Thursday, 11 June 2015

Fear of Flying

I knew Blizzard was going to give in and let people go flying. I learned that Blizzard always give in, back in 2090, when they fixed a problem in how group disbandment worked. Before patch 3.1.0, people who left a group in the middle of an instance would be teleported to wherever their hearthstone was pointing. This so-called "ghetto hearthing" was identified as a potential problem for Blizzard's not-yet-released Dungeon Finder. People might queue for instances just to ghetto hearth out: hearthstones had a 60 minute cooldown at the time. When Blizzard announced that they were fixing the problem by teleporting instance-abandoners to the nearest graveyard, there was an outcry from the usual entitled sources, who felt it was their right to be able to ghetto-hearth. To use the other 4 people in their instance in order to hearth out.

And incredibly, instead of telling these entitled idiots that we all have to play nice with each other,  Blizzard gave in to them, and announced they would reduce the cooldown on hearthstones to 30 minutes. I knew then that Blizzard would always give in.

Flying has always been problematic, for reasons that I'll not go over here, as they've all been given a good airing recently! And because it's problematic Blizzard has tried to curtail flying in many ways (and I'm sure they're sorry they ever introduced it). I don't want to get into why it's been problematic, or what Blizzard could have done to  make it less problematic. I want to focus on how Blizzard attempts to curtail it.

Flying was introduced for level-capped adventurers in the Burning Crusade, and at that point Blizzard had a proper 3D layout for Outland, but not for Azeroth. So for ages, flying was not possible in Azeroth. No convincing in-game reason was ever given, it was purely an implementation problem.

But once you have level 70 players flying in Outland, there was a hidden problem left for Northrend. Why couldn't level 70 players fly their mounts in Northrend? The real reason was that it would have made quest design a whole lot harder. Blizzard introduced it in Outland only for level-capped players - so after they had already explored the whole of Outland on horseback, and done all their quests on the ground. Blizzard wanted the same freedom for their designers in Northrend. So they introduced the idea of "Cold Weather Flying", which was a neat way of sidestepping the problem. You needed to train a new skill to be able to fly in such cold conditions. That kinda got them over the hump for Northrend. But wait - if I can fly in Northrend, what magical property of Kalimdor prevents me from flying there? Again, no in-game reason, and this was one of the many reasons Blizzard decided that for their next expansion, they'd redesign the old world to allow them to support "Old Weather Flying".

 So the idea of a "Flight Master's License" was introduced for the old world.  Not a skill that was trained. A licence. We're used to the idea of levels as being proxies for our level of ability. So I am more skilled at level 70 than at level 1. But the problem for the Cataclysm was that I am already well-skilled in flying, having learned how to do it in Outland, and having perfected the cold-weather variant in Northrend. Blizzard couldn't keep inventing new skills that were particular to Azeroth but not Northrend, and it wasn't a viable long-term solution to their problem, which was that they wanted to keep us from flying until we'd done all the quests they'd designed to be done on the ground.

So they embraced the problem in the Cataclysm expansion. and designed their new quests with flying in mind right from the start. They allowed flying right from level 60. In fact, the stonecore dungeon entrance couldn't even be reached on foot, and the Vashj'ir zone was based on everyone swimming (i.e. flying) underwater.

That was a great solution, but for some reason Blizzard didn't extend it to Pandaria. I don't know what they saw that made them change tack (perhaps they deplored the "fly-in, kill, loot, fly-out" play pattern that Cataclysm brought), but change tack they did, and in Pandaria, flying was again only available at the level-cap.

This was a retrograde step for two reasons.
1. We already had the expectation that we would be able to fly in Pandaria from the start
2. There was no convincing in-game explanation for why we couldn't fly in Pandaria.

 Sadly no explanation was given as to why my existing flying skills don't work there. "Wisdom of the Four Winds" was the gate to flying in Pandaria.What is it anyway? A skill? I already have flying skills. A licence? Who is the licensing authority? I already have a "Flight Master's License". Something else? What? This was a lazy solution. There was no in-game reason. As a player, I understand the designers' design reasons, but a solution that made sense to Dàchéng would have been nice.

Now we have reached Outland again, and my well-honed flying skills yet again seem to have faltered. I'm sure Blizzard by this stage were ruing the day they ever allowed flying, but their attempts to put the genie back in the bottle were always doomed to failure. However their "compromise" is as ugly a solution as the Wisdom of the Four Winds. None of the preconditions for "unlocking the ability to fly in Draenor on all [your] level 90+ characters" make any in-game sense.

