Wednesday, October 31, 2007

Derived Stats and Spell Damage

The damage done by a spell or special ability (non-crit) generally takes the form of:

TotalDamage = BaseDamage + Coefficient * PowerStat

For melee attacks and abilities, the PowerStat is Attack Power. For spells, the PowerStat is +damage or +healing.

What's interesting here is that Attack Power (AP) is built into the game from the very beginning. It's a derived stat, meaning that it's value comes from other primary stats. Different classes use different primary stats to boost AP. Warrior, paladins, and shamans boost AP through Strength. Rogues and Hunters boost AP through Agility. Druids use a mixture depending on what form they take.

In contrast, spell damage is tacked on to this system, and only starts appearing at the high levels. Even though it behaves in the same manner as AP, it is not a derived stat, but a completely separate one. It's also different in that it's actually a collection of PowerStats, each boosting a specific spell school.

In my opinion, the Attack Power system works better than the spell damage system. It's simpler and more elegant. Additionally, it works with the five primary stats, and allows different classes to value gear in a different manner. A ring with +Agi means different things to a warrior, a rogue, or a hunter.

It's probably too late for WoW now, but reworking spell damage to mirror Attack Power would have a lot of benefits. Let's call the new PowerStat Spell Power (SP). For example, mages and priests could get Spell Power from Spirit, while warlocks and druids could get Spell Power from Intellect.

This immediately differentiates the classes. You could tweak the rate at which Spirit or Intellect is converted into SP and balance each class without seriously affecting itemization. For example, maybe 1 Spirit = 3 SP for Mages, but 1 Intellect = 2 SP for Warlocks.

The other major advantage comes with paladins and shamans. The melee-magic hybrids. For these two classes you could have Strength convert to Spell Power as well as Attack Power. This would mean that both sides of the class would scale with the same stat.

We don't really need a power stat for each separate school of magic. The only place where it's significantly used for damage spells is the tailoring epic sets. 90% of the other damage gear in the game only uses +damage/healing.

As for +healing, let's pretend that you get twice as much +healing as +spell damage. (It's not exactly twice, but it's pretty close.) So 1 Spell Power = 2 Heal Power. Taking a look at the equation above, we can go:

TotalDamage = BaseDamage + Coefficient * PowerStat
TotalHeal = BaseHeal + Coefficient * ( HP )
TotalHeal = BaseHeal + Coefficient * ( 2 * SP )
TotalHeal = BaseHeal + ( Coefficient * 2 ) * SP

Instead of using a different PowerStat for healing spells, we can use the same PowerStat and simply double the coefficent on healing spells. It accomplishes the exact same thing!

Converting the spell damage system into a mirror of the Attack Power system would accomplish many goals. It would simplify the number of stats on gear. It would allow you to differentiate classes by getting them to place different values on the primary attributes. It makes the primary attributes more valuable for spellcasters. It allows melee-magic hybrids to scale with one stat, rather than chasing two. It causes both offensive and defensive spells for healers to scale at a similar rate, rather than having extremely powerful healers who can't hurt a fly.

Tuesday, October 30, 2007

Housecleaning

I've upgraded to Blogger's new templates. As part of this upgrade I've decided to drop the pseudonym "GSH" and simply use my real name when signing posts. I'm not really sure why I'm doing this. To be honest, the double layer of indirection (Coriel -> GSH -> Rohan) was getting a bit weird, especially when responding to emails.

I'm totally fine with signing a blog post or a forum message anonymously, but it somehow seems wrong to sign an email as GSH. It may be the fact that a blog or forum post is a one-to-many relationship, while an email is one-to-one.

Since I'm using my real name now, I figure I might as well use a more real email address as well. It'll save me the trouble of actually remembering to check the one I made specifically for this website.

Other than that, the upgrade broke my essential posts section, so I'll have to rebuild that. It doesn't look like I'll be able to easily do it in a nice compact form for the various series of posts.

If you notice anything that has changed for the worse, or is missing, please post a comment and I'll try to fix it.

Saturday, October 27, 2007

Server Mergers

Several people, including Tack, have suggested server mergers as a solution to the problem of consolidating talent. I don't think this will solve the problem. I've been watching the show Numb3rs lately, and I feel inspired to make an analogy which may not actually reflect reality or mathematics.

