they WANT spam regarding bugs, even the same bug to be reported over and over. Too lazy to type much out in the ticket? just copy and paste this-


================================================
*what type of bug are you reporting?*

-- Skills


================================================
*Please enter a short description of the bug*

-- Warforged Enhancement line: "Great Weapon Aptitude" does not have any effect at all


================================================
*How often does this problem occur?*

-- Always


================================================
*Please describe the problem in detail:*

The enhancement line for Warforged known as ""Warforged: Two Handed Fighting- Great Weapon Aptitude" does not have any effect on the odds of triggering a magical weapon effect.

The enhancement description states that it will increases the chances of producing magical weapon effects, this is non functional.
================================================


*Please give step-by-step instructions on how to recreate this problem. Please include as much detail as possible - this is the most important part of the bug report!*


Test attack any target for an extended period of time with and without the enhancement line and notice that there is no increase to the amount of magical weapon effects triggering. See this post for greater detail- http://forums.ddo.com/showthread.php?t=264963
================================================


If you want this fixed like we do, contribute by submitting a bug report. Also feel free to substitute the words "Great Weapon Aptitude" with "Barbarian Capstone- Barbarian Might". Since that enhancement has also NEVER worked, it needs bug report spam too.

[Reos]

This has certainly been an enlightening post. The question in my mind remains, are THF feats really worth it? I'm hoping they are, or I'm going to need a LR for my WF Pally...