Tobold's Blog
Wednesday, August 05, 2015
 
Next WoW expansion release date

All subscription MMORPGs have the same universal curve of growth, plateau, and slow decline of subscription numbers. Nothing Blizzard announces this week about the next expansion can change the basic trajectory of slow decline WoW now is on. It is stupid to think that any design feature is responsible for subscription numbers, that WoW was better while it was growing than it is now. If Mists of Pandaria had been released in 2007 and Burning Crusade in 2012 instead of the other way around, that wouldn't have changed the basic universal subscription number curve at all. It's not them, it is us, the curve is based on network effects and people getting bored with any game, not design features.

Having said that, any expansion results in a blip on that curve. And the financial impact is significant: If 10 million people buy a $40 expansion and there is a 3 million subscriber blip for one quarter, we are talking about additional revenue in the order of 500 million dollars here. Which is exactly why Blizzard has always said they wanted to release expansions faster: It costs them only a fraction of the added revenue to make an expansion. They would release an expansion every year if they could. But large organizations have their own internal production dynamics and Blizzard has proven again and again that they are unable to speed up the pace of expansions by much.

Now previous expansions were announced in November Blizzcons, while the next expansion is being announced this week at the Gamescon. From that some people concluded that Blizzard is going to announce the expansion for this Christmas. Dream on! Due to the mythical man-month they can't just throw more resources at the problem and speed up production by much. If the next expansion comes "faster", they'd be lucky if they could manage to release it in one year, around the time of the next Gamescom. It is not that they can't see all the good reasons for an annual expansion, or don't want to make faster expansions; they just can't manage it. The next expansion for World of Warcraft is going to be released in 2016, and more likely in the second half than in the first.

Comments:
I think you you are probably right. While they can't now throw people at the problem, they could have thrown people at the problem. E.g. had a sizeable dev team working on 7.0 concurrent with 6.0. In this fictional improvement, even a December ship is not inconceivable. But "I have no way to predict the future but by the path" and our experience is you are correct, disappointingly late 2016.

I would say almost all products, not just MMORPG, follow what business schools call the product life cycle, q.v.

“Why are people leaving WoW?” from eight years ago.
http://www.raphkoster.com/2007/06/15/how-open-big-virtual-worlds-grow/
 
It is stupid to think that any design feature is responsible for subscription numbers, that WoW was better while it was growing than it is now. If Mists of Pandaria had been released in 2007 and Burning Crusade in 2012 instead of the other way around, that wouldn't have changed the basic universal subscription number curve at all.

Is that not an assertion that game design as a field of study is useless, that there are no good mechanics or bad mechanics, that literally nothing else Raph Koster says has any meaning?*

I agree that, over the long-term, decline is probably inevitable. But design features can absolutely have an impact over aggregate subscribers, speeding up or slowing the decline based on player impact. How you play Warlords today bears little resemblance to how you played back in Wrath, even if everyone else you played with was there. The types of daily activities you are asked to do, how quickly or slowly you can progress your character, the gameplay arc of your average night, and so on.

Every expansion since Cataclysm has seen a ~25% decline, but Warlords has accomplished that in six months instead of two years, not even counting the initial 10 million expansion bump. I find it a bit absurd to suggest that nothing of what Blizzard's design team did either accelerated that outcome or could have slowed it. Do you really think it's all the same?

In any case, I have little faith in Blizzard's ability to pull out of the tailspin with the next expansion simply because the designers they do have are dumb. Really dumb. The best we can hope for is that they hired enough people to have two entirely separate design teams working on individual expansions.

* I'd argue Raph hasn't had anything relevant to say in years regardless.
 
I'd argue that while the shape of the curve is universal, the height of the peak and the width of the curve are not. They depend on the fundamental quality of the game, as well as historical context. Game design matters in that the original World of Warcraft game design set the parameters for the curve which the game has followed since.

What is less important is minor design features that tend to be overestimated. People talk about raid design in Burning Crusade, but at the time there was raid attunement and no LFR, and only 3% of the player base raided at all. That certainly didn't affect the subscription numbers of the 97%.
 
