WoW Lessons Learned

Ion sat with PC gamer to talk about lessons learned from BfA, and what’s coming up.

The big take aways for me are thus:

  • The devs thought the fact that Legion artifacts were front loaded in terms of skills, then only minor increments over time was some something players disliked.  That was a mistake.  (Extrapolating here – but the enemy of legion was RNG, which BfA went all in for.)
  • Azerite gear is a large enough misstep that they are putting in a new system.  Instead of unlocking skills on gear, you unlock it on the neck (permanent item).  So they learned that they should not take stuff away from people while leveling.  Odd that was even a lesson to learn in the first place.
  • 8.2 will be akin to 7.3, in terms of horizontal features.  I liked the content in Argus, it was just unfortunate that it made everything before that completely irrelevant.  Also begs the question if this is the last patch of the expansion.  That would be pure folly, due to the next point.
  • The lack of testing and feedback collection in beta sent them down the wrong path.  While I can understand that Azerite gear was very nebulous up until the last month or so, it does bear mention that all the issues were fairly clearly stated in the beta forums before launch.  Pretty much every complaint from then grew into a wildfire.  That’s like being put between a rock and a hard place, you have a set release date and a you know a key component cannot be fixed in time.
  • If Blizz really wants longer lead time on system development, there’s no way 8.2 is the last patch… and I’d bet that there are 2 more to come.
  • Blizz has also come to realize that making decisions now to avoid an issue in 3 years isn’t practical.  Either that issue never really existed, the market will shift and it won’t matter, or the decision causes major negative feedback that you lose players.
  • Related, there are some pieces of the game that just should not be changed for the sake of change.  (I mentioned Pet Battles are relatively untouched).


All in all a relatively good read, and a rare occurrence of a dev admitting that their design decisions were poorly received.  I’m generally curious as to how 8.2 will be received.  A lot of the pain points from launch have been, or will be addressed.  There are some clear lessons learned from the devs, as the feedback on this particular expansion has been some of the most vocal I’ve ever seen.

Quite important to note that when the core of the content is based on a particular foundation, and that foundation is considered a flaw by the players… it is a whole world of pain to make the necessary changes without causing everything else to fall apart.  We’re about 9 months out since launch, in terms of dev work, that’s a pretty quick turnaround for such big changes.

Here’s hoping it sticks the landing.

WoW Sub Numbers

It has been many a year since WoW published subscription numbers – pretty much this time in 2015.  Since then we’ve had nothing but speculation, mostly from 3rd party sites.  It would be fair to say that the general trend has been downwards.  This is entirely subjective, based on the number of people present in any given area – or simply the number of large scale world quests available.  Dips and spikes.

WoW did let us know how many copies of BfA sold initially – 3.4 million.  The wording here is a bit suspect, as it’s unrealistic that this would be the total number of sales on day one so much as by day one.  It would be fair to argue that this number would encompass both those who were actively playing in Legion, those who stayed after the free weekend in July, and those curious about the traditional WoW expansion fever.  I was in the 2nd category.  I am certain there is a long tail when it comes to expansion purchases, but a tiny fraction of those on “day 1”.

A recent tweet from the makers of WeakAuras intimates that the subscription numbers are a tad different.

It’s an interesting bit of “news” in that it can’t really be substantiated, right?  Does it align with subjective viewpoints?  Sure.  Is it mathematically accurate?  Not so sure about that.  Is it possible Blizz exposed data that it shouldn’t have?  Yeah, 100% on that front.


I barely squeezed out a month out this expansion.  But that’s my experience.  Plenty of folk still having fun.  And without substantiated numbers from Blizz, human nature is to always trend towards the less pleasant of all rumors.    Doubtful that will make a difference though – they are still making money hand over fist.

WoW News

I guess Blizzcon is 90% Diablo content now.

Two bits of interesting news.  Mike Morhaine is stepping down from Blizzard leadership.  No reason given, but after 27 years of doing something, I’d be tired of it too.  I consider Mike part of the old guard, like the doctors from BioWare.  More passionate about the games than the market.  Instead of oozing charm, he was full of geek.  I thought that resonated well.

