Public Quests

Let's discuss Public Quests. First, let us start by defining what exactly a Public Quest is.

A Public Quest is:

1. In the game world. It is not instanced content.


2. Requires multiple players to defeat. It's generally group content, not solo content.


3. Repeatable. A player who completes the event once can do it again the next time it starts.


4. Requires a minimal relationship between players participating in it. You don't need to be in the same guild, group, raid, or perhaps even faction to participate.


5. Has some direction from the game. Either a questgiver or UI element directs the event. A random wandering monster, even if it meets the other criteria, is not a Public Quest. A quest to go kill that random wandering monster, on the other hand, is a Public Quest if it meets the other criteria.

So you can see that Public Quests are essentially ad hoc, repeatable group content aimed at the players in the immediate vicinity of the event. The idea is that these players would stop what they are doing, from an impromptu group, complete the Public Quest, and then return to what they are doing.

From this definition, we can immediately see three major issues with Public Quests.

First, the PQ relies on players in the immediate vicinity. If not enough players are around, maybe because it's late at night, or out of the way, then the Public Quest can't be completed successfully.

Second, the PQ also relies on players to stop what they are doing and switch to the PQ. The players might prefer to ignore the PQ. Maybe they don't think the rewards are worthwhile. Maybe they want to work on this quest right now. Maybe they'll do the Public Quest in a few minutes when it restarts. Maybe they've already done it once, and don't want to repeat themselves.

I know that in RIFT, I'd close a rift, then ride by 10 minutes later and see a new rift in the same spot. Rather than wanting to participate again, my thoughts would be more along the lines of, "I've already done this, no need to do it again."

Third, because the content is ad hoc, and the relationship between the participants is tenuous, you can't make the PQ require too much from the players. There might not be a tank, there might not be a healer, there might not be someone who can crowd control. You can't guarantee that the players will be in the same group.

The best implementation of Public Quests I've seen comes from RIFT. But it's not the normal rifts that dot the landscape. Rather it is the Rift Zone Events.

The Zone Events are rare, occurring every two hours or so. That's a lot of incentive to participate in one when it happens, because you might only see that one in your play session. Rather than saying, "I'll finish this quest, I can always do another rift after," it becomes easier to say "I'll do this Event now, there might not be another chance later, and I can always finish the quest when this is done."

The rewards are high, commensurate with the rarity. As well, the "vicinity" of a Zone Event is the entire zone, so the event draws from the widest possible pool of players (for non-instanced content, at least).

If I was designing a Public Quest system, I would model it after Zone Events in RIFT. They would occur infrequently and would be broadcast to the entire zone. Only one event would active at a time, and it would be different than the last event that occurred, to increase the novelty factor. It wouldn't necessarily need to involve the entire zone, just make the entire zone aware of it, and give players enough time to travel to the right spot, with UI support to tell them where the "right spot" is, of course.

Critical Hits, Commenters

Follow Up on Critical Hits

The commenters on my last post have convinced me that having a stat with negative effects is a bad idea.

However, I still believe that in PvP, the presence of resilience means that healing will scale much faster than damage will. Inevitably, healing will have to be reined in, probably through a debuff in arenas and battlegrounds that reduces the overall effectiveness of healing, like in Wrath.

I still think that Blizzard should normalize crits at 200% across the board. Since that debuff will probably come sooner or later, Blizzard may as well bring the debuff in sooner, and maybe at slightly higher values. But a normalized crit would make critical strike more valuable to healers, and simplify the game.

Commenters

Sometimes it seems like Tobold is at war with his commenters.

To be honest, I think he takes commenters too seriously, and doesn't wield the delete button as aggressively as he should. People respond to their environment, and push the edges a little. Prune aggressively and people will respond to that, and live up to expectations.

I mean, take the last post of mine. Two one-line "this idea is stupid" comments. Both promptly got deleted. And the next two comments, by The Crossbowman of Sarcasm and Azuriel, are solid, insightful, and useful. I strongly believe that you don't get comments and insights like those two if you allow the terrible comments to stand and even flourish.

