Heroic Boss Kill Order by Percentage

[bals]

Bals, you're very knowledgable, and I usually trust your advice. But please explain why. Are you suggesting there is a better metric than wipe rate?

Also, updated the original post with a more thorough explanation.

this is one reason:

A lot of heroic attempts on amber shaper get flagged as normal by world of logs. Likewise Garalon. Your statistics are based on flawed data, in addition to the myriad other flaws with this approach.

one night on our first few weeks killing ambershaper 9/11 attempts were flagged as normal when they should have been heroic.
our first week of attempts we had 41/44 attempts flagged as normal.
Garalon has similar issues - I assume it only flags it as heroic if you break his carapace. A lot of guilds probably have wiped before then due to bad kiting and other issues.

[Deleted]

Sha HC wipes are also marked normal if you don't make it to P2 although they are on heroic difficulty.

[Mathrys]

Garalon has similar issues - I assume it only flags it as heroic if you break his carapace. A lot of guilds probably have wiped before then due to bad kiting and other issues.

And once you break his carapace its pretty much a kill which is why the data is so skewed..

[Deleted]

To get an accurate list you have to remove...

... the buggy WoL encounters Garalon, Ambershaper, Sha of Fear
... Tsulong because many guilds killed him using the shaman bug
... Protectors for the stated reasons
... Spirit Kings since they were once known as Reset Kings --> 200 wipes although you effectively did just 50 real pulls.

Once you do this you get a list, many people would actually agree with. So, the only mistake you made is that your data is flawed

Edit:

Feng - 25%
Stone Guard- 22%
Garajal - 22%
Elegon - 16%
Wind Lord MelJarak - 13%
Blade Lord Tayak - 11%
Will of the Emperor - 8%
Imperial Vizier Zorlok - 6%
Grand Empress ShekZeer - 5%
Lei Shi - 4.6%
Protectors Normal - 4.0%

Looks okay to me (being 16/16H) although it strikes me a bit that Windlord should be easier than Bladelord. However, I do know guilds that killed Mel'jarak before Tayak.

[Reckles]

Awesome, guys, thanks! I appreciate that info a lot more than "wtf, this list suxorz." But yes, 41/44 attempts is huge and data that buggy makes the list inaccurate, especially the Garalon stuff. I wonder why WoL does that. Anyways, I'll adjust the front page with this info. If there's anything else, let me know. Do y'all agree on the metric though, if the data gets fixed?

[visize]

*Edit: deleting my snarky comment...ok, so why do you think % of guild's that complete a boss is more accurate than kill ratio, since many guild's haven't attempted HoF and Terrace?

It was my interpretation that this thread 'Heroic Boss kill order by percentage' was meant to provide a comprehensive list of the most linear kill progression. Kill ratio is more indicative of how many guilds have the boss on farm; #ofkills / #of attempts.

[KaPe]

Don't forget that people like to hide their wipes and only make the actual kills public.

[Deleted]

load of rubbish even in respect to percentage of wipes

[Durahal]

I don't get why you are trying a very complicated approach to something that is very easily figured out just by looking at the total kills in WoL/WoWProgress.

[Asmodejjj]

Awesome, guys, thanks! I appreciate that info a lot more than "wtf, this list suxorz." But yes, 41/44 attempts is huge and data that buggy makes the list inaccurate, especially the Garalon stuff. I wonder why WoL does that. Anyways, I'll adjust the front page with this info. If there's anything else, let me know. Do y'all agree on the metric though, if the data gets fixed?

The thing we're trying to ram through your skull is that your list is worthless. There's already lists out there of easiest to hardest bosses, in fact, they've been there for months now. Your data is wrong in several ways, as others have mentioned. There is literally no reason for this thread, or for the arguments/insults going on in it. Just accept that this thread is a mistake, and have it closed.