<Professor Charles Eppes>

Imagine that a WoW server is like a bucket with a hole in it. The people on the server are like drops of water. If you fill up the bucket, water leaks out, and the bucket eventually empties.

A high-end raiding guild is like a plug. You put the plug in the hole, and the bucket stops leaking water.

Merging two low population servers is like pouring water from one leaky bucket into another leaky bucket. The amount of water increases, but the increase is only temporary.

The bucket still has the hole and, without the plug, the bucket will still eventually empty.

</Professor Charles Eppes>

Might work a little bit better with blurry visuals and random fluid dynamic equations being written out with chalk.

The analogy is crude, but you can see why I believe server mergers won't fix the problem. If you're lucky, a server merger may result in the creation of a hardcore guild which revitalizes the server. But that's not guaranteed, and the option for the hardcore to transfer is still very attractive.

Thursday, October 25, 2007

The Consolidation of Talent

I believe that Blizzard spends a disproportionate amount of resources on endgame content. The number of people who will see the Black Temple or Sunwell is a tiny fraction of the playerbase. The amount that this fraction pays in subscription dollars does not cover the cost of content produced for them, or the amount of customer support they use.

I also believe that these resources are being spent well. That it is important to retain the hardcore, to create content for them and keep them in the game. It is my belief that the hardcore have many beneficial effects for an MMO. They provide a spectacle for the other players, are the engine which drives the economy. They serve as aspirational models, and players that everyone knows about.

The hardcore act as hubs in the social network of the game. Without the hardcore, the world would seem emptier, and the network that binds people together, keeps people playing, would fray, and I believe the casuals would fall away.

So if the above is true, let us consider the impact another aspect of WoW: Paid Server Transfers.

The hardcore seek out the hardcore. They want to progress, and the best progression is found with like-minded people. And because they are the hardcore, they are willing to pay the fee to move to a different server. High-end raiding guilds recruit across servers. There are servers where the hardcore have congregated, which boast multiple Illidan-killing guilds. Servers like Mal'Ganis or Korgath.

This Elitist Jerks thread is an example of the situation. The majority of high-end guilds are on PvP servers. The hardcore on PvE servers would like to be able to transfer to such servers. And the phenomenon magnifies itself. If some of the hardcore leave the server, it becomes harder for the remaining hardcore to succeed, making it more likely that they too will transfer.

But what happens to the old server when they do transfer?

If the hardcore are nodes in the social network of the server, removing the nodes damages the network, and will eventually causes casuals to leave. And remember that the hardcore will consolidate themselves on a few servers. For every server the hardcore flock to, there are five or more servers being abandoned by them. Casuals will not pay to transfer servers. If the social network of their server is too badly damaged, they will simply leave the game.

On the other hand, if the hardcore are not hubs in the network, if the server will happily survive without them, then you really have to wonder if spending all those extra resources on the hardcore is worthwhile. Wouldn't it be more profitable to cater to the customers who don't require as many resources to satisfy?

Paid Server Transfers allow the hardcore to consolidate themselves on a few servers, damaging the many servers they leave behind. Blizzard needs some mechanism to push the hardcore in the opposite direction, to cause them to spread out. In the past, overcrowding and server stability was one of the main forces opposing the drive to congregate. Unstable servers caused Death & Taxes to move to Korgath, revitalizing that server. But as Blizzard's technology improves, it is less likely that unstable servers will push the hardcore to spread out.

Sunday, October 21, 2007

Ask Coriel: Improving a Retribution Paladin

Firelight, from Steamwheedle Cartel, writes in:
As a raiding retribution paladin i'm really looking forward to 2.3 and the upgraded DPS it brings for me!

I wondered if you could take a look at my gear and tell me what you thinked about it, and what i could do to improve! Obvious upgrades for my are t4 shoulders and legs, and i'm only 5 nethers away from upgrading my chest armour as well.

Being one of the only ret-raiders in my server (that i know of) i've been finding it hard to gain acceptance from my peers who just whisper about me behind my back (retnoob etc) so i'm just wanting to be accepted by being able to do more dps!