J. Allen Brack is promoted to his position.  I didn’t do much digging, but there’s an interesting video with J. Allen, Ion, and Tom.  I just think their personalities are really quite focused in this particular video.  As a people watcher, I find this quite entertaining.  Not from what they are saying, but how they are saying it.


Azerite Changes

Somewhat large changes afoot.

  • More drops:  You’ll eventually get up to 370 Azerite gear from world quests.  A quite inelegant solution to the problem, but it certainly fixes it.
  • More traits: A new outer ring on some gear in 8.1.  I don’t understand this one, personally.  It seems like it makes the situation worse, not better.  The problem now is that there are too many traits, without interesting choices.
  • Trait tuning: To make things more interesting.  Without actual examples, this feels more like number tuning.

By adding more traits, that’s further diluting the pool of available gear.  By adding Azerite gear drops to WQs, it goes completely against the design choice to target azerite gear drops for their specific traits.  WQ rewards are entirely random.  Combined, this appears to be making the problem worse.  Practical examples will help straighten this out.

Interesting vs Meaningful

This post as a base.  It’s a decent summary of the Azerite issues, with no potential solutions.  Which is smart in a sense, since people will focus on the problem, rather than the solution.  Lore responded.  One particular item:

The point about traits being “useless and uninteresting” is interesting considering that you also make the point of “every gear change requires simming.” These two points are kind of at odds with each other. The way to solve the simming issue would be to make the traits more simplistic in nature. Similarly, making traits with more outside-the-box designs leads to more complicated questions of “is this better or not,” which in turn encourages more simming. Either way, it’s an interesting challenge, and one we’re taking to mind as we move forward with traits in future updates.

Useful things are not always interesting.  A toilet is useful, but it is far from interesting.  A Ferrari is interesting, but far from useful.  A robot that vacuums the floor is both useful and interesting.  They are not binary, or in conflict.  It’s like saying something is red and big.  They are simply descriptors.

Azerite traits are generally uninteresting, because they have no impact on gameplay.  No matter the trait you have, the buttons you press stay the same, 99% of the time.  They are generally useful since they do apply a +damage/healing effect.  Some are much less useful than others (getting +haste on my Brewmaster feels bad man).

Azerite was meant to replace artifacts, tier sets, and legendary items.  All 3 of them had interesting impacts to gameplay.  So much so, that the majority of Shaman Ele changes in 8.1 are cut and paste from that model.

That the set of traits are so poorly worded that they require simming is a different point altogether.  Passive damage boosts require simming, and it’s actually practical to do so.  An interesting trait that changes your rotation… that’s something much harder to sim and compare.  It’s also a whole pile much harder to balance.  Pretty much why they were always restricted in the past.

I do get what is trying to be achieved here.  Simplification is every IT person’s goal.  It’s extremely hard to do.  I don’t quite understand what the changes above will actually do to fix that issue.  At least they are trying.



Testing. It’s not Hard

There’s a point where this becomes petty/funny/sad.


The item on the left is a normal drop.  The item on the right is a Warforged version, which should be an upgrade but is in fact worse that a Normal BfA dungeon drop.



The items for sale on the Timewalking Vendor have not had their ilvl updated, but do need a 120 player.

These are item specific – not related to how the stat squish has impacted the overall balance.  End Times can’t seem to be completed right now, for example, as Sylvanas will 1 shot most tanks.

All of this was corrected in a hotfix on the 25th.

Timewalking is something that comes up every few months.  It opens old dungeons from a previous expansion, and applies scaling so that max level players can go back into them.  Completing these dungeons awards drops as well as tokens.  Those tokens can be exchanged for items that were previously on-part with heroic dungeons, or cosmetic gear, or heirlooms, or mounts.

It’s a neat side activity. This is the way most dungeons work in FF14, so it’s not like the tech is new or hard to do.  Blizz has done this a few times in the past.  An entirely different topic as to why this system is always available…

Bug Hunting

IT has bugs.  That’s life.  Some are obvious, some are not.  Some are easy to fix, some are not.  Fixing them is one conversation that I don’t think I need to get into.  This is about finding them.

The 3 examples put above would have been obvious from a single dungeon run, through every Timewalking dungeon.  Guaranteed, 100% reproduction.  That these were pushed into production means that either a) it was not tested or b) it was tested and the project manager accepted the test results.

