Jab Nerfed Again :D

[Totaltotemic]

Apparently the new formula is 15% + 15% per tick, resetting when we get chi.

The instant tick from Soothing Mist now has an 85% chance to not have generated chi.
2nd tick - 59.5% chance to not have had chi yet
3rd tick - 32.7% chance to not have had chi yet
4th tick - 13.09% chance to not have had chi yet

Contrast that with a static 30%:

Instant - 70%
2nd tick - 49%
3rd tick - 34.3%
4th tick - 24.01%


They think that this is supposed to help. They've actually nerfed the generation from the first and second ticks. Sure, it's more consistent now... more consistently on the 3rd or 4th tick rather than the 1st or 2nd.

[Dreyo]

Are you sure the first tick doesn't have "one" as the 15% per tick, being 30%? Can anyone test this? Clip the cast before it can tick again, and measure the proc rate? Er...maybe that won't even work, does it care about prior ticks even if they weren't from the same channel?

We need more detail.

[Astraios]

I don't think it's possible to take logs on the PTR. I gave the general breakdown of my healing, but it's basically ReM/uplift on twin consorts, ReM/uplift/Chi Torpedo on lei shen. The other 25% is mastery/sck/expel/SoM/etc.
ReM being clearly > Uplift on both fights, in terms of raw healing.
If anyone can tell me how to save logs from PTR, I can save them next testing.

[Totaltotemic]

Are you sure the first tick doesn't have "one" as the 15% per tick, being 30%?

I don't think so. That would result in a massive increase in chi generation from Soothing, but their goal is to just make it more consistent, which it is.

[Deleted]

So you didn't use any T30 talents at all?

[Dreyo]

I don't think so. That would result in a massive increase in chi generation from Soothing, but their goal is to just make it more consistent, which it is.

GC also said they wanted to increase the proc rate overall, didn't he? Am I making that up?

Edit: Yep.

Soothing Mists now generates chi more consistently and average generation rate is slightly higher.

And based on the measured results posted earlier, the average chi per channel has gone up.

[Aure]

It should be 40% and then 20% more proc chance for every tick you don't get chi.

Maybe 10% instead of 20%, I dunno, the proc rate needs to be higher than 15% + 15% for every tick you don't get chi..

Maybe i got something wrong, but if the initial tick had more chance to generate chi, we would recast SM always after the first tick. And that would be pretty stupid for such a long channeling


I love the fact that GC dedicated his post almost entirely to mistweavers monks but refused to comment all the requests asking for new chi dumpers, 40yd eminence, shitty lvl30 talents and so on...

[Deleted]

I don't think so. That would result in a massive increase in chi generation from Soothing, but their goal is to just make it more consistent, which it is.

15% + 15% = 30%, similar to the proc chance of static 30%. The formula states 15% + 15% per tick, thus I also think that the first heal counts as first tick.

[Dreyo]

Also, even if it was 30% on the first, 45% on the second, etc. it wouldn't increase the average generation rate that much. Each of those higher-chance events are predicated on other, low-chance events to happen first. It's a conditional probability problem. The end proc rate will be a little bit higher than 30%, but not very much higher than 30%. Based on the posted data, there's an average of 3.18 Chi per 9 ticks, for an aggregate average proc rate of 35%. A straight 30% proc chance would be 2.7 Chi per 9 ticks, for comparison.

It'd be nice to know if the proc rate only cares about ticks on your current channel, however. That may be something that only GC can answer - pulling it out of measured data would be rather annoying.

[Totaltotemic]

GC also said they wanted to increase the proc rate overall, didn't he? Am I making that up?

Edit: Yep.

Soothing Mists now generates chi more consistently and average generation rate is slightly higher.

And based on the measured results posted earlier, the average chi per channel has gone up.

No, that's correct, but let me get into this more. It does actually increase the total generation with the 15% + 15% by severely decreasing the possibilities of not getting chi for 5+ ticks.

The instant tick from Soothing Mist now has an 85% chance to not have generated chi.
2nd tick - 59.5% chance to not have had chi yet
3rd tick - 32.7% chance to not have had chi yet
4th tick - 13.09% chance to not have had chi yet
5th tick - 3.27%
6th tick - 0.327%
7th tick - 0%

Contrast that with a static 30%:

Instant - 70%
2nd tick - 49%
3rd tick - 34.3%
4th tick - 24.01%
5th tick - 16.8%
6th tick - 11.76%
7th tick - 8.24%
8th tick - 5.76%
9th tick - 4.04%
10th tick - 2.82%

And it keeps going on and on forever. Lopping that entire possibility past 7 ticks off and cutting the standard deviation considerably does in fact make it more consistent and have a slightly higher mathematical chance for chi. Just because I'm on a roll here, let's look at how ridiculous it would be with a 30% base + 15% per tick:

Instant - 70%
2nd - 38.5%
3rd - 15.4%
4th - 3.85%
5th - 0.385%
6th - 0%

That's an obvious massive increase in chi generation. It would put Soothing Mist at less mana per chi than Expel Harm, which would just be ridiculous and would cut out all semblance of mana spending for us.

