Rogue Guide by Oto

[Youfie]

Hey, i was wondering as combat daggers.

Cadaverous armor ( 8 agi, 8 strenght + 60 AP ) vs Tier 1 chest ?
Will cadaverous armor win before you get 5x set bonus?
- I also have around 30.5% crit chance unbuffed and 710 ish AP

Also, this is my current gear: http://realmplayers.com/CharacterViewer ... =Pomfittes ( with cada ) not updated yet. and i also use Rune of the guard captain ( 20 ap 1 hit )

Should i go with Muggers belt or:
http://db.vanillagaming.org/?item=14502

My current hit is 5% from talents ofc, and 7% from gear. so a total of 12%

If you have +0 Skill with your current gear, yes you should get Mugger's Belt (better than Distracting Dagger + T1 Belt since you have CHT), until you can get your hands on ACLG gloves, then CHT + T1 Belt + ACLG and you'll be set for a while .

Regarding chests, from what I calculated, Cadaverous & T1 are very close to each other wih a gear setup similar to yours. Which one is better kinda depends on the buffs you use. T1 is slightly ahead, unless you can have buffs like Druid crit / Ony's head & shit, with which Cadaverous is slightly ahead. T1's extra crit benefit more from AP (like DM 200 AP buff), but Cadaverous AP scales nicely with crit.

My calculation are probably not perfect anyway, but you get the idea : for your current gear they're basically equivalent in my opinion, depending on buffs.

And the 5/8 T1 is basically useless in PvE, it doesn't really change anything in my opinion.

I'm sorry but your maths on crit>hit after 8%hit until you reach crit cap are incorrect because you don't take in account :
-Lethality
-More sword procs thanks to hit
-More hoj procs thanks to hit
-More poisons procs thanks to hit
-More crusaders uptime thanks to hit
-More WF proc thanks to hit (horde obv), then have to take in account the AP increase thanks to WF buff
-Duration of the fight

I don't see how the duration of the fight favours either crit or hit (but I might be mistaken), and Lethality is actually helping +Crit getting ahead of +Hit.

As a matter of fact, more +Hit means more Swords / HoJ / Poison / Crusader / WF procs. But does it really counterweight the superiority of +Crit?

I mean, with two weapons of 2.8 & 1.8 speeds + SnD, you swing an average of 1.0945 hits per sec at your target with 13,5% Miss (took these numbers cause I already had them).
Gaining 2% hit (11,5% Miss) cause this number to become 1.1038 hit / sec.

This means +2% Hit causes the average time between two HoJ proc to go from 45,68s to 45,30s (assumption : 2% proc = One proc every 50 successfull hit/crit/glance on average) : isn't it very un-significant? I mean, when you compare it to crit?

(Didn't took into account Specials here to calculate HoJ proc since we're calculating benefit fro +Hit and we're assuming your already hit-capped on specials)

Maybe my math is flawed here (and it's obviously incomplete above, just did a simulation for HoJ), but I'd like to know how you modeled your own DPS gains from +Hit versus +crit for those procs, I'm very curious.

[enjoyr14]

I don't see how the duration of the fight favours either crit or hit (but I might be mistaken), and Lethality is actually helping +Crit getting ahead of +Hit.

As a matter of fact, more +Hit means more Swords / HoJ / Poison / Crusader / WF procs. But does it really counterweight the superiority of +Crit?

I mean, with two weapons of 2.8 & 1.8 speeds + SnD, you swing an average of 1.0945 hits per sec at your target with 13,5% Miss (took these numbers cause I already had them).
Gaining 2% hit (11,5% Miss) cause this number to become 1.1038 hit / sec.

This means +2% Hit causes the average time between two HoJ proc to go from 45,68s to 45,30s (assumption : 2% proc = One proc every 50 successfull hit/crit/glance on average) : isn't it very un-significant? I mean, when you compare it to crit?

(Didn't took into account Specials here to calculate HoJ proc since we're calculating benefit fro +Hit and we're assuming your already hit-capped on specials)

Maybe my math is flawed here (and it's obviously incomplete above, just did a simulation for HoJ), but I'd like to know how you modelized your own DPS gains from +Hit versus +crit for those procs, I'm very curious.