Finally, as advice to aspiring bloggers, I don't think you should second-guess what your audience wants. If there's one thing I've learned, it's that I cannot predict what other people find interesting. Completely random posts get linked. One-off posts that you spend 10 minutes on get multiple Likes on Google Reader, while crafted epics that take you hours go uncommented on.

Just write what you want to write, rather than worrying about what you think people want you to write about. It will sort itself out.

Critical Hits and a PvP Idea

Ghostcrawler had an interesting post on critical hits today.

Back in Vanilla, the rule was that melee criticals did 200% damage, spell criticals did 150% damage, and damage/heal-over-time spells could not crit. But over time most of the DPS casters got talents or abilities that pushed their criticals to 200% damage. Pretty much the only people left at the 150% mark are healers and physical dps who get some damage from spells.

So GC is mulling over simplifying the entire thing with a rule that all crits do 200%.

Personally I think it's a good idea. It simplifies things immensely. Pretty much every DPS already has 200% crits on the majority of their damage.

As for healing, because damage crits do 200%, crit rating is expensive. So when you look at crit rating for healing it seems even more overpriced compared to the throughput it gives you. And then you add to that the fact that healers disdain crit because you can't count on it in the short term. At least bumping healing crits up to 200% would normalize the value of crit rating.

The one area of concern is PvP. A healing crit can be very swingy. Ghostcrawler suggests that the potency of the Mortal Strike debuff be increased to compensate. But then that can make having an MS class absolutely mandatory when fighting a healer.

I have a different solution. The "swingy-ness" of damage in PvP is mitigated by resilience. Perhaps we should use the same mechanic. Have resilience reduce incoming damage and also reduce incoming healing. So if you have a lot of resilience you're taking less damage, but you're also getting less healing. That will shift power away from the healers and make a random critical heal, even at 200%, much less of a game-winner.

Resilience would essentially slow the rate of change of your health bar in both directions. A lot of resilience would mean your health decreases slower, but it also increases slower while being healed.

Leveraging Successful Players in the Dungeon Finder

Here's a random idea I thought of yesterday because of all the recent talk about the Dungeon Finder and the deja vu experience.

My theory is that there exists certain players who are more successful than other players when it comes to random groups. Maybe it's skill, maybe it's gear, maybe it's patience, maybe it's even luck. But these people are more likely to successfully complete random dungeons than the "normal" player.

What if we could identify who these High-Success players are? It would be as simple as looking at their success rate for heroics where they queued solo, and comparing it to the average success rate.

Once you've identified who these players are, then when a heroic group is in trouble and needs a replacement player, instead of taking the first person in the queue, search the queue for High-Success player and toss them into the group. If the heroic group has wiped, and people have left, adding a stronger-than-average replacement would make it more likely that the run would still succeed.

You would make it more likely that runs don't completely dissolve. You refrain from throwing the first player in the queue, who might not be able to handle a weakened group, into a half-finished run that may not be successful, leading to a wasted queue and poor experience.

Plus, this would be entirely invisible to the player base. You can't really tell that something like this is happening. The High-Success player gets a bit shorter queue, but at the price of having to "carry" a weaker group.

Deja Vu

I had a really odd experience running random heroics recently.

In the last two randoms I queued up for, the queue time was about 10 minutes, but the dungeon popped at about 5 minutes in. Both times I zoned in and found myself at the same place: Shadowfang Keep, in front of Baron Silverlaine.

Coincidence? Or a glitch in the Matrix?

Difficulty of Automatic Transient Content

Alright, you guys are probably getting as tired of Tank Time/Call To Arms as me, so hopefully this will be the last post on the subject for a while.

Maybe this Call To Arms issue is showing us the upper limit of difficulty for transient content.

Maybe transient content can only be so difficult before we start seeing problems in the system, such as lack of critical roles. I know I complained that rifts in RIFT are nothing more than zergs, with no skill or strategy involved. But perhaps that's the way it has to be. It's entirely possible that if those pieces of transient content were too difficult, the system would break down. Arguably, that's what happened with Warhammer Online Public Quests, they got too difficult unless you had the numbers.

