I spent a while looking through logs and found zero misses, so I was sure that bosses are not able to miss, and found similarly reduced dodge and parry chances relative to base avoidances and shuffle/EB uptimes in logs as well.

I appreciate you dredging that old post up to get an actual blue response on the subject.

This does mean that Night Elf is worth .5% avoidance, not 2%, which makes them quite garbage. Night Elf bonus is evidently dodge again, instead of increased miss chance, so it's still worth 2% avoidance and still far and away the best damage "reduction" race (quotes since it's avoidance). Hooray.

So basically if i understand your guide well we have to gem:

Red: 160 Agility ?
Yellow: 80 Agility&160Haste or 320Haste or 80 Agility&160Crit or 320Crit ?
Blu: 80 Agility&160hit ?
Meta: 216 Agi&3% crit bonus?

More along the lines of
Red: Exp/Haste
Yellow: Haste
Blue Hit/Haste

I personally use the stamina/armor meta, because 2% armor is quite big imo and 3% crit damage is next to useless for damage reduction

The 2% armor only adds about 300-400 armor depending on your gear. If you are choosing the stam gem over the agi gem you are much better to go with the 2% magic damage reduction. In long term mitigation though assuming you have enough stam to survive a few blows the agi gem will win out due to the dodge and crit granted.

Though I hear there is a new legendary meta in the new patch which will be the way to go anyway.

From the Brewmaster Basic Discussion thread:
Apparently the 1600% was a tooltip error, and we've only had to do 800% the entire time. They're testing 1600% on the PTR.

blue post:
"General
Attack Power gained from Vengeance is now capped at the player's maximum health."

First, the cap is your HP, not some fraction of it or something. It is certainly not common to be getting in excess of 500k AP from vengeance, so this is going to have no effect on the normal course of play. I imagine this was just a fix to prevent things like taking some ridiculous hit that you're completely supposed to die from with Diffuse Magic and Zen med up, then turning around and destroying the boss. A monk with fort brew, zen med, diffuse magic (or even dampen harm for that matter), guard, pws, life cocoon, maybe some spirit shell could live through a hit for many millions of damage, which would result in completely imbalanced levels of attack power for a while. Further, you could leave Guard out of that list, get hit for 500 million or whatever by some aoe you're supposed to dodge, then guard and be invincible for 30 seconds.

Secondly, because the amount of damage you have to do in order to get a statue bubble scales with the AP you have, vengeance doesn't have nearly as big of an effect on your statue bubbling as it does on your damage. What has a far greater effect is how many targets you're hitting. On Wind Lord I (and i'm sure most monks) are doing as much or more healing than most healers are capable of, simply by aoeing all of the adds. While it's personally cool to be tanking, do a million dps, and top the healing meters, i can't say it's really very balanced.

They nerfed ramp-up times to cope with vengeance generated by single hits. I guess it's recent 80 lvl pally thing - due to Vengeance calculation bug he was getting to few millions worth vengeance and absorbing all the damage with Sacred Shield. So, there you go - a cap. Instead of a bugfix : )

The Agility meta gives 0.24% dodge before diminishing returns and 0.29% crit from Agility (including +5% stats buff) and about 1% more self-heal(if it's not overheal) and statue absorbs from the increased critical effect (note that it's not critical strike chance), depending on your crit chance.
The Armor meta gives 0.5-0.6% physical damage reduction depending on your armor and about 60k hp which does nothing for your damage reduction.
So I don't think it's such a foregone conclusion as you make it sound.

The mentioned 2% Magical Damage reduction meta is also interesting, but in the current content only Lei Shi (and maybe Elegon if your raid is a bit undergeared) deal dangerous magical damage imo, so I don't think it's such a good coice at the moment but that can very well change with Throne of Thunder.

P.S.: The numbers I used were 951.2 Agility/1% dodge and 780Agility/1%crit.

P.P.S.: Protectors might also favor the magical meta, but I'm tanking Kaolan for Elite mode and DPSing for non-elite, so I'm not really sure.

edit: I used the rogue values for agi/crit by accident

Those numbers seem sound at first glance. The thing people don't keep in mind when comparing the metas is that the Armor/Stamina one does nothing for dps, while the Agi/Crit Damage meta is clearly designed for dps. It has real tanking value as well and is probably comparable, although it does depend upon the scenario. In my experience, tank damage matters a lot in progression raiding, and the gain in damage is often worth taking an even/slightly inferior gem for the total benefit provided.

Here is my basic take on Monk secondary stat priority:
Stamina and Mastery prioritization is highly suboptimal for damage reduction. It just is, mathematically. It provides healers with much more room to accomodate for burst damage, but will result in an increase in total damage taken.
Haste and Crit prioritization is much better for damage reduction over the course of a fight, but has a lower margin for error.
A wrench gets thrown in the works when a Monk is terribly undergeared for a fight featuring spike damage, and in this case the best way to handle the situation is to deviate from the mathematically superior total damage reduction priority in favor of the burst damage reduction priority. This is also very strongly a function of player skill. A Monk-playing Robot with instant response times could probably have tanked a guild's first Heroic Gara'jal on a Brewmaster Monk using a standard Haste build without difficulty, but most players were going the Mastery/Stamina route and still found it pretty painful.