I never said that hit>crit i just said his maths was incorrect, and for the un-significant part the same goes for crit, going from 40% to 42% crit would be better than going from 10% to 12% hit? when considering all procs (sword/hoj/poisons/wf(horde)/crusader...) etc? And ye your simulation for HoJ proc is not true (1proc every 50hits) but we both know that.

For the fight duration, on a short fight crit might be better than hit because of AR, tea, BF, flurry juju, crusader uptime(#rng) etc... whereas hit might be better on a longer fight (and then it might not be true, hit could be better than crit on a shorter fight too, just because of procs which is a huge part of vanilla dps).

For example if a fight last 30sec, you'll be 50% of the time under CD, but if it last 120seconds you'll be 12,5% of the time under CD, and on a shorter fight your yellow damage should be higher (in percentage when comparing it to white hits) than the ones on a longer fight because of those CDs, and the RNG you will have plays a big roll there, but then is crit better than hit? considering a certain number of crit / hit, your gear, your spec, the actual procs that you can have, the duration of the fight etc?

I'm just saying it's incorrect because he doesn't take in account procs (and all it brings), which is a big part of our DPS, it just takes in account a number of hit and crit with a 0/0/0 spec, no particular piece of gear, no enchant, and no procs.

[Kha]

[size=200][color=#40BF00]You will need +8% hit to always hit with your special abilities against a raid boss.

wrong. the 8% hitcap for yellows is only for 300 weapon skill.

[Kha]

Distracting Dagger is just an example, its best when you don’t have Mugger’s Belt[i], otherwise every other fast Dagger is nice to have with a high dps rate.

wrong again.

[pomfittes]

If you have +0 Skill with your current gear, yes you should get Mugger's Belt (better than Distracting Dagger + T1 Belt since you have CHT), until you can get your hands on ACLG gloves, then CHT + T1 Belt + ACLG and you'll be set for a while

Thank you for the reply, i currently have 2 epic daggers, and i dont use distracting. My question is that with 12% Hit total, can i ignore the +5 from mugger belt and go with an agi belt? or should i stick to +5 mugger belt while i get the hands with +5 daggers from mc ?

[Solmyr]

It's so endearing watching dagger rogues quarrel .

Do carry on.

[Youfie]

I don't see how the duration of the fight favours either crit or hit (but I might be mistaken), and Lethality is actually helping +Crit getting ahead of +Hit.

As a matter of fact, more +Hit means more Swords / HoJ / Poison / Crusader / WF procs. But does it really counterweight the superiority of +Crit?

I mean, with two weapons of 2.8 & 1.8 speeds + SnD, you swing an average of 1.0945 hits per sec at your target with 13,5% Miss (took these numbers cause I already had them).
Gaining 2% hit (11,5% Miss) cause this number to become 1.1038 hit / sec.

This means +2% Hit causes the average time between two HoJ proc to go from 45,68s to 45,30s (assumption : 2% proc = One proc every 50 successfull hit/crit/glance on average) : isn't it very un-significant? I mean, when you compare it to crit?

(Didn't took into account Specials here to calculate HoJ proc since we're calculating benefit fro +Hit and we're assuming your already hit-capped on specials)

Maybe my math is flawed here (and it's obviously incomplete above, just did a simulation for HoJ), but I'd like to know how you modelized your own DPS gains from +Hit versus +crit for those procs, I'm very curious.

I never said that hit>crit i just said his maths was incorrect, and for the un-significant part the same goes for crit, going from 40% to 42% crit would be better than going from 10% to 12% hit? when considering all procs (sword/hoj/poisons/wf(horde)/crusader...) etc? And ye your simulation for HoJ proc is not true (1proc every 50hits) but we both know that.

For the fight duration, on a short fight crit might be better than hit because of AR, tea, BF, flurry juju, crusader uptime(#rng) etc... whereas hit might be better on a longer fight (and then it might not be true, hit could be better than crit on a shorter fight too, just because of procs which is a huge part of vanilla dps).

For example if a fight last 30sec, you'll be 50% of the time under CD, but if it last 120seconds you'll be 12,5% of the time under CD, and on a shorter fight your yellow damage should be higher (in percentage when comparing it to white hits) than the ones on a longer fight because of those CDs, and the RNG you will have plays a big roll there, but then is crit better than hit? considering a certain number of crit / hit, your gear, your spec, the actual procs that you can have, the duration of the fight etc?