If this happened in my team, this would merit disciplinary action.  After 3 times, they would either be dismissed or re-assigned to other work.


It’s a game.  It’s made to be relaxing for some, challenging for others, fair for all.  It’s just symptomatic of BfA in general.  Which is sad – and not what we should expect from Blizzard.  I should be focusing on the fun items, not these brain fart items that keep popping up.  I’m missing the polish.

And I can only fathom how Blizzard employees are feeling about this.  These items deflate overall confidence… and it takes a long time to build that back up.  It also causes finger pointing, and niche work.  No good ever comes from that.  Again, in most industry the lead of these teams gets replaced so that it at least appears something is being done to rectify the situation.

Story Telling: Branch vs Root

I’ve completed the Horde storyline for BfA.  It is quite well done, and I’d argue that it’s a better one that the Alliance version.  Mainly because it operates under a branch model, while the Alliance is a Root model.  Let’s explain this a bit.

A branch story starts from a given point and then expands to multiple lines that develop the plot and characters.  There’s a large arc, but each zone has it’s own flavor.  It’s the most common type of story, and it allows you to see the various pieces in play at almost any given time.

A root story goes the other way.  It’s a bunch of different pieces that eventually meet up in the end.  Most crime fighting novels go this route (Holmes & Poirot are very good examples).  Things do not seem at all connected until, poof, there you go.  It takes a really good writer to pull this off. Stephen King’s The Dark Tower series is a really good example.  Lost (the TV show) is an example of how that can go off the rails.  A very solid plan is required to make this work, so that the reader doesn’t feel cheated.  They can go back, look at the bits and come to the same conclusion.  (e.g. watching Memento is fun the 2nd time).

So that leaves the Horde with an expansion story, with tons of troll lore and very few Horde bits.  The Loa, Blood Trolls, and G’Huun main arcs all work well against each other, and it’s honestly a lot of fun to dig into those zones.  Mechanically there are only so many blood trolls you can kill before moving on.  And there’s next to no interactions with Sylvanas that I’ve seen so far.  Both good since the Troll storyline is really awesome, but also bad, since this entire war started because of her.

The Alliance instead gets 3 near completely distinct storylines that never truly intersect past the Drust King Gorak Thul events on the War Campaign to find Jaina.  The events in Stormsong really don’t have much impact at all on the main line.. but it does set up the Old Gods.  Drustvar… sure, a bit.  Tiragarde is golden.

Capital Cities

Ziggurats are cool to look at.  They are less cool to explore.  The Alliance “center” is relatively small and easily traversable.  Sure, there are memes about the rope ladder on the boat, and a stupidly placed fence, but it still is easy to move around.

Horde need to take a flight path from their main center of operations to their boat.  A flight path.  And then jump down and run 3x as far as any Alliance player would need to move.  It is a beautiful city that is a major pain to move around in.  So I guess like a real city?

Other Isle

I think the Alliance get the better deal here in terms of movement on the other isle and placement of activities.  Both start off with 3 spots near the coasts, and after hitting the proper reputation level, can unlock more inland flight points.  It’s that the Alliance islands have WQ that are placed more inland, or at least over much more difficult terrain than the Horde islands.  Stormsong is quite bad in terms of initial Horde bases and where the WQ spawn, with Tiragarde not too far behind.  Drustvar has nearly everything in the top left (even as Alliance I won’t do the dark forest items).

Vol’dun is passable as it’s pretty flat (cept the north).  Nazmir has most activities in the North East.  Zuldazaar is ok, if you avoid any quest in the middle of the map.  There’s a lot of vertical to do if you do end up trying.


I normally play Stormrage, which is 95% Alliance and just massive in terms of overall players (> 900,000).  I made a Horde Demon Hunter on Uldum, where the split is closer, but there are still only 50,000 Horde players.

(Side note.  Demon Hunter from 100-110 was Legion Invasions for 1.5 levels in 15 minutes of effort.  110-120 was with Azeroth Auto-Pilot to get me to the next main quest and some optional ones.  Still read the text and cut-scenes, which was good.  No dungeons or Island Expeditions.  It was painful until 111, when I had a full set of BfA green gear.  When I dinged 120, I had 26,000g and had yet to use the Auction House once.)