Like with every single investor call, thirty seconds after it concludes everyone on the internet suddenly becomes an MMO design and player retention expert. I saw a reddit thread extolling how "envy" was the necessary ingredient that WoW lost; so because we started letting people beyond those 3% raid and get epics, the envy was lost and the game went under.

Everyone wants to find that one fundamental flaw. They treat it like the WoW-Mobile just crashed and want to know what bolt or screw broke. Instead they miss the fact that the WoW-Mobile has been driving for ten years and EVERYTHING is getting worn out. Sure one bolt might break here or there but that's a symptom, not the problem. This strained metaphor meant only to explain that there is nothing "wrong" with WoW. As you pointed out, it's just old and following the same damn curve every other game does.

Sure, the curve might be longer or shorter depending on dev choices, but it's still inevitable.
 
Note how the just announced Q2 subscription numbers of WoW are exactly on the curve that one would expect without assuming any effect of Warlords of Draenor. See curve here.
 
Blizzard has claimed that the WoW team absorbed a large amount of workers (the Titan team) while working on Warlords, but that they couldn't contribute to WoD because they needed to be brought up to speed (that might also be a factor that delayed WoD - the current team would have to spend time tutoring the new guys). I *think* they also said that all those new guys would be ready to work on the next expansion. If that is correct, there is a legitimate chance that Blizzard has finally managed to shorten the gap between expansions.

The announcement this time is 3 months earlier than all other announcements, which indicates that they are, indeed, ahead of schedule. I'd take to indicate a release either summer 2016 (if they announce it a year ahead as they've always done) or Spring 2016 (if the wait between announcent and release is shortened exactly as much as the period between releas and next announcement).

There's even an outside chance for a 2015 release, because if they have managed to double the pace, I doubt they'd announce the next expac at the same time they launched 6.2. Bad PR move. I don't believe it will be released in 2015, but there is a glimmer of hope.

Conclusion: WoW expacs have been regular as clockwork up until this point. The Gamescom announcement breaks this, which indicates that previous experience may be invalidated. We just don't know at this point.
 
"If Mists of Pandaria had been released in 2007 and Burning Crusade in 2012 instead of the other way around, that wouldn't have changed the basic universal subscription number curve at all."

Such an insane statement, as Az already mentioned. Especially because the decline point isn't a set number. Some MMOs decline after the first month, others after three, and some not until ten+ years. The 'why' is very complex, but quality of the game and design decisions are a HUGE factor in that.

Anyone looking at the decline of WoW's growth since WotLK and just saying 'whelp, that was unavoidable' is a fool. Clearly vanilla and TBC did a lot of things right to keep the game growing over that timeframe, just like clearly WoD had design factors that caused people to leave at an accelerated rate.
 
At this point they COULD if they wanted to.

There's no reason you couldn't have three teams going now, each working on the 2015, 2016, and 2017 expansions at the same time. Once the 2015 team wraps up their expansion, they become the 2018 team. Since each expansion radically changes everything anyway it would be relatively simply not to cross those streams.

I don't think this would really be a great investment decision, but it's hardly impossible. They just chosen not to do it. I think that spike you see from expacs are mostly nostalgia buys. Ex-players who want to pop back in for a month or two. If you had a new expansion every year it would just be too much and water down that money which is kind of the point.
 
The fool is you. See the curve I linked in the comment above, it corresponds exactly to Raph Koster's universal curve, and the expansions are just blips that cause short-lived deviations from that curve. Every single MMORPG that has been around for a while, and meanwhile that includes EVE Online, has declined from peak numbers. Are you saying that EVERY SINGLE MMO COMPANY is making the same mistake of designing a brilliant initial game and good early expansions followed by bad later expansions?
 
Avoiding the battle.net forum for a week or two until all that whining and doomsaying dies down a little ;)

Old game is old. That Koster curve looks natural to me.

I basically do the same ingame now as I did 9 months into MoP, 9 months into Cata, 9 months into WotLK or 9 months into BC. Noone who whines about Warlords having less content could tell me what overall he had more to do in earlier expansions.