I'm just saying it's incorrect because he doesn't take in account procs (and all it brings), which is a big part of our DPS, it just takes in account a number of hit and crit with a 0/0/0 spec, no particular piece of gear, no enchant, and no procs.

You're right about the duration of the fight affecting the relative weight of special / whites, indeed .
I kinda disagree about your statement regarding RNG & fight duration tho. Yes, RNG can play a bigger roll on shorter fights than on longer ones, but this isn't like one should think of it.

If you aim at modeling DPS accurately, you just gotta average RNG and assume everything happens as frequently as it should - otherwise let's just assume I crit 100% of the time with 20% crit or miss all my attacks, or whatever, you know.

By the way isn't my simulation for HoJ okay to you? Doesn't 2% mean, on average, 1 every 50?
Kinda does to me. I'm no maths expert tho, so any enlightenment on that matter is welcome.

Distracting Dagger is just an example, its best when you don’t have Mugger’s Belt[i], otherwise every other fast Dagger is nice to have with a high dps rate.

wrong again.

I kinda agree. From what I simulated, Distracting Dagger + Nightslayer > Mugger's Belt + Bonescrapper, then CHT + Muggers is better, and then ACLG / CHT / T1 rules.

If you have +0 Skill with your current gear, yes you should get Mugger's Belt (better than Distracting Dagger + T1 Belt since you have CHT), until you can get your hands on ACLG gloves, then CHT + T1 Belt + ACLG and you'll be set for a while

Thank you for the reply, i currently have 2 epic daggers, and i dont use distracting. My question is that with 12% Hit total, can i ignore the +5 from mugger belt and go with an agi belt? or should i stick to +5 mugger belt while i get the hands with +5 daggers from mc ?

As I said, since you have CHT in off hand, you should stick to +5 Mugger's Belt until you get Ancient Core Leather Gloves (ACLG) from MC, yes.

[enjoyr14]

You're right about the duration of the fight affecting the relative weight of special / whites, indeed .
I kinda disagree about your statement regarding RNG & fight duration tho. Yes, RNG can play a bigger roll on shorter fights than on longer ones, but this isn't like one should think of it.

If you aim at modeling DPS accurately, you just gotta average RNG and assume everything happens as frequently as it should - otherwise let's just assume I crit 100% of the time with 20% crit or miss all my attacks, or whatever, you know.

By the way isn't my simulation for HoJ okay to you? Doesn't 2% mean, on average, 1 every 50?
Kinda does to me. I'm no maths expert tho, so any enlightenment on that matter is welcome.

The thing with RNG in vanilla is that you should think about it, because you do have a lot of RNG in your dps, and he just completely ignores it in his maths that's all i'm saying, you can't state that crit>hit after a certain amount without taking in account every other source of dps, your not playing alone with a 0/0/0spec and no gear.

And ye your simulation for HoJ isn't okay because it's not maths it's randomization, just like /roll so you can't really define it, else you could say "I gain 2%crit so I'll have 2 more crits every 50hits" and it just doesn't work like that, you could have 5HoJ proc in 10hits just like you could have 0 in 50 hits (and I don't think the proc is THAT rare because if it was it wouldn't be as good as we all know it is), and by getting more hit you increase your chances to proc (you don't with more crit).

To calculate it precisely you sould take many hit samples (like ~15 of Xnumber of hit), roll for every hit and average every samples (for example one time with like 10% +hit on gear and another time with like 15% +hit), then see how it is affected by hit (and even then I don't really know how blizzard's RNG is randomized so it might just not be as simple as a /roll, but that's the closer you can get from it).

I know RNG is hard/long to calculate, it shall not stay in the void tho, because like I said, it's a big part of our dps.

[Youfie]

You're right about the duration of the fight affecting the relative weight of special / whites, indeed .
I kinda disagree about your statement regarding RNG & fight duration tho. Yes, RNG can play a bigger roll on shorter fights than on longer ones, but this isn't like one should think of it.

If you aim at modeling DPS accurately, you just gotta average RNG and assume everything happens as frequently as it should - otherwise let's just assume I crit 100% of the time with 20% crit or miss all my attacks, or whatever, you know.

By the way isn't my simulation for HoJ okay to you? Doesn't 2% mean, on average, 1 every 50?
Kinda does to me. I'm no maths expert tho, so any enlightenment on that matter is welcome.