Perhaps Blizzard would be better off saying: here's easy content, and here's hard content. You get a Dungeon Finder for the easy content, but anything difficult requires you to build your own groups. Maybe difficult content needs to be reserved for extended forms of play, such as raiding, where you are involved with the same group over multiple days.

Or if the 5-mans are hard enough, it might would be better to leave them off the Dungeon Finder, signalling to the playerbase that you need to find your own groups, or that you should be doing them in-guild. But even long Dungeon Finder queues might be preferable to that alternative.

Again, a lot of qualifiers for this idea. It might be completely wrong, and we are still well below the "difficulty ceiling" for transient content. I think the real test of this proposition is coming in 4.1. By all accounts, the new troll instances are harder than the current heroics. So they will stress the automatic group creation system even more.

The Slow Death of 25-man Raiding

Unfortunately, because Blizzard decided to use the same achievements for both 10 and 25-man raiding, it's somewhat hard to disentangle exact statistics. But I believe there are some clear trends in what data we do have.

Take a look at these graphs, from wowtrack.org, which display where guilds are in progression. In particular, mouse-over the 25-man/10-man lines in the legend (upper-left corner) and watch the graph change.

You'll see that there's a clear difference in the shape of the demographics for 10s and 25s. 10s are shifted to the left and have a much larger proportion in the non-heroic area before sloping down. 25s, in contrast, display something close to a binomial distribution with slopes down to zero on both sides of the peak.

What this graph indicates to me is that no new 25-man guilds are forming!

Instead, the 25-man guilds appear to be the same 25-man guilds that existed at the end of the last expansion, and they are moving through the content more or less as a single wave. Note that the median 25-guild has one or two heroic modes already down, making it very likely that they are experienced guilds.

In contrast, 10s have a lot of guilds clustered on the left side of the high point, indicating that there are a lot of newer guilds there. And if you look at the numbers below, you see that there are about ten 10-man guilds for every 25-man guild.

And this makes sense, to a certain degree. Where would new 25-man guilds come from? A guild new to raiding would start raiding with their first 10 raiders. There's no need to move up to 25s. Maybe a large guild with multiple raid teams might merge some, but then you have issues where Team A is several bosses ahead of Team B, or too many raid leaders, or conflicting schedules or loot rules. Why break up three functioning teams to create one team which may or may not be successful?

But if no new 25-man raid teams are created, then the number of existing 25-man teams will slowly fall just through everyday attrition. In the normal scheme of things guilds die, and new guilds are born to replace them. That's what's happened all throughout WoW's past. But now, no new 25s are appearing to refresh the blood.

That's why I think 25-man raiding is doomed. Not because of any inherent negative quality in them, but because I just don't see where the new 25-man guilds are going to come from.

Given the path we're on, I would be surprised if 25-man raiding existed in the next expansion. The number of existing 25-man guilds will probably be less than half of what it is now, just through regular attrition.

Tanking: Power and Responsibility

We all know Uncle Ben's dictum: With great power comes great responsibility.

But what about the converse? Does great responsibility require great power?

In my previous post, I suggested that a Trinity game might be better off by sharing the responsibility of tanking between two people.

But perhaps we should go in the other direction and embrace the responsibility. For example, in the description of tanks, emphasize that they are the leader of the group. In a random, rather than having a separate leadership checkbox, have the tank automatically be the party leader in a random dungeon.

And give the tank's greater control over the group. As Dinaer suggests, give the tanks two votes in the vote-kick, with three votes kicking someone. Make it a little easier than the other roles for tanks to initiate vote-kicks.

Maybe there are other powers we could give the tank to help ensure a smooth run.

Having the responsibility without having the power to carry out that responsibility is a hard thing to bear. If we cannot do anything about the responsibility, maybe we can give them enough power to make it easier to shoulder that responsibility.