WoW will not somehow magically get revolutionary new game play, and after ten years it doesn't surprise me the least that subscriber numbers drop.

That said SoO went on for too long and I hope that the next expansion comes a bit earlier. I have not canceled my subscription in SoO because after 10 years the 13€ were just another fixed cost of living to me anyway. And now it's paid with gold.

Warlords so far has an OK raid content speed for my beer'o'clock raid. Maybe even a bit too fast.
 
"Are you saying that EVERY SINGLE MMO COMPANY is making the same mistake of designing a brilliant initial game and good early expansions followed by bad later expansions?"

Yes, more or less (few can say they had a good initial release+expansions, so no, not 'every single mmo company'). How else do you explain EVE growing for almost 10 years, while WoW only growing for 5ish, FFXIV still growing after 2+, and a near-countless amount of MMOs not growing a few months after release?

Or in other words, simply saying "all games decline" is true if the timeline has an end-point of 'forever', but if the timeline is 1, 5, or even 10 years, that's not true. The far better question is WHY isn't it true for some on the 1, 5, 10 timeline? And why is WoW, with its near-limitless resources and massive momentum due to a giant playerbase, not the gold-standard others try to reach in this regard?
 
Ugh, you made me agree with syncaine again. How does this keep happening?

"Every single MMORPG that has been around for a while, and meanwhile that includes EVE Online, has declined from peak numbers. Are you saying that EVERY SINGLE MMO COMPANY is making the same mistake of designing a brilliant initial game and good early expansions followed by bad later expansions?"

This is a ludicrous statement. The curve that nearly every major MMORPG release follows bears absolutely no resemblance whatsoever to that curve. They start with a huge spike at launch, followed by a sharp drop-off over the next 3-6 months, and then a slow but steady decline. They have no growth period at all, and most of the decline is almost immediate. The decline then decelerates, not accelerates.

You are picking the only two MMORPGs out of dozens to follow that curve, and then saying "this is the curve that all games follow, it cannot be stopped." Only a very small percentage of games follow that curve, regardless of how pretty it is.

You are also contending that NOTHING Blizzard has done or could do could possibly have any effect on subscription numbers? Really? The developers themselves admit they made massive mistakes with WoD, the players have heavily criticized it, and you are really going to contend that plays no part in the increased decline? If they had released a great expansion that everyone loved, their numbers would still be the same?

The product life cycle assumes an unchanging product. You don't see car companies saying, "oh well, Model X is just following the inevitable decline." They redesign and update the model. And if they do a good job, and the new version is well received, they will sell more than ever. There is nothing inevitable about it.
 
You don't see car companies saying, "oh well, Model X is just following the inevitable decline." They redesign and update the model.

You don't see car companies saying that they noticed that Model X sales were dropping after 5 years, which incidentally was the year in which they changed the windshield wiper design, and conclude that it is the windshield wiper design that caused the decline.

They design a COMPLETELY new car and say that this is "the new Model X". And expansion of a MMORPG changes the game much less than the difference is between two designs of Model X car.

How else do you explain EVE growing for almost 10 years, while WoW only growing for 5ish, FFXIV still growing after 2+, and a near-countless amount of MMOs not growing a few months after release?

I already explained that in a comment above: It is the quality of the INITIAL game and the historical context (how competitive the market is for that genre for example) which determines the height of the peak of the curve, and the number of years to peak. For example if EVE Online had launched into a market full of space-trading / -fighting MMORPGs, it wouldn't have had that longevity.

They start with a huge spike at launch, followed by a sharp drop-off over the next 3-6 months, and then a slow but steady decline.

That is still the same curve, just very much compressed with a time to peak of a few months. And apparently you don't know many MMORPGs, because MMODATA.net has at least a dozen curves of games that follow the Koster curve over multiple years.
 
"You don't see car companies saying that they noticed that Model X sales were dropping after 5 years, which incidentally was the year in which they changed the windshield wiper design, and conclude that it is the windshield wiper design that caused the decline."

