Everything You Always Wanted to Know About Warriors But Were (Not So) Afraid to Ask

Blizzard again rises the bar and sets a new standard.

The warrior protest in World of Warcraft produced some results. I documented the protest itself here, then gathered more comments here.

For me the protest was more a crusade about the communication than getting involved in the problems of the warrior class in the game. I’m really happy and pleased by how Blizzard reacted. Kalgan, one of the devs that engaged some PvP debates during beta, wrote down an incredibly exhaustive, deep and informative comment to the issues that the playerbase underlined during these months.

The only ‘bad’ element is that this post arrives only after the protest. I believe that engaging the community in a discussion is extremely useful. For both the parts involved. I’m happy to see that Blizzard did a relevant step in the right direction and I hope now it won’t stop or feel satisfied.

With this entry I re-read (with my broken english) what has been written, summarizing and underlining those parts that I consider relevant.


Kalgan starts explaining that the rage-generation of 1h+shield is intended to behave differently than 2h. This because the rage builds up as you deal damage AND when you are hit. In the case of 1h+shield you are supposed to have more aggro on you and “hold” more opponents at the same time. So the rage-generation is compensated because while you deal overall less damage you are also being hit a lot more frequently. The issue then become how you are able to take the aggro. He explains that you can do this by switching stances and with the help of the Tactical Mastery talent and the Bloodrage skill.

Then he passes to discuss and analyze why warriors cannot be directly compared with rogues. This is obvious since, you know, classes are supposed to play differently in this game. Built on completely different mechanics. This was a non issue for me but it needed to be explained because it’s a very common assumption of the players when they compare directly the classes. So there’s not much to argue or discuss here, aside the fact that this was absolutely needed on the “communication level”.

Direct quotes:

If rage and energy were generated at equivalent rates, it wouldn’t have been particularly meaningful for them to be different mechanics. If we intended for them to be the same, we probably would have chosen to make warriors use energy just like rogues do, yet all this would have done is water down the core differences between the character classes.

Finally, if when it was all said and done warriors dealt as much damage as rogues (due to similar rage-energy translation into damage and similar power generation rates), warriors would be flat-out better characters then rogues (better armor, more hit points, same damage), so again, it simply isn’t reasonable to draw direct comparisons between the two.

All of that having been said, we agree that there are some issues with rage generation in general, and we plan to address these so that warriors dish out some more damage and push slightly more buttons than they do today.

Then he continues to explain how 1h+shield, 2h and dual wield are considered and balanced. To begin with he explains how 2h was balanced against 1h+shield. To do this they calculated extensively and exhaustively the overall time to kill 100 creatures. So including the downtime. 1h+shield is supposed to kill one single creature more slowly, but with less downtime between each fight. While 2h is supposed to kill more quickly but taking more damage, so more downtime. These tests brought directly to a rule:

2h weapons needed to produce approximately 20% more DPS then 1h+s. What this meant was that in order for DW to be an interesting (but not overpowered) choice, DW needed to also produce a total increase of 20% DPS over 1h+s (just like 2h does).

Given that, we needed to find a way to modify some combination of damage, hit chance, attack speed, or other basic combat factors in order to achieve 20% rather than 100%.

The miss rate, one of the most debated issue, is also considered intended. Balanced between 1h+shield, 2h and DW with the formula here above and with an overall ratio that is balanced to offer a “reasonable amount of unpredictability”.

This is instead what ISN’T working and will be addressed (soonâ„¢):

Now, regarding improving the warrior’s overall rage generation. We’ve found that there are indeed a few issues with rage generation as a result of bugs. Warriors (and druids in bear form) are intended to gain some rage on block,dodge,parry events, but apparently this isn’t working correctly. This needs to be fixed asap in order for warriors to meet their total expected rage generation.

In addition, while our tests have indicated that players are getting the intended 5% miss rate while using a single max-skill weapon against even level enemies (yes, we’ve tested this exhaustively, and cannot reproduce the 9% miss rate you’re reporting for normal swings), we have found the bug causing abilities (ie: heroic strike, sunder armor, etc) to miss around 9-10% of the time under the above conditions instead of the intended 5%. This does mean that warriors are getting the “miss” even about twice as often as they should. Obviously, this needs to be rectified asap, since it directly impacts the warriors DPS and rage use (since missing wastes 20% of the rage cost of the ability).

The next topic is the rage potions. Recently commenting this I wrote: “The design decision is to not make the game a potion fest. This destroys the game when you HAVE to go around with a hundreds of potions because you know the standard is about to use them ALL the time. Potions are ok if they are odds. To be odds they need to be limited or “odd” becomes just “catass”, where this catass can just own everything everywhere because he has super loot and potions to use all the time.”
Kalgan explains my point of view:

We understand that Rage potions being on the same timer as other combat-timer potions means they don’t amount to “free” rage. If rage potions were on a separate timer, we’d actually have to balance the game around the assumption that rage potions are used in a significant number of fights. Therefore, rage potions aren’t intended or expected to be what provides a baseline amount of rage for a fight.