Flying Mounts

Sometimes I wonder if flying mounts were a good idea.

The thing about flying mounts is that flying mounts basically obsoleted geography. You fly up, then fly directly to your destination in the air. There's just air around you, a blank emptiness.

Whereas for ground mounts, where you are matters. Sometimes you cut across country, sometimes you follow the road. You're "in" the scenery.

Not to mention that when you see people on the roads, they're much closer to you. They're sharing the road. When flying, it's basically seeing someone pass you at a great distance. Another airplane, or ship on the horizon.

Flying mounts are cool, no doubt. But I think in a lot of ways, they started the trend of separating players from the game world. I wonder if the novelty of those mounts was really worth it.

Vampires

Apropos of nothing, I came across a F2P game today called Forsaken World. Looks like a pretty standard Eastern-ish MMO, but one of the classes available is Vampire. That sounded kind of interesting, so I took a look at it.

Apparently Vampires in this game are healers.

That just blew my mind. Vampire tanks, sure. Vampire DPS, definitely. Vampire healers...uh what?

Also, they seem to shape-shifters, which actually makes a lot of sense given the mythos, and is a pretty cool spin on the class.

And according to Melmoth at Killed in a Smiling Accident, the vampires barely wear clothes and wield giant crucifixes as weapons.

Shape-shifting vampire healers in skimpy outfits who wield giant crucifixes. The mind staggers at the design process which came up with that.

More Thoughts on Call To Arms

After reading comments and assorted blog posts about Call to Arms, here are some more thoughts.

Tanking isn't Quantum Mechanics

Gevlon thinks that tanking is too difficult. Difficult enough that offering an incentive won't improve things.

On the other hand, he's offering 1000 gold to everyone present at a guild-first kill of Nefarian, which seems to imply that you can motivate people to complete difficult tasks with rewards. On the scale of things, killing Nef is much more difficult than tanking.

First off, tanking isn't that hard. It looks hard at first, but practice improves everything. Personally, I think that if more people step up and try to tank, more people will find that it really isn't as bad as they thought it would be.

(Except for Ozruk. You're on your own there.)

Second, the reward is also aimed at experienced tanks who have finished gearing up in heroics. If the reward pushes those tanks to do just two daily heroics instead of one, it's a major win.

Tanking Shortage Is Not Caused By Raid Slots

I used to think that part of the reason tanks are rare is because 25-man raids require fewer than 20% tanks.

But the truth is that 25-man raids are dying (a topic for another time), and the vast majority of raiders are in 10-mans. And 10-mans have the same 20% tank ratio as a 5-man group.

So I'm not really sure this reason is valid anymore, if it ever was.

Social Cohesion

The thing about PvE in WoW is that it tends to be very egalitarian once you're in a group. There's a sense that everyone is part of the group, and everyone is contributing more or less equally, and thus everyone has more or less the same shot at the rewards. This may not be strictly true, but that's how the default loot systems work. Even Need before Greed is strictly based on armor-type, rather than any performance-based quality.

This Call To Arms explicitly breaks that illusion. Now, though the group completes the task together, one member of the group gets singled out for an extra reward. And going against that egalitarian grain often rubs people wrong.

For example, a lot of the high-end guilds give loot priority to their tanks and healers. This is deliberately unfair to the DPS, but the DPS accept it because it helps the group progress faster. The DPS sacrifice for the group, and that can restore the cohesion.

But in Call to Arms, there is no external group for the DPS to invest in. I think this will be the biggest challenge for Call to Arms. Tanks often feel "abused" by the dps and healers. Will it get worse if the game deliberately sets them apart and gives them extra rewards?

Envy is a deadly sin, but it still causes damage.

The Value of Experimentation

I may have mentioned this before, but I think our society is becoming too risk-averse in many ways. What's wrong with running experiments, with trying something new?

Honestly, if Blizzard had floated the idea of the Dungeon Finder before it was available, I think everyone would have enumerated all the things that could possibly go wrong and insisted that the Dungeon Finder would be a failure.