Three noticeable impacts of Uldum

  • The Auction House is much slower.  Prices are higher, but many items just won’t sell.
  • No guild invite spam!  I get one every 2 minutes on Stormrage.  Saw 1 total on Uldum.
  • The world feels empty.  I’d have 6-7 groups running for World Quests, now there are times when I don’t see any.  When I do meet people, it’s nice to see a split between factions.  Stormrage… I’d see one Horde per day, from another server.

Alternative Measures

I like to look at all sides during a debate.  Rarely is someone ever 100% correct.

WoW Tokens.  The interesting mechanic that Blizz put in place a long while ago so that people in need of gold could buy it with real money.  Sure, it helped get rid of a lot of the black market gold (not all, but most) but it also served as an interesting measuring stick for the game.

On US servers tokens were around 215,000g in the early summer.  It has been at that level for most of Legion.  Good enough.

When BfA launched, it took a massive nosedive.  Down to  a low of 100,000g.  That lasted exactly one month to the patch cycle.  It is now moving back up, currently at 140,000g.


Which is interesting on many fronts.

I have a passion for numbers, and even more so for the context behind those numbers.

BfA Launch

Prices on items dip for one sole reason – supply outstrips demand.  There were more people fishing for gold than there were people who wanted to buy tokens for time.  Which makes sense, since expansions typically have a spike in players and they are not necessarily buying tokens to get time.

I sold a staff for just 800,000g within a few days of launch.  Money was flying all over the place.  It was relatively easy for people to swap their gold stores for a month token at launch.  There’s a pile of analytics to be seen on this.

Interestingly, the price has started to spike on the 1 month anniversary of BfA.  Why?  Quite a few reasons I’m sure. My bet is that people have stopped buying tokens for cash, meaning that the supply is drying up, in addition to less people actually buying tokens.  I’d wager that people are figuring that with LFG/Warfronts out, they really don’t have any need for gold now.

It’s still quite early to see what this actually entails, we’ll need to see at the 2 month date.

Fishing for Cash

Blizzard is offering the Dreadwake mount for free for anyone that buys 180 days (6 months) of game time.  Offered the day after the livestream of 8.1, and at the 1 month mark.  It does look neat, and it’s not a horse.

Cynicism active. This reminds me a lot of the 1 year sub offer in Cataclysm that offered guaranteed access to MoP beta and Diablo 3.  That is not a good mental link.

Patch 8.1

More info from MMO-Champ.

High Level

  • Azerite trait rebalancing
  • A new Warfront
  • 2 new Island Expeditions
  • War Campaign continues
  • 2 new raids, on in Zuldazar and another below the Shrine of Storms
  • New Pet Battle dungeon
  • Kul Tiran and Zandalari Trolls are coming, requiring Exalted with Proudmoore / Zandalari factions
  • Heritage armor for Blood Elves and Dwarves
  • Champions of Azeroth rep levels will be account bound, making the +15 levels on the neck not requiring the grind

The Good

More content is good, can’t deny that.  Another Warfront on a separate timer will mean that people won’t twiddle their thumbs for weeks waiting.  More raids, and faction bound are an interesting twist.  A re-focus on the actual war between factions would be nice to see, since it’s been all but invisible so far.

The CoA faction change is good.  Heritage armor is nice to see.  The two new Allied Races added with a grind on existing factions is very nice, rather than a brand new one.  I still think that it should be baseline, and that the heritage armor is the actual rep grind… but hey.  The allied races should be selling features, but they are timegated behind 2-3 weeks of grinding.

The Bad

The lack of response on the fundamental issues with BfA.  The content itself is not the issue, it’s the systems that support that content.  Re-balancing the Azerite traits is good, but the entire system needs a massive overhaul.  Better view on that here

The Ugly

The need to wait for balancing.  Shamans in particular are getting a rather large re-tune, with the inclusion of two Legion artifact abilities as talents.  Which is better than now, but obviously worse than Legion, where you had 3 talents + the artifact bonus.  Even if it hit the PTR tomorrow, it’s still 2 months of testing.