It would also remove many legitimate boss kills though. For example, this week we had our first kills of Naj'entus-Akama and Anetheron-Kaz'rogal. Out of those kills we DE'd a piece of loot off of Naj'entus and Shade of Akama, leaving us at 1 piece of loot per boss. So requiring multiple pieces of loot from a boss would take half of our progress this week out of the ranking.
I'm wondering if there's any way, once all of the guilds are archived, to find new members and remove the gear they're currently wearing from the guild's loot pool. That would certainly solve the problems with transfers.
My second thought: Blizzard really cares a lot about raiders!
My rough estimate says there's 75 US guilds that can even zone into Black Temple. There's maybe 200 guilds that can engage Kael'thas. Even estimating a roster of 40 raiders per guild, and a similar number of guilds with this much progress in Europe, that adds up to 6000 people in Black Temple and 16000 people who have experienced all of the first two raid instances in TBC. And WoW's subscriber base is what, 8 million active accounts? Conservatively 6 million people? Even if your guild can just kill Gruul and Magtheradon, you're doing more than 99% of the WoW player base is.
It's unbelievable the amount of resources Blizzard has devoted to gameplay that only a small subset of players are experiencing, and I'm very thankful they've gone to all this trouble for my sake. I know it's cool to bash Blizzard for being slow to fix issues in raiding. But in the grand scheme of things, we're not remotely close to their typical customer, and they still deliver us a high quality gaming experience. I really can't complain.
]I'm curious to know what combination method you use for score estimation. I noticed a few times that there are "big" (about 30 points, which is about 10% for the bottom half of the Top 100) jumps in scores between guilds that have only one boss difference between them. I don't think this disrupts the general trends that much, but it would nonetheless be interesting to me (and I bet others here) if you were able and/or willing to discuss the mathematics associated with this separation.
The scoring system is really simple, and admittedly I'm no math whiz. Basically what I have is...
For each boss I associate two values: KillValue and DropValue. KillValue is a largish number while DropValue is a very small number. To calculate the score I get the corresponding values from each item that has dropped for a guild and then for the first item from a given boss I add in KillValue to the score, then for each additional item I add in DropValue with a decaying factor. So the points from a given boss are determined predominately by the KillValue with the DropValue only slightly adjusting the score to differentiate guilds that have killed the same bosses.
Here is a query out of my mob database for Karazhan and Gruul's Lair values, for some concrete numbers:
+------------------------+-----------+-----------+
| BossName | KillValue | DropValue |
+------------------------+-----------+-----------+
| Rokad the Ravager | 1 | 0.001 |
| Hyakiss the Lurker | 1 | 0.001 |
| Shadikith the Glider | 1 | 0.001 |
| Attumen the Huntsman | 2 | 0.002 |
| Moroes | 3 | 0.003 |
| Maiden of Virtue | 5 | 0.005 |
| Romulo & Julianne | 4 | 0.004 |
| The Big Bad Wolf | 4 | 0.004 |
| The Crone | 4 | 0.004 |
| The Curator | 5 | 0.005 |
| Terestian Illhoof | 6 | 0.006 |
| Shade of Aran | 6 | 0.006 |
| Netherspite | 6 | 0.006 |
| Chess Event | 4 | 0.004 |
| Prince Malchezaar | 8 | 0.008 |
| Nightbane | 8 | 0.008 |
| High King Maulgar | 10 | 0.010 |
| Gruul the Dragonkiller | 12 | 0.012 |
+------------------------+-----------+-----------+
Originally Posted by Sarutobi
Also, you say you're pulling the guilds from warcraftrealms. Would it be safe to assume then that you're selecting the guilds based on the number of active members?
I was just using warcraftrealms as a place to get guild names, so I was pulling whatever they provided me, which also just so happened to be (I believe) the "largest" 50 guilds from each faction on each realm. I'll have a manual guild submission thing up at some point so people can add guilds that aren't in the list. For now they can contact me at feedback@wowjutsu.com with exact Realm and Guild names and I'll add them manually.