And a lot of those negative effects did happen. But they were greatly outweighed by the benefits, and the most negative behaviors corrected where they could be.

Doing something, seeing what happens, and fixing the problems as they emerge is far more likely to produce advances than trying to theoretically construct the "perfect" system beforehand.

Carrying a Tank

Tobold asks a question: How do you Carry a Tank?

It's pretty rare, but I've seen a group carry a tank before.

I define "carrying" as when a character performs significantly below normal, to such an extent that it requires other characters to perform significantly above normal in order for the group to be successful.

I was once put into a Dungeon Finder group with three DPS and a bear tank, all from the same guild. The bear tank had a 150k health, so I thought all was good.

As we progressed through the instance I noticed two things. First, the DPS was very good, all pushing 10k. Second, the bear was taking a ton of damage. I was having to chain Divine Lights and pop cooldowns to keep him up. I found this amazingly odd, given that the tank came from the same guild as three top-notch DPS.

Mid-way through the run, I inspected the tank. The tank was in full PvP gear. So he had a lot of health, but was missing all the avoidance and mastery that would have significantly reduced his damage taken. He was also missing the talent Natural Reaction.

The tank had enough health so that he wasn't one-shot, but he took damage at a rate that required me to burn mana at a much higher rate than normal. However, the DPS performed at such a high level that they were able to end the fights before I ran out of mana, in addition to requiring the barest minimum of healing themselves.

It's fairly obvious that the group decided to have their druid respec and wear PvP gear, and let the DPS carry him through the instance, rather than face 45 minute queue times. I think the Natural Reaction thing was just a mistake, as the druid spec looked good otherwise.

The point is that because the tank performed at such a low level, the DPS had to perform at a much higher level so that the run was successful. The performance of the DPS effectively carried the performance of the tank in that run.

The moral of the story is that very high DPS forgives a lot of sins.

Solving the Tank Problem

Blizzard unveiled a new system to bribe incentivize tanks to run random heroics.

Everyone else will probably be commenting on it in greater detail. My take is that it is a reasonable experiment. Maybe it will work, maybe it won't. Also, the single best comment on this system comes from Martinel, posted on WoW Insider:

Well. Rogues, mages, warlocks, and hunters...being pure DPS classes, they won't really be eligible for this, will they?

I guess you could call this a hybrid tax...

*sunglasses*

..refund.

But seriously, how would you solve the tank problem for real? The truth is that, in every trinity MMO I've played, save one, the tank has been the limiting factor.

Maybe it would be instructive to look at the one game that was different: Age of Conan. To me, Age of Conan tanking seemed pretty hard, so it was quite surprising that you could find tanks.

After ruminating on it, I've come to the conclusion that the key difference was that Age of Conan required two tanks. At first this seems counter-intuitive, how can requiring more tanks for a harder job make it easier to find tanks?

But players have always cited the responsibility of tanking as a key factor in keeping people from tanking. Perhaps people are more willing to tank if they know that the responsibility will be shared, that there will be another tank in the group to help you.

And it's not really that hard to have another tank in the mix. You tank X, I tank Skull. You get the boss, I get the adds. I have better gear, so I'll tank the boss and you can watch and learn from me and do a little DPS. It's easier in instances than raids, because it's really not worth wasting time swapping out a tank for a single fight if you will just need to get them back again.

Healing could also benefit from having two healers, depending on how you heal. If healing is a little more passive, so the healers aren't constantly stepping on each others toes. I'm less sure of this, because AoC required two healers, but healers were pretty rare.

So maybe the proper group would be 2 tanks, 2 healers, and 2 DPS. Or maybe instead of DPS, have 2 Crowd Control, and have all three roles put out roughly equivalent damage. And design the class system to match that ratio, rather than having 50%+ DPS classes/specs.

Intellect Plate

Ghostcrawler, in his post on Stats on Gear, talks a bit about holy paladins:

I mentioned two out of three paladin specs above. That third spec has been a thorn in our side for a long time. Paladins in general have a knack for that. I kid. (Mostly.)