I think this pertains to the meta quandary in much the same way. The Agi/Crit Damage meta is just all-around superior as a gem. However, there are situations when the Monk needs a Stamina/Armor (or magic damage reduction) meta instead, despite its being suboptimal, just to live through punishing burst damage.

Also, the value of an armor increase is diminished as a damage reduction statistic by avoidance. It doesn't become less useful, mind you, but since you are not getting any benefit from armor when you avoid an attack, it is only useful on unavoided attacks. This is pretty obvious, but my point is that a .6% increase to damage reduction from armor is not a .6% damage reduction increase. It is a .6% * (% chance to be hit) damage reduction increase. So, going with a low estimate, let's say a Monk's avoidance chance against a boss is 34% (28% parry from base parry, Swift Reflexes and Shuffle, 15% dodge from gear and buffs, -4.5% parry and dodge from being a level 93 boss). This means chance to be hit is 66%, and the overall damage reduction would only be .396%.

Armor is good, really good, but it does not compare to avoidance in terms of total damage reduction. The historical issue with avoidance has always been the unreliability. Elusive Brew is a significant breakthrough in terms of player control, and is one of the most impressively designed tools Blizzard has ever given a tank class.

Unless this game suddenly becomes horribly designed at some point in the future, I will always recommend putting your gearing emphasis in favor of active mitigation rather than passive survivability, because it provides more control and puts more responsibility upon the individual to keep themselves alive. Yes, this is harder, but it is the right way, and that is what matters. Putting the pressure upon yourself to stay alive, instead of your healer, is the right way to tank.

Of course, gear however you feel comfortable, but I do not see a mathematical justification for gearing mastery/stamina until you have found a fight in which your inability to live through spike damage is causing wipes, and there is no gameplay alteration (e.g., changing cooldown usage) that can fix the problem.

I really agree with Pisshands on this. Outside of a scenario where you are doing content that is frankly designed for players in much better gear (aka many week1/week2 HM kills), Elusive Brew, Guard, and to an extent other CDs are the direct counter to the last 8 years of saying "well avoidance would be good, but I can't control it so I die when i'm unlucky." This becomes even more true in any fight with tank swaps where you essentially double your possible EB awesomeness. I die for 3 reasons:
1. I screwed up using CDs, usually because I'm also trying to play about 15 other characters like it's a RTS game and I have to tell them where to stand and what to attack every couple seconds or they are useless/dead. I'm definitely not good enough to do that to the extent I always used to on my druid when it barely mattered what buttons I hit other than defensive CDs.
2. I go 10 seconds without getting any actual (non hot/aoe/beacon) heals more often than I have cooldowns to cover. Usually because my healers are dead or someone else is standing in a fire because I didn't do a good enough job of playing the aforementioned wow 25 man raid leading RTS.
3. I get agro when I'm not supposed to.

After a few pulls on a new boss of me being a pretty much terrible tank because I just sort of mash buttons semi randomly and watch everyone else, the first 2 are almost a non issue. The 3rd is a bit more trouble, but I can usually get blessing of salvationed to victory without having to gimp up my resource generation. I think i'm going to experiment with dropping out to tiger at the loss of a lot of shuffle stacking.

I chose Gara'jal because the only other fight that hits that hard and was pretty thoroughly experienced by the body of Monk players that came to mind was Will of the Emperor, and the issue with that fight for most tanks is failure to dodge. If you don't perfectly dodge on Will, your slow feet are a bigger issue than gearing. Others may have experienced issues on that fight even while perfectly dodging, but I did not at all. Elusive Brew, DHarm and Guard trivialize the fight pretty effectively - on our first kill I had 66% avoidance on melee attacks. Insanity.

In the heat of the moment, overall damage reduction is less important than living through that burst damage. There are two ways to use Guard and EB: maximum damage reduction or as a lifeline in emergencies. I promote the emergency usage route on cooldowns and I gear for overall damage reduction. I think this provides a good balance.

However, I do think there is a limit to overall damage reduction gearing through Haste, and I think that gear values in t15 and beyond will most likely result in a shift from Haste gearing toward either Crit or Mastery. This is not to say that Haste will no longer possess value, but after reaching 9k+ Haste with Ascension, the amount of energy regeneration begins to become unwieldy. The 25% buff to Mastery will certainly make it a more attractive statistic, and Crit is already a rather valuable damage reduction statistic at present, so I think we could see a point at which transitioning from Haste to Crit or Mastery may be worthwhile. To be clear, this is all just conjecture at this point.