[Isotope]

I think more accurate is (from easiest):

1. Stone Guard
2. Feng the Accursed
3. Gara'jal the Spiritbinder
4. Elegon
5. Blade Lord Ta'yak
6. Four Kings
7. Wind Lord Mel'jarak
8. Garalon
9. Will of the Emperor
10. Imperial Vizier Zor'lok
11. Lei Shi
12. Amber-Shaper Un'sok
13. Protectors of the Endless
14. Grand Empress Shek'zeer
15. Tsulong
16. Sha of Fear

Based on number of guilds that manage to kill these.

I 90% agree with this list. The few things I disagree with are Spirit Kings at 6, then Will down at 9. If you can get spirit kings down you can kill Will. 6-9 can pretty much go either way. The same for Vizier and Lei Shi, pick one between the two and just focus on that.



@OP

Please at least see more than 70% of the heroic content before making a "GO HERE NEXT THREAD." I laugh at guilds on my realm trying to do heroic protectors when they haven't even killed Vizier let alone Heroic Will at least.

[Reckles]

I don't get why you are trying a very complicated approach to something that is very easily figured out just by looking at the total kills in WoL/WoWProgress.

Durahal, if you all don't agree with it, that's fine, but ease="kill:wipe ratio" adjusted for gear level. I do thank everyone who messaged me and helped determine reasons for why the data is wrong. Data aside, the reason for kill ratio being used over kill amount is that amount is hugely skewed by number of attempts. What people care about when attempting a fight is how long it will take them to kill it. Sure anyone could move straight to sha of fear H after killing normal mode, but it would take them 200+ wipes before a kill, so getting Feng down after 6 wipes would mean Feng is easier. For another example, look at normal will of the emperor vs normal elegon. A lot of people get stuck at normal elegon and can't even attempt will, so the WoWprogress shows normal will under elegon. There are many ways to interpret ease, but # of kills is hugely unreliable statistically. It is a list showing worldwide progression, not boss ease. Dividing amount of kills by number of attempts though, gives you my list here and something useful. I think people see the buggy Garalon and Unsok up top and call bullshit, as they should. Doesn't mean that kill amount is any better though. I am looking for more useful data though, since WoL wasn't nearly as useful as I imagined it would be.

Asmodejjj, thanks for trying to help but for the above reasons, I'm trying to bring light to the fact that the progression list other people rely on is inaccurate. While mine is wrong (but getting better, I've tried to update it), it doesn't mean theirs is correct, just that I need better data. So I request from the moderators to keep this open until we, hopefully as a group, find the most accurate way to determine the mean kill:wipe ratio for each boss.

To answer Durahals question though, we raid two nights a week, have 16 heroic bosses to kill before 5.2 hits and have one tank still wearing blues, so efficiency is very important. Plus, I can do this at work, lol

[Deleted]

The validity of these percentages depends greatly on how the heroic detection works for each of these encounters on world of logs.

Some is more of an educated guess as to how its done. Some is from the World of Logs forums

Stone Guard: Detection is based on Floor tiles and the buff that you get when you take the crystals. This happens relatively early in the fight, so the only wipes on this fight that are missed in the % will be the ones that are right at the very start, which is almost impossible. 22% is probably reasonable

Feng: Detection based on the shadow phase and the strength of spirit buff, which (unless you take shield first) you need to at least push past 1 phase, there will be plenty of wipes not shown here by people that wipe to the arcane, which is often done first. 25% is probably a bit generous by a few percent.

Gara'jal: Detection based on Frail Soul which you get when you come out of the spirit realm, so any wipes before the first guy comes up from the totem aren't included, most wipes don't really occur this early in the fight, but it will throw the percent off by a couple again.

Stone Guard: as has been mentioned it used to be "reset kings" however detection is based on the shields which are new in heroic so if you wiped straight away, it'll be a normal mode wipe. So, luckily, this probably reflects the difficulty reasonably well. It may miss a few wipes that will occur due to annihilate at the start or to the first flanking, but otherwise this percentage should be fine.

Elegon: Heroic detection (I'm guessing) is based on "Destabilizing Energies" the debuff that the soaking players get from the add, this occurs relatively early, so again percentages are relatively ok.

Will of the Emperor: heroic detection based on titan sparks, and/or titan gas being present, so this is pretty much accurate.