Heh. I'm pretty sure we're all down with this. Ghostcrawler goes on to say:

I'm talking about Intellect plate. We don't like Intellect plate, but we haven’t come up with a good alternative for it. The pitches we hear most often have downsides we don't like. Yes, a Holy paladin could wear mail… at which point their silhouette would look a lot like a shaman instead of a paladin. Yes, Holy paladins could derive their spell power from Strength… at which point they also hit with their weapons nearly as hard as Ret paladins. Yes, we could convert all that Strength to spell power, and convert the Hit to Spirit and the Expertise to Mastery or whatever. The Spirit to Hit conversion does the job, but it's not super elegant or intuitive. I’m not sure we want to pile more onto that design.

I like Intellect plate. I'm really happy that the devs have refrained from expecting Holy paladins to wear mail. To me, it's one of the iconic elements of the class. Paladins wear plate. I played through the times where Holy Paladins would dress in cloth, and they were terrible.

As for wearing Strength plate, in my mind the real problem with that solution is the non-armor item slots like rings and weapons. Right now, holy paladins have their own armor and share the non-armor slots with the other healers. If we were expected to wear Strength plate, would we also wear Strength rings and use DPS weapons? Or would we wear Strength plate but Intellect rings? Or would we become the new hunters, and simply roll on everything.

It sounds a loot more complicated than the current system. The current system is simple. You're a paladin so you wear plate. You're a healer so you go for Intellect and Spirit. That's pretty much all you need to know to get a good start gearing your Holy paladin.

Intellect plate isn't ideal. But I think it's the most workable solution. Between the larger amounts of loot dropping from bosses, tier tokens, and vendor items, I think that having Intellect plate isn't too much of an inconvenience.

Now, if only Blizz would make a two-handed healing mace with Intellect and Spirit.

Progress Updates

My guild is chugging along, moving into heroics. We're currently 4/13 with Halfus, Magmaw, Atramedes, and Chimeron dead. We're working on Maloriak at the moment. Of those listed, I've seen all but Magmaw.

The heroic fights are okay. They're like ICC heroics, just the normal versions with 1 extra mechanic and the damage dealt and dps required dialed up.

Truth be told, when WoW is over, I think we'll look back at Ulduar as the high point of raiding over the game's lifetime.

Loot-wise, I think Blizzard needs to pay a little more attention to where stuff drops. There are five different epic plate healing belts, including one from reputation and one from crafting. Yet there is only one epic plate healing bracer, and it drops from an end-boss.

Or healing weapons. Putting the two healing weapons paladins can use on the end-bosses was not a good choice. I think some care should be taken that end-bosses don't share item slots or types in their loot tables.

(I guess you can tell which two items I'm missing to finish i359.)

It would have worked if people cleared Blackwing Descent before starting on Bastion of Twilight. But sadly, people don't approach bosses in that manner when multiple raids are available. They bounce around in order of difficulty, and you can almost guarantee that all the end-bosses will be killed at the very end.

Learning Programming Using WoW

Ian Lamont sent me the following:

Thought you might be interested in an MIT class project: Developing a basic computer science curriculum around Lua scripting in WoW. We published it earlier this month at the following location:

http://www.ilamont.com/2012/02/teaching-computer-science-through-wow.html

Now, I'm a software dev by trade, so this was pretty interesting. Games have traditionally been a path into programming, and WoW does have some interesting features for aspiring devs.

One thing about WoW is that it is an "information-rich" environment. There's all sorts of API calls you can make to get information about different parts of the game world. Then you can easily change things, and make the same calls and get different data.

In the link above, this is illustrated with simple calls like "IsOutdoors()". You can move your character in and out of buildings to get different results.

That's something which is actually somewhat hard to set up from scratch at the beginner level. You have to set up a datasource, and then connect to it, then learn how to query it and interpret the results.

As well, it's easy to move into simple, useful mods in WoW. Sometimes it seems like it is pretty hard to make a simple app these days. Making a small mod is a lower bar.