If the system records guild roster history, then you'd only need to discount loot from people who only showed up recently. If someone has been in a guild for six months, and then shows a piece of gear from a boss in logical progression from what the guild has already killed, then there's no reason to discount that loot.
But another wrinkle is when a guild has someone running with an out-of-guild raid. We have a few OOG folks who run in our raids with us, and their guild either doesn't raid or is not as progressed, when the OOG folks get loot, it will be credited to their progression, not ours.
It would also remove many legitimate boss kills though. For example, this week we had our first kills of Naj'entus-Akama and Anetheron-Kaz'rogal. Out of those kills we DE'd a piece of loot off of Naj'entus and Shade of Akama, leaving us at 1 piece of loot per boss. So requiring multiple pieces of loot from a boss would take half of our progress this week out of the ranking.
I'm wondering if there's any way, once all of the guilds are archived, to find new members and remove the gear they're currently wearing from the guild's loot pool. That would certainly solve the problems with transfers.
However, your kill of Supremus (with non-sharded loot) would have precluded your killing of Naj'entus, so I would assume that the system would recognize you had downed him despite having only 1 piece of his loot in your guild.
However, your kill of Supremus (with non-sharded loot) would have precluded your killing of Naj'entus, so I would assume that the system would recognize you had downed him despite having only 1 piece of his loot in your guild.
Currently, no. Many guilds have Azgalor down but not Kazgoral -_-
But I wasn't talking about Hyjal! Seeing as how BT is linear with no optionals (as far as I know).
Hyjal is a linear instance as well, a guild could not kill azgalor without first killing kazrogal. Our page also lists us with azgalor down and no kazrogal, we may have sharded something but i can't quite remember
Here is a query out of my mob database for Karazhan and Gruul's Lair values, for some concrete numbers:
+------------------------+-----------+-----------+
| BossName | KillValue | DropValue |
+------------------------+-----------+-----------+
| Rokad the Ravager | 1 | 0.001 |
| Hyakiss the Lurker | 1 | 0.001 |
| Shadikith the Glider | 1 | 0.001 |
| Attumen the Huntsman | 2 | 0.002 |
| Moroes | 3 | 0.003 |
| Maiden of Virtue | 5 | 0.005 |
| Romulo & Julianne | 4 | 0.004 |
| The Big Bad Wolf | 4 | 0.004 |
| The Crone | 4 | 0.004 |
| The Curator | 5 | 0.005 |
| Terestian Illhoof | 6 | 0.006 |
| Shade of Aran | 6 | 0.006 |
| Netherspite | 6 | 0.006 |
| Chess Event | 4 | 0.004 |
| Prince Malchezaar | 8 | 0.008 |
| Nightbane | 8 | 0.008 |
| High King Maulgar | 10 | 0.010 |
| Gruul the Dragonkiller | 12 | 0.012 |
+------------------------+-----------+-----------+
That's pretty interesting, and seems fairly reasonable. I am mainly curious now how you scale the later instances, as a 1/1000 scaling factor does seem just like a subtle variation as you intended. The separation still seems a bit large, though, and I can't pin down the reason, barring some misunderstanding on my part. (As a concrete example, I was looking at Forgotten Aspects {53} vs. Vis Maior {54}, two fairly well known guilds.) I am ignorant on the simplicity or difficulty of Hyjal and Black Temple, though, so the case in question may be out of context.
In Karazhan, I can understand how Chess should be undervalued (and maybe even moreso than you have it, but that's not important), but options in SSC/TK do not necessarily reflect such ease. Nonetheless, and out of curiosity, are you devaluing Void Reaver, The Lurker Below, etc... in the same way that you seem to be doing here?
Very nice Site! I've always hoped something like this would be created.