[Dreyo]

Your analysis is flawed, but in a subtle way. You have the pieces there, but you're fitting them together wrong. You note, accurately, that your chance on not having generated a chi up to and including the 5th tick is very small - .385%. But here's the thing: in order for that to happen, you had four ticks previous to the fifth where you now have not generated chi. The conditional probability for that is (not coincidentally) the number for the 4th tick, 3.85%. Which means that these allegedly high-likelihood chi generation events themselves don't happen very frequently. With the cap of 100%, we can actually model exactly what the long-term average proc rate would be. But just looking at the numbers that you gave, you can't just say that it's "ridiculous" - because you need to tie it in with the conditional probabilities.

Based on the measured data posted earlier in the thread, they undoubtedly changed something between Live and when the poster took the measurements. And that resulted in the proc rate going up from 30% to about 35.3%, on average. I'm working on the full probability spread now - it being a finite set of probabilities makes it much simpler. Will post an update.

[DirewolfX]

For a full channel:
1 chi - 3.18%
2 chi - 34.13%
3 chi - 42.47%
4 chi - 16.91%
5 chi - 3.04%
6 chi - 0.28%
7 chi - 0.01%
*8 and 9 chi are possible, but the chance is miniscule

[Astraios]

So you didn't use any T30 talents at all?

My bad, I used Zen Sphere for the extra HoT/reliable extra stack healing.

However, I'm not used to placing it, and on both fights it would average 2-3 million healing when my ReM/uplift are averaging 20+ mil. It's part of the bottom %'s. As far as I'm concerned that tier is now minimally powerful in any extra healing after the nerd. (I suppose chi burst could be good for lei shen, but proper Zen Sphere would beat it if you timed 2 to explode whenever thunderstruck hits, assuming the raid even gets that low. Something to look into, CB might be better, but I like having something extra to cast, even if it's pretty insignificant.) Granted, Chi wave used to be overpowered when it averaged 400k healing/cast for free, but I think its something we need mechanically (a strong smart heal or an AoE smart heal, our lack of this is currently being remedied by...dwarfing the other healers HPS with ReM/uplift). I would gladly take that loss in HPS 400k/15 ~ 23k hps off of ReM and into chi wave as a smart heal.

I think the disc DA absorb nerf is showing up on meters, and my ReM is eating a lot of that extra healing that disc was preventing previously. That is my guess as to why I'm showing up as 100k HPS easily. This tier has, in the fights I've been in at least, a lot of burst damage mechanics/movement mechanics/blanket healing, and nearly every one of them seems to line up perfectly with our CDs (TFT/CB/Revival). This is making us extremely strong, combined with 2pc, enough to offset whatever healing loss jab caused.

It was really funny because I went in expecting to suck and feel like a wasted raid spot (except we had 23 for some of the testing, so any extra healing/dps from another player is helpful haha)...and then proceeded to crush meters while feeling like I had no idea what I was doing [getting used to SoM/SCK/less chi/less fistweaving/no chi wave, boss mechanics, etc] and I have plenty of room to optimize (though as time goes on other classes [esp. disc] tend to gain as much as we do, but theirs tend to push us off the meters and not the other way around.)

[Dreyo]

I think the disc DA absorb nerf is showing up on meters, and my ReM is eating a lot of that extra healing that disc was preventing previously.

Okay, now THIS sticks out. Monks are blanket throughput healers, in the style that most of us play them as, and I think it's entirely possible that a lot of our healing is going to overheal due to disc priests that now will be registered as actual healing. Reduce the blanket preventers, and the blanket healers will get a lot of work for free.