On the other hand, WoW does have one huge philosophical negative as an environment to learn to program in. Programming is more than just processing data and displaying it. The other half of programming is automation. Not just slicing and dicing information, but "doing" multiple actions.

And here WoW goes to great lengths to prevent the user-programmer from being able to automate actions. You'd have to make clear the idea of what is theoretically possible to automate, like moving your character in a pattern, but is prevented by the game company.

Either way, it's an interesting idea. Learning to program can be challenging at first, because it often seems like there's this entire framework that needs to be set up before you can accomplish anything. But in WoW, that framework already exists and has useful data. And mods and the mod community provides a ready path for the new programmer to make small apps that may even be useful to other people.

But in the long run, I wonder how a programmer who doesn't grasp the concept of automation is going to fare.

Ask Mr. Robot

Here's a nice website I used the other day: Ask Mr. Robot

It allows you to load your character from the Armory, and then you hit an Optimize button and it tells you the best way to gem, enchant, and most importantly, reforge your gear. There's options to not use Maelstrom Crystals, etc.

I tried it out for my Retribution spec, because it was a big mess. Then I executed all the suggestions for reforging and regemming. Surprising, I could actually feel the difference when I went to do dailies in Tol Barad. Mobs were dying faster.

It managed to pretty much nail the hit cap with 8.01% hit (cap is 8.00%).

The defaults look pretty decent. They're taken from theorycraft sites like Elitist Jerks. But there's also the option to change the default stat weights to whatever you prefer.

So, Ask Mr. Robot. Very easy to use, and a very nice site to help optimize your character.

Alternative Holy Paladin Masteries?

Continuing on from the last post, let's say that this whole shield mastery just didn't work out. What would be an interesting ability to replace it?

So far, Masteries should not apply all the time, and they should be somewhat unique. So what's interesting about holy paladins? In the past we've been known for our single-target healing, but directly increasing those is pretty boring. We have Hand spells, but they're used too infrequently to really be a useful mastery.

Beacon of Light is another iconic spell. But the last time Beacon got too strong, it got overpowered, and ended up pigeon-holing us. As well, it's actually possible that a Holy paladin might not have Beacon in her spec. It's unlikely, but possible. Having a mastery tie into a technically optional ability is probably not the best of ideas.

My thought is that there are four possibilities.

1. Judgement

Judgement is a signature paladin spell. It's also something that we cast a lot. (You should be Judging on cooldown for the mana return.) So maybe something that triggers off Judgment would be interesting. Some ideas:

"After casting Judgement, the effectiveness of your next 3 healing spells is increased by X%."
"Casting Judgement causes 5 players with 5 yards of the target to be healed for X% of the damage done."
"Casting Judgement places a debuff on the target that lasts for 15 seconds. The effectiveness of your heals on the target of the debuffed target is increased by X%."

(The third is a bit awkward, and possibly overpowered, but it's a nod to our tank healing days.)

2. The Holy Power system

The Holy Power system is fairly unique among healers. Playing off this might be interesting:

"Increases the effectiveness of Word of Glory and Light of Dawn by X%."
"Your direct heals have a X% chance to generate a point of Holy Power."

The problem with the Holy Power system is that too much HP generation causes us to have excess mana, and that always leads to trouble.

3. Criticals

In the past, paladins have always cared about critical heals, mostly because of Illumination. Tying crits to mana is a bad idea, but there are other options:

"Increases the bonus healing done by critical heals by X%."
"Your critical heals heal the target for an additional X% over 6 seconds."

This might even make critical strike rating a more attractive stat.

4. Instant spells

This is new to Cataclysm, but we actually have a significant chunk of instant spells: Holy Shock, Holy Radiance, Word of Glory, and Light of Dawn. Maybe:

"Increases the effectiveness of your instant cast healing and damage spells by X%."

(Damage spells just for fun and flavor.)

The problem with this would be PvP as instants are very powerful there.

Conclusion

That's just some brainstorming over possible replacement Masteries for Holy Paladins. Any other ideas?