In regards to the issue of server transfers with gear farther in progression then the guild they are joining. The only thing I could think of that was not already mentioned would be include a series of conditions that flag boss kills as suspect. This information could be displayed on an admin page (somewhere) for you (or whoever is an admin) to review, verify (check guild websites, forum progression threads, etc) and then either approve or deny.
Conditions that would 100% indicate a transfer:
A Guild has Vashj loot but isn't 5/6 in SSC.
A Guild has Kael/MH/BT loot but isn't 3/4 in TK.
A Guild has BT/MH loot but hasn't killed Kael.
A Guild has BT loot but isn't 1/6 in MH.
A Guild has loot from MH boss Y without killing MH boss X.
Conditions that almost surely indicate a Transfer:
A Guild has Magtheridon loot but hasn't killed Gruul (I've never heard of this happening)
A Guild has SSC/TK loot but hasn't killed Gruul. (Haven't heard of a guild skipping Gruul and doing VR or Lurker/Tidwalker, etc)
A Guild has Kael loot but hasn't killed Vashj. (I don't think any guild has killed Kael before Vashj).
A Guild has Alar/Solarian loot but hasn't killed Void Reaver. (I don't think anyone skips VR)
Conditions that WILL create false positives from time to time
A Guild has Leo the Blind loot but isn't x/6 in SSC. (Leo is usually done 5th, but I've guilds kill him as their 3rd boss because they lacked tanks, so perhaps set x = 2 so there are fewer false positives)
The above example with Leo the Blind can be applied to other unlikely boss combinations as well.
This won't catch everything, but it should catch the larger gaps in progression, and gaps that simply can't exist.
The only downfall is that manual verification by you or other moderators that would need to be done. However since server transfers from a higher progressed guild to a lower progressed guild don't happen that often, hopefully the verification list would be managable.
I wish you the best as you continue to work on the site.
But I wasn't talking about Hyjal! Seeing as how BT is linear with no optionals (as far as I know).
After Supremus, what stops you from killing Teron, Gurtogg or Reliq before Akama, besides the retardation in such choice. A guild with a large number of terribles could probably kill Gurtogg and not Teron (luck binding both sides). Either way, I suppose any guild killing "further" content in BT could be assumed killing other things (just due to natural difficulty of bosses), but Hyjal is definitely more linear.
Cool list but it doesn't list two horde guilds on my server that have had Mags down for several weeks (RareSpawn and Baahaa) and one guild even has several kills in SSC (RareSpawn). My guild is listed but I was just curious as to why it would leave those two out.
It has my server pretty accurate but 2 problems I see from just checking my server:
#1: Progression has always been about doing it first. On our server my guild is shown as 4th despite having killed the same bosses as # 3 only we did them first.
Edit: I see we're behind #3 because of the point system. We've taken no loot from the animal bosses in Khara and therefor lack a point or two. IMO the animal bosses should be a zero value 'boss' for the point system. If you even bother to go kill them the odds you get any loot worthwhile (I remember we got a mail belt of +defense) is quite low.
#2: We killed Kazzak but nobody took any of the loot so we don't show as having killed him =\ (Edit: This is basically the same as my Edit under #1)
Last edited by Torrential : 07/03/07 at 5:33 PM.
Reason: Further inspection of the site
Very nice site, looking forward to see the EU version. Couple of small issues that come to my mind:
- what about servers where PvE progression is more or less unrelated to guilds? Take for example my server Earthen ring;EU/RP (when I was playing the game that is :P ), most of the PvE progression is by "raid groups" that consist of people from various guilds. The guild name you see on the left is not the people that I went to Naxxramas with. So might be interesting results there, even though it's not really a big deal...
- also the issue of ranking guilds that have the same progression; in the EU version, there will be several guilds with all the content cleared, will the one with the fancy golden wings be the guild with the guy who looted an item from Rokad the ravager?
My second thought: Blizzard really cares a lot about raiders!