You will need to change your talents once 2.3 comes out. I would suggest taking some points from Benediction and finishing off Fanaticism and Improved Blessing of Might (Build). Also, if you don't need Blessing of Kings, you could take 3 points from Protection and put them in the new Pursuit of Justice (Build - the last point goes in Pursuit of Justice). I'm a fan of speed increases as less time running equals more time DPSing. If you need to keep BoK for raiding, consider putting the Cat's Grace enchant on your boots.

On gems and enchants, I think you are putting a little too much weight on critical strike rating/agility, and not enough on Strength. Your crit rating is nice and high, but your AP is a bit low. Also, always go for Strength instead of pure Attack Power, because you get bonuses from Divine Strength and Blessing of Kings.

As well, once 2.3 comes out you're going to need some more hit rating, and you'll get 4% extra crit from talents and racials. Not to mention that Vengeance will last twice as long, so you can keep it up with a lower crit rating.

I would go:

Red slots = Bold Living Ruby (+8 Str)
Yellow slots = Rigid Dawnstone (+8 Hit) until you have 95 hit rating, then Inscribed Noble Topaz (+4 Str/+4 Crit)
Blue slots = Sovereign Nightseye (+4 Str/+6 Sta)
Meta = the one you have is good, but you might also want to consider Relentless Earthstorm Diamond (+12 Agi/+3% critical damage)

As for enchants, you look pretty good. Put a +6 all stats on your chest. If you can squeeze out an extra +16 hit rating (maybe from gems), consider using the Lower City head enchant. It will give you a little more AP and some more mana.

To be honest, you look pretty solid already. You just need to tweak things when 2.3 hits, and I personally would place more emphasis on Strength over crit rating when gemming.

Also, and I'm just throwing this out as an option, if you switched from armorsmithing to weaponsmithing, you would get access to some really nice weapons. But that's a big change, and you do have a decent weapon in Hammer of the Naaru, and you may not have a backup chestpiece.

Any other advice readers--especially Retribution Paladins--would give Firelight?

Saturday, October 20, 2007

The Paladin Community

Looking over the comments on Threat Reduction for Retribution, it seems that people on the outside don't really understand the paladin community. You see comments like "First you asked for X, and now you're asking for Y?"

The paladin community is the most fractured of all the classes. There are four main factions: Holy, Protection, Retribution, and Hybrid. The different factions want different--often opposite--things and that leads to the confusing nature of paladin feedback. Each side generally champions one tree, and does not really care about the others. The only exception are the Hybrids, who generally support all three trees, but oppose efforts to overspecialize them.

The next complication is that the Holy faction is at war with the Retribution faction. Retribution derides Holy as "healbots", and Holy calls Retribution "retnoobs". If I can indulge in some armchair psychology, the reason both these factions are at each other's throats is because a paladin is both melee and healing/support. Protection and Hybrids still try do both melee and support, though in different ways. But Holy gave up melee, and Retribution gave up healing, and by attacking each other, they try to convince themselves that what they sacrificed was not of value.

So for example, the Retribution faction asked for threat reduction and increased DPS in raids. The Holy faction opposes out of habit. Protection doesn't really care, and the Hybrids are generally supportive.

Then Retribution gets permanent threat reduction, and the Retribution faction is happy. However, the way the threat reduction is implemented hurts the Hybrid ideal, and thus the Hybrids are the paladins objecting.

The final complication is race. Alliance paladins generally view the Blood Elves as interlopers, and the Blood Elves are more than happy to antagonize them back. As well, a large percentage of vocal Blood Elf paladins are Horde raiders who rerolled after seeing the effect of Alliance paladin healers. Where the Alliance paladins were sort of "pushed" to the healing side when they started raiding, the Blood Elves embraced it from the very beginning, and are often the most militant of the Holy faction. Ironically, the Horde-specific Seal of Blood is extremely good for Retribution, a fact which seriously annoys the predominantly Alliance Retribution faction.

So that's a quick guide to the paladin community, and why different paladins complain about different buffs/nerfs. I belong to the Hybrid faction, and thus I'm generally supportive of all three trees, but not of talents that overemphasize one aspect of the paladin at the expense of the others.

Wednesday, October 17, 2007

Question about Raiders

I'm thinking through an idea for my next post, which which will be relatively complex. But I want to make sure that I am not missing a possibility somewhere, so I'm asking for feedback on a couple of underlying assumptions.

Assumption 1: 25-man raiders are a small minority of the playerbase.