The problem of the “hitpoints” is another hot topic in the discussion. Again here the damage comes directly from the experience of the players with other games and the assumptions that the same mechanics need to work in the same way in World of Warcraft.

You’re claiming an imbalance exists is because a warrior that focuses on getting gear with stats other than stamina gear can end up having a lower total health then another character that focuses on stamina equipment, I still don’t see the argument as entirely valid. That scenario seems more like a conscious choice on the part of the players involved, and evidence that WoW has interesting and meaningful equipment choices.

With respect to the implication that warriors should have more base health, our combat durations, damage values, ability/spell statistics, etc are all balanced around a warriors having their current health values. Changing those health values yet retaining similar combat times, downtime, etc for the game as a whole would require a huge number of changes across the game, impacting the balance of the game in general.

The problem of spells dealing damage to a warrior without a protection value coming from the armor factor is considered intended but again felt broken by the players because of assumptions coming from playing other games. They know that the spell damage isn’t reduced by the armor factor and the damage of spells (compared to warrior’s styles) is balanced keeping this key mechanic in mind.

Therefore, a warrior’s base melee attacks are deliberately “overpowered” with the expectation that the targets will have some amount of physical damage reduction due to armor.

The next problem is the “survivability”. In particular when warriors are compared to Paladins. Basically what Kalgan writes is that this issue will be partially solved with a paladin nerf more than a warrior boost:

While paladins are indeed intended to be the most survivable of any class in the game (players sometimes assume warriors are, which is not the case), paladins aren’t intended to make the best “tanks� in the game, or to have superior offensive abilities then warriors. What this means is that survivability alone does not a tank make.

The other key concept to tanking is the ability to hold agro on mobs (especially simultaneously on multiple mobs). This is the key difference between warriors and paladins in PvE. Simply put, warriors are intended to be the best at holding agro against multiple mobs (which will also be a more clear distinction once the Seal of the Crusader bug is fixed). So, assuming a group scenario that includes healers, a warrior’s survivability rivals that of a paladin’s, with the warrior having very significant advantages in terms of holding agro.

A quicknote is about the use of bandages compared with food:

Bandages heal considerably more quickly then food.

The rest of what Kalgan writes is simply about dissipating more and more wrong assumptions. Again this is needed and it’s an OPTIMAL way to communicate with the playerbase. It allows the players to UNDERSTAND how the problems are perceived by the devs and allows them to finally understand, accept and SHARE these design choices.

I’m more than sure that the time Kalgan ‘wasted’ to write all this will be rewarded. The communication isn’t simply important to keep the playerbase calm. It is important because the community itself is the biggest and more precious resource for a dedicated dev team. This message, that I read and analyzed, definitely improved my faith and trust into Blizzard. I’m sure that many other players feel the same way.

To conclude this is a quick list of what will hopefully change in the near future:

Cumulative List of Fixes:

1- We’ve found that there are indeed a few issues with rage generation as a result of bugs. Warriors (and druids in bear form) are intended to gain some rage on block, dodge, parry events, but apparently this isn’t working correctly. This needs to be fixed asap in order for warriors to meet their total expected rage generation.

2- We have found the bug causing abilities (ie: heroic strike, sunder armor, etc) to miss around 9-10% of the time under the above conditions instead of the intended 5%. This does mean that warriors are getting the “miss” even about twice as often as they should. Obviously, this needs to be rectified asap, since it directly impacts the warriors DPS and rage use

3- We agree that Bloodrage is too punishing right now. We plan to improve Bloodrage so warriors don’t feel as hesitant to use it.

4- (about the Improved Thunder Clap talent) This talent needs some love.

5- (about Improved Execute) The full consumption of rage upon missing is a bug that will be fixed.

6- (about Bloodthirst) I agree that this talent can use some improvement. It turns out the current design leaves it vulnerable to always being either too narrow or too powerful, without much in-between. As such, this talent is a likely candidate for change.

And a possible nerf in the future:

– (about Sweeping Strikes + Cleave or Retaliation) In truth, it was never really the intent that sweeping strikes would work on multi-target abilities (it is intended to turn single-target abilities into dual-target abilities). However, in the interest of not weakening warriors right now, we’re inclined to allow the combination to work as long as it doesn’t become highly abusive and create balance problems.

I archived permanently the original message here on this site in the forum. So that it won’t vanish as it always happen on Blizzard’s forums.

Posted in: Uncategorized | Tagged:

Leave a Reply