The thing with RNG in vanilla is that you should think about it, because you do have a lot of RNG in your dps, and he just completely ignores it in his maths that's all i'm saying, you can't state that crit>hit after a certain amount without taking in account every other source of dps, your not playing alone with a 0/0/0spec and no gear.

And ye your simulation for HoJ isn't okay because it's not maths it's randomization, just like /roll so you can't really define it, else you could say "I gain 2%crit so I'll have 2 more crits every 50hits" and it just doesn't work like that, you could have 5HoJ proc in 10hits just like you could have 0 in 50 hits (and I don't think the proc is THAT rare because if it was it wouldn't be as good as we all know it is), and by getting more hit you increase your chances to proc (you don't with more crit).

To calculate it precisely you sould take many hit samples (like ~15 of Xnumber of hit), roll for every hit and average every samples (for example one time with like 10% +hit on gear and another time with like 15% +hit), then see how it is affected by hit (and even then I don't really know how blizzard's RNG is randomized so it might just not be as simple as a /roll, but that's the closer you can get from it).

I know RNG is hard/long to calculate, it shall not stay in the void tho, because like I said, it's a big part of our dps.

Man, I agree with you when you say statements made earlier about Crit > Hit, while they might still be true, ignore some procs that you listed (Swords, WF, Crusader, Poison, HoJ).
The question on this particular matter is now whether these procs are enough to counterweight the theoretical superiority of Crit assessed without taking them into account, but I agree with you.

However, dude, I strongly disagree about the way you see / understand RNG.

Modeling something random into its averaged effects does NOT mean you ignore it.

I'm perfectly aware you can get 10 procs in a row or never get one in your entire life, regardless of the proc chance being 0.01% or 99.99%, it's the way probabilities work, sure.

But in the long run, mathematically speaking, something with a X% to happen WILL happen X% of the time, it's just damn high-school-level statistics (Bernouilli's Law of large numbers) ffs.

It's not because something can happen twice in a row or once every 100 years that it can't be modeled properly.

Hand of Justice having a 2% chance to proc means it will proc once every 50 hits on average, period. I CAN proc 4 times in a row or never for a long period of time, but in the long run, in average, it will be 2%.

As the sample size grows, the observed probability will approach the theoretical one, end of story.

That's why I don't see anyting wrong with modeling HoJ's DPS [averaged (after glance, crits, dodges, misses & shit) MH damage] + [averaged Instant Poison damage (if you got some on your MH)] every 50 hits.

[enjoyr14]

The question on this particular matter is now whether these procs are enough to counterweight the theoretical superiority of Crit assessed without taking them into account.

That's all i've been saying (more or less).

Modeling something random into its averaged effects does NOT mean you ignore it.

He (#Oto) ignores it in his maths, I never said you did.

But in the long run, mathematically speaking, something with a X% to happen WILL happen X% of the time, it's just damn high-school-level statistics (Bernouilli's Law of large numbers) ffs.

ffs.

It's not because something can happen twice in a row or once every 100 years that it can't be modeled properly.

Hand of Justice having a 2% chance to proc means it will proc once every 50 hits on average, period. I CAN proc 4 times in a row or never for a long period of time, but in the long run, in average, it will be 2%.

That's why I don't see anyting wrong with modeling HoJ's DPS [averaged (after glance, crits, dodges, misses & shit) MH damage] + [averaged Instant Poison damage (if you got some on your MH)] every 50 hits.

Yes you can modelize it, but it still doesn't make it true because you have to take TIME in account, and I stated it in every post I've made.

As the sample size grows, the observed probability will approach the theoretical one, end of story.

Yes, as the sample size grows, we will approach the theoretical probability, thanks I went to highschool too.

But that's not the problem, the problem is how HIT affects this probability, on a SHORT (#totalrandom) and on a LONG (#randotheoritical) fight, and when you know, only then you can compare hit properly with crit (and not assume this is better than that when reaching X because only white and yellow hits are to be counted).
And btw, you have to take in account that some procs are equivalent to a hit and can proc other stuff(#nerfswordspec in 2.2.0) so the "sample size" will be the same, but the time until you get to the sample size will be different.

In his maths, he just ignores procs right away when it's actually a big deal, that's all I've been stating since my first post.

But in the long run

It's not always a long run (just look at the duration of MC's boss, it's around ~50seconds), I said it in my first post that duration of the fight should be taken in account, and I say it again.

Time is at the essence.