My rough estimate says there's 75 US guilds that can even zone into Black Temple. There's maybe 200 guilds that can engage Kael'thas. Even estimating a roster of 40 raiders per guild, and a similar number of guilds with this much progress in Europe, that adds up to 6000 people in Black Temple and 16000 people who have experienced all of the first two raid instances in TBC. And WoW's subscriber base is what, 8 million active accounts?
As it is with the real-world population, Chinese subscribers are the largest demographic in Azeroth, totaling 3.5 million players, Blizzard said. Next up is North America, with 2 million players, followed by Europe, with 1.5 million--the remainder of subscribers are in other territories,
So 3.5 instead of 8.
On topic, thank you for this, Aspir. I was very curious exactly how far off a self report system like Bosskillers would be, but then I noticed we were there. A guild member took it upon himself to go there, so I reasoned that more likely then not, most guilds at or above us would have had a guy like that anyway, so reporting wouldn't have been wildly off. Interestingly, using only ourselves as a data point (which I know, is useless), Bosskillers was reasonably accurate (validating with your site's positioning of us, at the same point... c'mon weekly update, scan those new loots!)
Everybody is your brother until the rent comes due.
- I don't think the three "animal bosses" in Karazhan should count at all. They don't drop loot that anyone wears and most guilds kill maybe one before figuring out it's a waste of time. I don't think another guild should rank ahead of mine just because they killed Shaditkith the Glider.
- It looks like all SSC/TK bosses are weighted ahead of Magtheridon. My guild has killed Mag but are just getting started in those instances, but I honestly feel that we are more progressed guilds that skipped Mag in order to do Void Reaver or Lurker. I think that Magtheridon should be worth less than Vashj/Kael but more than the other SSC/TK bosses. At least put him ahead of Void Reaver.
I'll chime in with the consensus - nifty, nifty site. Lots of info, but not crowded, either.
I notice a few guilds here and there with credit for Gruul, but none for HKM. It seems obvious why - no one has HKM loot worn. It might indicate a transfer, but it could also indicate that guild just didn't like HKM's loot.
At first, I was thinking it would be good to boost a guild's score by taking these "gatekeeper bosses" into account. For instance, if the Armory indicates members had to have killed Gruul four times to get the drops showing, then credit them with at least four HKMs as well. The possibility of transfers makes me wary of doing this, however.
Assuming Aspir's database has the resources to track it all, it might also be instructive to compute a different index - call it RawKillIndex - that simply counts all the times a given guild has killed a given boss, based on the loot seen. So if the Armory shows - as conclusively as is necessary to satisfy WowJutsu's current requirements - that a guild's membership has 14 drops from a 3-drop boss, then that guild gets 5xKillValue for that boss. If that boss has a gatekeeper, then that guild gets at least 5xKillValue for the gatekeeper, too.
You could even figure in loot lists - guild has 10 drops from a 2-drop boss, but 8 of those are things that boss only drops once per kill, etc.
Meanwhile, I've heard tales of guilds getting Mag before Gruul, but granted, I have no hard evidence of such.
Magtheridon drops t4 chest, whereas lootreaver drops t5 shoulders. I don't see why it should be placed otherwise (and besides the fact that there was keying requirements before the patch that forced you to do Magtheridon first)
<Eej> YOU"RE GONNA PULL
<Eej> IF YOU SQUEEZE OFF ANOTHER ARCANE BLAST
<Spectear> You've obviously never played with Manly.
<Spectear> That's hardly a reason to stop DPS. Very Manly Staff
Magtheridon drops t4 chest, whereas lootreaver drops t5 shoulders. I don't see why it should be placed otherwise (and besides the fact that there was keying requirements before the patch that forced you to do Magtheridon first)
Not to mention it isn't as if Aspir concocted this website overnight. You can damn well assume that he was working on it well before the attunements were dropped. That being said, T5 > T4. If you think 1 boss is easier than the other, well go kill it -- That is what progression is all about.
I agree that the 3 Kara bosses should be dropped from counting as well, really nobody would use the loot so...
07/03/07, 4:00 PM
Similar Threads