Visier: Detection based on the different attenuation spell, and perhaps also echoes spawning. Numbers perhaps a bit skewed, I think any "goes left" attempts probably counted as normal if you wiped straight away. Of course this also doesn't count people pulling with no intention other than to practice the attenuation, which depending on your viewpoint may/may not count as a wipe, but it certainly would be recognized as a heroic attempt, even if you pulled with 5 people.

Blade Lord: Tempest, so wiping on the first strike would be counted as normal.

Garalon: isn't counted as a heroic attempt until phase 2 so this number is massively off

Wind lord: isn't heroic until you get a recklessness, so wipes before then won't count (again off a bit).

Amber Shaper: isn't counted as heroic until amber globules spawn, so like....almost all heroic wipes won't be counted...gg

Empress: isn't heroic until..you get a corrupted dissonance field? I think, so some wipes missed here too.

Protectors Normal: detected as heroic due to the adds so most wipes will be counted
Protectors Elite: detected as heroic due to order, so won't count until you get to the last phase at least, else you could have been planning to kill kaolan second or something.

Tsulong: massively thrown off by the bugged shaman parses, but otherwise hc detection done with dark of the night, so mostly numbers are ok here

lei shi: scary fog...which you get straight away

Sha of Fear: 2nd phase only, so any wipes in phase 1 are ignored.

As you can see, there are plenty of discrepancies in the fight detection of a wipe, so this already throws the numbers of, and this is before you get to things like, Raid x has been out longer and therefore is on farm for more people who will slowly increase the percentage, or fights that you reset, or wipe straight away (providing detection works immediately at the start of the fight).

If I had to Estimate, the percentages would have to be: (keep in mind that this may still not be an accurate representation of their difficulty due to the previous statement)
Stone Guard- 22%
Feng - 21.5%
Garajal - 21%
Elegon - 15%
Wind Lord MelJarak - 13%
Blade Lord Tayak - 11%
Spirit Kings - 10%
Garalon - 9%
Will of the Emperor - 8%
Imperial Vizier Zorlok - 5%
Lei Shi - 4.6%
Protectors Normal - 4.5%
Grand Empress Shek'Zeer - 4%
Tsulong - 3%
Amber Shaper Unsok - 2.8%
Protectors Hardmode - 2.1% (<- this really is a super guess, I think that perhaps this may even require less wipes than normal if you have already done normal and have gear from all the other bosses, for ex. some guilds took like 5 pulls on this?)
Sha of Fear - 1.5%

[Durahal]

Sure anyone could move straight to sha of fear H after killing normal mode, but it would take them 200+ wipes before a kill, so getting Feng down after 6 wipes would mean Feng is easier. For another example, look at normal will of the emperor vs normal elegon. A lot of people get stuck at normal elegon and can't even attempt will, so the WoWprogress shows normal will under elegon. There are many ways to interpret ease, but # of kills is hugely unreliable statistically. It is a list showing worldwide progression, not boss ease. Dividing amount of kills by number of attempts though, gives you my list here and something useful. I think people see the buggy Garalon and Unsok up top and call bullshit, as they should. Doesn't mean that kill amount is any better though. I am looking for more useful data though, since WoL wasn't nearly as useful as I imagined it would be.

To answer Durahals question though, we raid two nights a week, have 16 heroic bosses to kill before 5.2 hits and have one tank still wearing blues, so efficiency is very important. Plus, I can do this at work, lol

I hope you realize that you can't go straight to H Sha without killing the other bosses on heroic. Even then, looking at just the number of boss kills by guild progression doesn't suggest this, so this argument doesn't hold any validity.

As many people have pointed out to you in previous posts, the data you would obtain by doing this would be much less skewed than the method you have tried to do, just look at the first list you generated...

And I also raid on a strict 2 night a week schedule, so I understand the importance of making the most of what little raid time some guilds have. But another thing you have to keep in mind is that some bosses will be easier for some guilds than others based on raid compositions and individual talent (ie. having very strong healers as opposed to very strong dps).

But I agree that a list would be helpful, I'm just saying there is a much easier way, and it kind of already exist on WoWProgress.