Assumption 2: A disproportionate amount of development time is spent creating content for them.

Can anyone refute these assumptions?

Also, assuming 1 and 2 are correct, why is it worthwhile for Blizzard to create content usable only by such a small portion of the player base?

I have my own ideas about that, and I've written about them before, but I'd like to see if anyone has any other reasons that I may not have considered.

Edit: For reference, here is the last article I wrote on the subject.

Tuesday, October 16, 2007

Threat Reduction

Blizzard is giving Retribution Paladins threat reduction:
After further discussion and testing we’ve decided to add threat reduction deep in the paladin's retribution tree. Fanaticism will now reduce threat caused by all actions by 6/12/18/24/30%, in addition to its current effect.

It seems like Blizzard is equalizing all the melee DPS specs at 30% threat reduction. Rogues, Feral Druids, Fury Warriors, Enhancement Shamans, and now Retribution Paladins will all have the same level of built-in threat reduction.

The only negative is that this talent will make it very hard for Retribution Paladins to tank or off-tank. So much for my Ret OT build. To be honest, I'm not sure I really approve of building such a large, permanent disadvantage into a talent. It seems against the spirit of talents.

In general, a paladin with talent points is always equal to or better than a 0/0/0 paladin at every task. This holds true for pretty much every class. But now it's possible that a Ret paladin will be worse than a 0/0/0 paladin at tanking. And that doesn't really seem right to me. Maybe the increased damage from the other Retribution talents will make up for the loss of threat.

I mean, Shadowform is an awesome talent for Shadow Priests. Would it still be a good talent if it was permanent? 99% of the time a Shadow Priest will be in Shadowform, but being able to turn it off is a lifesaver for that 1%.

Still, from a pure DPS point of view, this is a massive buff. Now, just kill Seal of Vengeance and give all paladins Seal of Blood, and Retribution will be in solid shape.

Then Blizzard can start work on the harder task: fixing Holy.

Friday, October 12, 2007

Hit Caps Post Updated

I've updated my Hit Caps for Bosses post to reflect the Expertise changes in 2.3. Take a look and tell me if you see anything that should be changed.

Also, if anyone can point me to the definitive Dodge cap number, that would be greatly appreciated.

Thursday, October 11, 2007

2.3 Patch Notes

The 2.3 Patch Notes are out. Some first impressions follow.

Expertise: We have added a new stat and associated rating called expertise and expertise rating. Expertise rating converts to expertise at the same rate that weapon skill rating formerly converted at. Each point of expertise reduces the chance for your attacks to be dodged or parried by 0.25%.

Weapon Skill: All items and abilities that granted weapon skill have been changed. In most cases, they were converted to expertise or expertise rating. Ranged attacks do not benefit from expertise, so ranged weapon skill has generally been replaced by critical strike bonuses or hit bonuses. In a few cases, talents have been changed to other effects to avoid granting players excessive amounts of expertise


Bah, I have to go rewrite my +Hit Caps guide. Basically, this removes the disparity between the first 5 points of weapon skill and the subsequent points, which is a good thing. Hit caps will generally rise across the board for physical damage classes. At first glance, expertise rating is roughly equal to hit rating for melee dps classes, and twice as good for tanks. It also raises the possibility of capping out both hit rating and expertise rating, and always hitting/critting the mob.

Healing: Almost all items and enchantments that provide bonus healing now also provide a smaller number (approximately 1/3) of bonus spell damage. There are a few items and enchantments where this was not possible, such as random-stat items and Zul'Gurub enchantments, but this is now the case on virtually all other items.

Awesome! Simply awesome. I can solo in my raiding epics, and even add extra damage in between heals. Heck, I might even be able to disenchant my spelldamage set now and free up a entire bag worth of bank space!

Blessing of Light: Lower ranks of Flash of Light and Holy Light are now properly penalized when used with this Blessing.

This nukes downranking Holy Light pretty hard. Which is how it should be. In my opinion, costs are what maintain balance, and messing around with costs leads to degenerate gameplay. To be honest, I think Blizzard should do away with down-ranking entirely, not even offering it as an option. I think the game would be better for it.

Cleanse and Purify range increased to 40 yards.

Another great change. No more healing and then finding out you're out of range of a Cleanse.