Garrisons are hardly a "windshield wiper." A more apt comparison would be a new seat that is very uncomfortable and cannot be adjusted. The car designers all agree it was a huge mistake, and there is a large outcry from customers who hate it. Someone at the car company is getting fired.

"They design a COMPLETELY new car and say that this is "the new Model X". And expansion of a MMORPG changes the game much less than the difference is between two designs of Model X car."

A lot of players would contend WoW is a "COMPLETELY different game" than it was at launch. You could point to some fundamentals, and it would be similar to the car company using the same frame and engine. I think the amount of change with expansions is very comparable, if spread out over a few expansions rather than the less frequent car redesigns.

"That is still the same curve, just very much compressed with a time to peak of a few months."

I heavily dispute the idea that WoW's 3 years of steady growth or EVE's 10 years, are "the same curve" as a typical launch with immediate decline. However, you can't really deny that the curve you point to calls for deceleration in the decline like a bell curve, but WoD has caused an acceleration in the decline, which is really the point we're debating here. Even if you regard the curve as inevitable, the quality of the content still shows a drastic impact.
 
"They design a COMPLETELY new car and say that this is "the new Model X". And expansion of a MMORPG changes the game much less than the difference is between two designs of Model X car."

The Toyota Corolla would heavily disagree with you and the NGE, to pull one easy example.

"I already explained that in a comment above: It is the quality of the INITIAL game and the historical context (how competitive the market is for that genre for example) which determines the height of the peak of the curve, and the number of years to peak"

So the initial quality of FFXIV explains its peak of today? That makes about as much sense as suggesting that the MMO market at the peak of WoW's growth was inferior to the heavy competition it faces today, hence the drop. Or that today EVE has more Sci-Fi MMO competition than it has over it's 10+ year growth state.

You've dug a lot of holes here in the past, but this one might be your deepest. Just admit suggesting that the quality of added content not having an impact on a MMOs popularity was foolish, and lets move on to the far better question of why WoW is declining so rapidly, especially compared to how rapidly it was growing back in vanilla/TBC.
 
Syncaine, if you continue calling me names, all you will accomplish is that I'll delete your comments.

Added content has no impact on the base curve. Of course added content has an impact on the size of the blip that deviates from the base curve whenever an expansion is released.

That is VERY visible in the curve shown here.

WoD has NOT caused an acceleration in the decline, just the opposite. It was one of the more successful expansion in causing one of the bigger upward spikes in the curve. Every blip is followed by a return to the basic curve, and if the blip is higher, the drop back to the curve is bigger. World of Warcraft today is at EXACTLY the number of subscribers you would have predicted for today a year ago, before Warlords of Draenor even came out.

So the initial quality of FFXIV explains its peak of today?

You can see how thin your arguments are getting if you have to cite the only game that actually got completely redesigned. And yes, the subscription curve of FFXIV is explained by the quality OF THE RE-RELEASED new game.
 
This really isn't just "My Curve." It's known under other names in business schools. I just presented it from the angle of empirical data from multiple MMOs.

https://en.wikipedia.org/wiki/Bass_diffusion_model

A game that shoots up fast and falls quickly is one that quickly penetrates its entire available market, then declines as the players age out of the product. With granular enough data, it has the same curve, just squashed on the X axis.
 
What 'blip' did TBC create? Can you show me on the graph where the blip happens and then we return to the basic curve? How about the first few EVE expansions; where is the blip from those?

WoD had more initial buy-in because it was hyped as a return to vanilla, and is seeing an increase in the decline because its far, far from the design that made vanilla work. Suggesting that not only was WoD overall good for WoW, but that it also has no impact on the now rapid decline is... well its something alright.

And again, you're general flaw here is you want the X of the curve to end in infinity, which is silly and useless, but is the only way your writing about the curve has a chance to hold up. (The content stuff still doesn't, as TBC and the other examples show).