(now, whether those numbers on the meter are actually useful is another story)

[Totaltotemic]

Your analysis is flawed, but in a subtle way. You have the pieces there, but you're fitting them together wrong. You note, accurately, that your chance on not having generated a chi up to and including the 5th tick is very small - .385%. But here's the thing: in order for that to happen, you had four ticks previous to the fifth where you now have not generated chi. The conditional probability for that is (not coincidentally) the number for the 4th tick, 3.85%. Which means that these allegedly high-likelihood chi generation events themselves don't happen very frequently. With the cap of 100%, we can actually model exactly what the long-term average proc rate would be. But just looking at the numbers that you gave, you can't just say that it's "ridiculous" - because you need to tie it in with the conditional probabilities.

I am capable of estimation, this isn't the 3rd grade where I need to show my work to use common sense. I don't have to do a statistical probability spread analysis to see clearly that if it starts at 30%, the total generation chance is much more than "slightly increased", in fact you'll find that this specific case would end up being more than a 20% increase (30% to 36%), which I would classify as "ridiculous" considering the very brief test run at 35% was considered too powerful. If it does start at 30%, then the increase is way beyond "slight."

[Dreyo]

I am capable of estimation, this isn't the 3rd grade where I need to show my work to use common sense. I don't have to do a statistical probability spread analysis to see clearly that if it starts at 30%, the total generation chance is much more than "slightly increased", in fact you'll find that this specific case would end up being more than a 20% increase (30% to 36%), which I would classify as "ridiculous" considering the very brief test run at 35% was considered too powerful. If it does start at 30%, then the increase is way beyond "slight."

Except this is the measured data:

This is what I got on the day of the Soothing Mists change:

Out of about 1000 full casts of Soothing Mists, with 9 ticks per Soothing Mists I got the following:

1 chi: 1%
2 chi: 23%
3 chi: 42%
4 chi: 25%
5 chi: 9%

So the chi proc rate is roughly 35%, however it is considerably rare to only get 1 chi, but you cannot get more than 5 chi.

During tonight's testing on the Twin Consorts I observed a couple times that a full channel of soothing mists only generated 1 chi, which is pretty frustrating when you want to generate chi before Cosmic Barrage hits without spending 48k (if you could even find a safe spot in melee as a mw) mana through jab. So generating only 1 chi still happens, but rarely.

Oh, and sorry about any implications that you don't know how to estimate, but people tend to be really, really awful at conditional probability unless they have training. It's just a default to assume people are likely to get it wrong, which causes trouble sometimes. This is one of those times.

[Totaltotemic]

Except this is the measured data:

Which is flawed. Can't get more than 5 chi per channel? Wrong, you can get up to 9. I have a feeling that he tested that with Ascension and forgot to spend the chi when he capped, so he could never see gains beyond 5 chi per cast. It's more than 35%, probably closer to 37%, which is definitely more than "slightly increased."

[Dreyo]

Who gets to say how much slight is? Going from 30 to 35 seems "slight" to me. Even 37. Particularly when people have been asking for things like bringing it up to 50% or higher.

[Totaltotemic]

Who gets to say how much slight is? Going from 30 to 35 seems "slight" to me. Even 37. Particularly when people have been asking for things like bringing it up to 50% or higher.

That's only because people get easily confused between percentage point increases and percentage increases. 30% to 36% is a 6 percentage point increase, but it's a 20 percent increase. The reason people have been asking for massive increases is because we need a massive increase to fill the void that the removal (let's not kid ourselves, SCK is more mana per chi then Jab) of Jab from normal healing in order to save lives.

gnorrior has been talking about Renewing Mist's huge potential for healing in 5.2. Relying on that and Chi Torpedo essentially leaves us at WotLK Resto Druid level, can't save anyone's life but acts as a +Hp5 aura to the entire raid. On fights with big bursts of healing needed, Renewing Mist won't save anyone. On fights with big bursts and people being spread out, SCK and Chi Torpedo won't save anyone. We need to be able to use Uplift, and we can't use Uplift when we're relying on RNG to fuel a multi-target heal with RNG targets. The chance of being able to land it when you need it on people that need it more than 2-3 times in a row (barring Chi Brew, as raid-wide damage usually happens much more often than every 90 seconds) is extremely small. A huge increase to Soothing Mist is nice and all, but it's still RNG. Guaranteed after 7 whole ticks, yes, but still RNG in the 2-3 second window when real healing matters.

[Dreyo]

20% increase in Soothing chi gen doesn't seem unreasonable to me, is what I'm saying. If they're really asking us to use Soothing to power Uplift instead of Jab, then it needs to be a good bit higher than live. Going from 30% to 32% isn't going to cut it.

And yes, it is RNG, but this new implementation should significantly reduce the "drought" streaks that are what people actually dislike about RNG. Nobody complains about RNG when you get four chi in four ticks.