Crusader Strike (Retribution) cooldown reduced from 10 to 6 seconds.
Exorcism mana cost reduced.
Hammer of Wrath mana cost reduced.
Holy Wrath mana cost reduced.
Improved Seal of the Crusader (Retribution) benefits folded into the base spell. This talent now gives the benefits of the Sanctified Crusader talent instead.
Judgement of Light: The combat log will now show the mana gained from rank 5 of this ability as Judgement of Light instead of Seal of Light.
Pursuit of Justice (Retribution) is now 3 ranks and increases movement speed by 5/10/15% and also reduces the chance you'll be hit by spells by 1/2/3%.
Sanctified Crusader (Retribution) renamed Sanctified Seals, which now increases your chance to critically hit with all spells and melee attacks by 1/2/3% and reduces the chance your Seals will be dispelled by 33/66/100%.
Vengeance (Retribution) duration increased from 15 to 30 seconds.
Vindication (Retribution) frequency and duration increased and now reduces all attributes by 5/10/15%, not just Strength and Agility.


Various Retribution buffs that seem aimed at PvP for some reason. I'm sure that it will help there, but threat reduction seems a curious omission, given that DPS warriors and Enhancement Shamans got extra threat reduction.

Weapon Expertise (Protection) renamed Combat Expertise, now increases expertise by 1/2/3/4/5 and total Stamina by 2/4/6/8/10%.

Solid change. I predict that the standard Protection build will drop Reckoning for Weapon Expertise. As well, with the changes to Improved Seal of the Crusader and Pursuit of Justice, the new challenge will be bleeding enough points from Protection to fill out Retribution.

Corpses that belong to someone in your party, but that you cannot loot will say in the corpse mouseover who has loot rights on that corpse. This will help skinners who want to know who to talk to in order to skin their corpse, as well as master looters when special loot drops on a normal monster. Everyone will then see that there is a monster that the master looter has loot rights on.

It comes late, but no raid shall ever be faced again with an unlooted Core Hound. Let us mourn the passing of the defining experience of Molten Core: "Loot the Hound!"

Lowered the fireball damage of Defias Pillagers.

Their unholy reign of terror has finally come to an end.

Elite mobs outside of pre-Burning Crusade dungeons have been changed to non-elite.

The burning question of 2.3: Has Hogger been nerfed?

Other than that there's a lot of UI cleanup changes, new low level content, a new 10-man, and guild banks. There's also some crazy Arms/Fury warrior changes which I don't really understand. All in all, a very full content patch. I don't think PvE Retribution paladins are going to be too happy, but it looks fairly decent otherwise.

Wednesday, October 10, 2007

Time to Max Level

In patch 2.3, Blizzard is reducing the amount of experience required to level by 15% (for levels 20-60). Most people are happy about the change, though there are some people (such as Tobold) who questioning if this devalues the existing levelling game.

I would like to look at it from a sightly different angle. Let's define a concept called Time to Max Level. It's basically the amount of play time the average player would have to spend for her first character to reach the level cap.

Here's my question:

Should Time to Max Level depend on the value of the max level?

I think that it should not, that Time to Max Level should be independent of the numerical value of the max level. That there's a sweet spot, probably around 8 months, where someone who is new to the game and plays a couple of hours a week can eventually reach the cap. Reaching the cap is a major milestone, and should be in reach of every player. If that basic goal seems out of reach, it's very discouraging.

When WoW first came out, many reviews praised it for being easy for even casual players to hit the level cap. The actual number of the cap didn't matter, only that people could reach it.

But if it takes 8 months to reach 60, and 4 months to go from 60-70, that's 12 months to reach the max level. And when the next expansion comes out, that's another 4 months. Soon it will be impossible for a new player to reach the max level before the next expansion comes out.

Of course, if you add more levels, you need to make the levelling time faster for the earlier levels in order to keep Time to Max Level constant. So I think that Blizzard is on the right track with speeding up 20-60 levelling. I hope they continue this trend with the next expansion, speeding up 60-70, and trying to keep hitting the level cap within the reach of even the most casual players.

Sunday, October 07, 2007

Choosing Your Battles

With the removal of attunements, there is a point where a T4 raiding guild is faced with multiple possible paths for progression. Many guilds, in my opinion, are choosing a path which looks deceptively easy, but has a lot of negative repercussions.