Plus if we were having this conversation in 2014, one could still point to EVE as being the example that proves you wrong, so for now at least this only works if the timeline is 12+ years, which when you consider all MMOs ever released, is pointless for most of them. A curve of 2-3 years would be far more useful and accurate, but unfortunately a 2-3 year timeframe would contain plenty of examples that contradict your suggestion, so back we go to the mostly useless chart with X > 12 years.

And lets pretend AAR was in fact a complete redesign, rather than an overhaul of some mechanics while keeping the core engine, sound, graphics, and gameplay the same/similar (IE, an expansion); how does the more recent expansion effect the curve? Should be a blip right? Or is it another TBC example where a blip doesn't happen because the content added is sustainable and appealing to the core base, and a quality product will continue to benefit from positive word-of-mouth, getting new players while still retaining enough of the old ones?


 
I still am placing my bets that the next xpac will release holiday 2015 or spring 2016. I feel like financially blizzard has a lot of motivation to get it out sooner rather then later.

I know they always say it and fail but I'm banking they will get it out this time.

Almost 2 years between content updates is just not cutting it anymore.
 
@Tobold

The curve you linked perfectly fits the data because it is drawn to perfectly fit the data. If it did not fit the data, it would have been drawn incorrectly. Unless it was drawn by a small child, it isn't something of significance.

Aside from that the curve proves me right. Raph's curve is a standard bell curve that calls for the drop-off to decelerate. The curve you link to accelerates. If content has no impact, you need to explain why your curve deviates from the curve Raph predicts.
 
And again, you're general flaw here is you want the X of the curve to end in infinity

Oh great, now we are at the stage of the discussion where due to your lack of arguments you resort to putting words in my mouth and then fight strawmen. Where did I ever say anything about infinity? The curve I already linked to twice clearly shows that WoW will have less than a million players left in 2022. I think that somewhere around that point the game will be shut down.
 
"Oh great, now we are at the stage of the discussion where due to your lack of arguments you resort to putting words in my mouth and then fight strawmen"

Not really, you just don't have an answer to everything else written and are trying to derail the conversation with a short reply that doesn't address anything that I posted, so looks like we are done here.
 
The graph that MMO Champion posted http://www.mmo-champion.com/content/ tracks Koster's hypothesis almost exactly.

I think the only way Blizzard could have avoided this fate is to release a new game after one expansion... based on the original, of course. But new. Base the new games on the old one, of course... reuse a lot of the art, basic concepts, etc.

They "sort" of did this with cataclysm. (Albeit after 2 expansions.) They redid the terrain with the Deathwing damage and such, and redid many of the quests. What they DIDN'T do , and should have, is reset the level curve back to 60 levels. People that move over from the old game? Give them Heirlooms and whatnot to simplify their leveling if they want it. They did that to a degree... bot only for multi boxers.

WoW went so far because it reached so high. And it did that because the quality was there at launch. Every other game at the time was poorly thought out crap next to WoW. The game I was working on at the time, Dark age of Camelot, suffered the same fate as all the rest.

Now, if you'll recall, Everquest released a new game (Everquest II) but it was just not as well put together as WoW, and was crushed like the rest.

 
"The curve I already linked to twice clearly shows that WoW will have less than a million players left in 2022."

Actually, it shows a current predicted drop of about 2 million per year. So a year from now (barring another expansion in that time), it is predicting roughly 3.5 million subscribers. Two years from now (which should be well after any bump from the next expansion), the game should have 1.5 million subscribers, based on this curve.

Are you willing to agree with this prediction? I find it more likely a different curve will be drawn, and people will say, "look how perfectly this fits our new curve!"
 
Samus, as you pointed out yourself, the theoretical curve isn't linear at the end (even if it is in a linear stretch right now). Decline should slow down somewhat. So I stand with my prediction that the year in which WoW reaches less than 1 million subscribers is more in the range of 2020 - 2022.

Note that the theoretical curve only applies to subscription games. So if a game goes Free2Play, all bets are off.
 
Post a Comment

Links to this post:

Create a Link



<< Home
Newer›  ‹Older

  Powered by Blogger   Free Page Rank Tool