After a guild kills Gruul, there are three or four possible choices for the next boss to focus on: Magtheridon, Hydross/Lurker in Serpentshrine Cavern, and Void Reaver in Tempest Keep. Of these bosses, Void Reaver is the easiest, and Magtheridon is probably the hardest. Given that Void Reaver drops T5 shoulders, a lot of guilds choose to tackle him after Gruul. I believe that this is a mistake, which ends up hurting the guild progression in the long run.

Casual raiding guilds, especially those on a limited schedule, need to organize their schedule for progression. There are three competing constraints that a raiding guild needs to follow:
  1. Maximize the time spent on Progression instances.
  2. Minimize the time spent on Farming instances.
  3. Maximize the number of epics gained from the Farming instances.

You need to spend time on boss fights to learn them. The more time you spend on a fight, the better your progess is. But you still need to farm a beaten instance for gear in order to improve your raid. Spend too much time farming, and you won't get the experience on fights that you need to beat them. Spend too little time farming, and your raid won't improve gear-wise. (Not to mention that your raid will get discouraged from the lack of loot.)

The key to sustainable progression lies in how you manage your farming time.

So, keeping this in mind, let's look at the three options again. What happens after you put the target on farm?

If you kill Void Reaver or Lurker/Hydross first, your farming instance is Gruul's Lair. You'll get about 7 epics for your farm night, and you'll probably have to add Karazhan runs to supplement the extra gear. Or you'll end up working on SSC, but farming Gruul's Lair and Void Reaver, which takes a fair bit of time and nets 10 epics.

On the other hand, if you tackle Magtheridon first, it may take a bit longer to learn the fight, but once he's down you have a farm night of Gruul's Lair + Magtheridon, which is very fast and gives 12 epics. This allows you to give the maximum amount of time to the next raid instance.

In fact, I believe that a guild should ignore Void Reaver until SSC is complete and becomes your farm raid. Even though you can kill him earlier, going after Void Reaver earlier is non-optimal for the three constraints listed above.

In my opinion, raid progression should go in the following order:
  1. Karazhan
  2. High King Maulgar/Gruul/Magtheridon
  3. Serpentshrine Cavern
  4. Tempest Keep
  5. Mount Hyjal
  6. Black Temple

You should concentrate on one instance, and farm only the instance behind it. As you move down, you drop the oldest instance in favor of farming the next instance. For example, once you've killed Magtheridon, move into Serpentshrine Cavern and drop Karazhan. Once you've killed Lady Vashj, move into Tempest Keep and drop High King Maulgar/Gruul/Magtheridon.

Note that this list is not organized in order of boss difficulty. The objective here is not to cherry pick bosses as you are able to, but to optimize for the three constraints. Maximize progression time. Minimize farming time. Maximize loot from farming.

I believe that is the best recipe for long term success.

Thursday, October 04, 2007

Blizzard's Secret Weapon

I don't really understand game companies. They say a lot about copying Blizzard, but then they never do. There has been one constant about Blizzard games ever since the first Warcraft: Odds are the game will run on your machine.

And this is crucial. If your machine can't run the game, why would you buy it? Blizzard constantly sets the system requirements low. They make their games available for Macs. They deliberately stylize their art so that it still looks good on low end machines.

Warcraft, Diablo, Starcraft. Time and time again, Blizzard pulls the same trick, and no one else seems to learn.

And yet other companies do not seem to realize this at all. If you can't run the game, you won't buy the game. I play on a laptop, with a built-in video card. No nVidia, no ATi. And yet WoW runs, looks halfway decent, and thus I give Blizzard my money. I'm not going to go spend a couple thousand dollars just to play your game.

It's not the only reason Blizzard is successful, but I think it's a major one.

Wednesday, October 03, 2007

The 10-man to 25-man Transition

I used to believe that Blizzard made a mistake by starting raiding with a 10-man and then forcing guilds to move to a 25-man model. I thought that running two Kara groups and having to increase in size would be too much work and organization for most guilds, and would prevent the majority of guilds from really experiencing raid content.

I think I was wrong.

The data that has forced me to change my mind comes from WowJutsu. WowJutsu is an automated ranking site that crawls through the Armory. It notes when new loot from raids appears on characters and uses that to determine which guilds have killed which bosses. It's not perfect, but it is fairly accurate for the most part.

On the side, they have an automated list showing what percentage of raiding guilds have killed each boss. For example, only 3.37% of raid guilds have killed a boss in Black Temple. By comparing the number of guilds on two bosses that come right after each other in progression, we can see which bosses guilds are "stuck" on.

When examining the 10 to 25-man transition, we look at the end bosses of Karazhan and the first boss of Gruul's Lair, High King Maulgar. First, we make an assumption:

Assumption 1: If your guild has killed Maulgar, your guild has killed Nightbane.

This is a pretty reasonable assumption to make, in my opinion. From WowJutsu, we see that 74.51% of all raid guilds have beaten Nightbane, and 72.68% of all raid guilds have beaten High King Maulgar. This means that the vast majority of guilds who beat Nightbane go on to beat High King Maulgar. That they are able to overcome organizational challenge of getting 25 raiders. The conversion rate is a whopping 97.5%.

But maybe the assumption is wrong. Maybe there's a number of guilds who have killed HKM, but not Nightbane, and that's skewing our results. So let's soften it to:

Assumption 2: If your guild has killed Maulgar, your guild has killed Prince Malchezzar.

Honestly, if your guild has killed HKM, there is zero excuse for Prince Malchezzar to still be alive. Now we see that 88.88% of guilds have killed Prince Malchezzar. Again, the majority of guilds who beat Malchezzar go on to beat HKM. The conversion rate here is 81.8%. That's lower, but it's still very reasonable. 80% of all raiding guilds are able to navigate the transition from 10-man to 25-man.

I think that this is a reasonable result. It's not perfect, but between the Nightbane and Prince Malchezzar numbers, it's evidence that going from a 10-man to 25-mans was not a mistake, and the transition is within the capabilities of most guilds.

Tuesday, October 02, 2007

Respawns and my Guild

After some thought, I think I might hate trash respawns because they are disproportionately affecting my guild.

My guild raids 3 days a week, for three hours a night, from 6:30 to 9:30. We can go longer, but there are people who have to leave at 9:30, so we have to do swap-outs.

Most trash now respawns on a 2-hour timer. That means that trash respawns in our third hour of raiding. So we're always faced with the choice of reclearing trash and extending the raid to get more attempts in; or calling the raid early. We very rarely get to end the raid on time.

If trash respawned on a 4-hour timer, we would almost never see it respawn, and thus I would not really care about respawns. Or if we raided for four hours a night, it would be an inconvience, but a relatively minor one as we could reclear and get in several more attempts before ending on time.

Is a three hour raid window unusual? I don't think so. I think it's a very natural slice of time for a more casual guild to schedule a raid. Unfortunately, it just happens to mesh badly with 2-hour trash respawns.

Monday, October 01, 2007

My Gear Philosophy

One activity that I find very interesting is the making of gear lists. You see people making lists of their ideal gear, and then running instances over and over until they get the exact gear that they want.

This is something that I don't do. Mostly because Lady Chance and I have an understanding: if I want an item, it will not drop.

The only items that I have ever really wanted was the paladin Tier 2, Judgement Armor. I was in a guild that cleared Blackwing Lair. Yet I entered TBC with only one of the eight pieces (the gloves, I believe). I never got my helm to drop from Onyxia. I lost the roll for the legs to a paladin wearing a necklace from Scarlet Monastery.

So I follow a different philosophy: If you do stuff, loot will drop. Do a variety of instances, and upgrade whenever possible. Even if the item is not the "ideal" item for that slot, take it if it is better than your current gear. Spend your DKP freely, and do not hoard it for the perfect item. As long as you keep doing instances, things will work out. Work on all your sets at the same time. If you are Holy, and a good Ret piece drops, spend DKP on it.

Of course, show some courtesy to your fellow players. If something is off-set for you, and main-set for someone else, let them have it.

As long as you keep doing stuff, loot will come. I don't think it's worthwhile to worry about what the "absolute" best piece of gear for the slot.

About the only exception to this philosophy is crafted gear. It's worthwhile to look over the crafted gear in the game, and see if there are any pieces there worth working towards. But this is because crafted gear is not random.

So that's my philosophy. Don't sweat the "ideal" drops, and continuously upgrade. Eventually you'll end up with decent gear.