This is just a gamified solution, with not even the pretense of an in-world reason. Not even an inscrutable reason like Wisdom of the Four Winds. Instead we are asked to collect 100 "treasures". Explore all of Draenor on the ground. Complete all the quests that are part of the Draenor Loremaster and Securing Draenor achievements. Achieve three Tanaan Jungle Revered Reputations. These are hoops for the player to jump through that make no sense to their characters. What has any of this got to do with flying? It's another step away from a believable virtual world.

Wednesday, 10 June 2015

Personal loot changes

These changes to "Personal Loot" in 6.2:

"rather than treating loot chances independently for each player—sometimes yielding only one or even zero items for a group—we’ll use a system similar to Group Loot to determine how many items a boss will award based on eligible group size"


Let's imagine that our 10-man raid is set to "Personal Loot" and will get awarded 5 items of loot. Just how will this be distributed? Does Blizzard look at the first team member and decide "Well that's a 50% chance of them getting personal loot. Let's toss a coin. Heads! You win". Then the next person in the group has a 4/9 chance of being awarded loot? I.e. 4 items left to distribute among 9 players. Is that how the mechanics of this will work? I don't know, but I'd like to. Do you know for definite?

Wednesday, 3 June 2015

First they came for the pet-collectors ...

I know it's a while since I posted about the price of WoW Tokens, but that's mainly because I'm dumbfounded by it.

Blizzard engineers continue to play with the algorithms governing the in-game price of  WoW Tokens. By now I'd come to doubt that the relationship between price, supply and demand is anything more than tenuous. I note that we have never seen any data on actual volume of transactions, and also that has ceased to report the "time taken to sell", presumably because it was stuck at a fixed "4 hours" for so long. I think that in their effort to smooth price fluctuations, Blizzard's engineers have erred in reducing current supply and demand to a minor input in their equation.

I continue to buy the game-time tokens from time to time when I see them cheap on the auction house.

However, I have cancelled my subscription, and won't be renewing it when my gold runs out. To me, these tokens are the point of no return. In effect, Blizzard are now selling epics for cash (with the intermediate step that you must first exchange your cash for gold, then exchange the gold for epics). this is something many people (I include myself) have been warning about for many years, and it has inevitably happened. Take a trip down memory lane with my favourite mage blogger, Larísa of the Pink Pigtail Inn. The context of that article is that selling pets in-game for cash is the start of a slippery slope. It has proven to be so, and all the arguments people made that Blizzard might sell fluff for cash but never what matters (epics) have turned out to be wrong. Larísa, knew it, I knew it, you knew it, deep in your heart.

I know there is plenty of fun to be had in Azeroth that doesn't rely on earthly cash or Azerothean gold: I love exploring and finding little nooks and items not shown on the map. I love to stand, as Painter Mikkal does, and just gaze at the beauty of the world. I love finding routes up to "unreachable places", I love chatting to the many amazing people I have met, and I love killing orcs. But the group finder tool dealt a heavy blow to my sense of immersion in a virtual world; and now being able to buy epics like this has been the final straw for me.

I'm sure I'll keep some gold back so I can resubscribe for a while post-Draenor, and I've still plenty of game time tokens on my account, so I'm not going real soon; but my heart is no longer eager.

Monday, 4 May 2015


Izlain has posed this question as the first TalkBack Challenge in the Newbie Blogger Initiative:

How did GamerGate affect you?

Are you mad, Izlain?

Tuesday, 28 April 2015

WoW Tokens: What the ...

I am at a loss to understand the disparity in the gold (game-time) WoW Token price between the EU and NA regions. Many people have tried to explain it, but none of the explanations ring true.

However my investigations have now uncovered the truth. Let's look  in the NA region first. The token price rolls between about 24 000g  and 20 000g. Let's examine that.

If you want some Azerothian* gold, you would buy a $teel (RMT) WoW Token from Blizzard for $20. If you need the gold instantly, you stick it on the AH immediately, and are take whatever the going rate is. If you can wait a little, you sell it only when the price nears 24 000g (I showed yesterday how the peaks and troughs are predictable).

What about buyers of the gold (game-time) WoW Token? You buy that at auction. If you need the token instantly, you buy it at whatever the going rate is. If you can wait a little, you buy it only when the price nears 20 000g.

So we can pretty much see that for most people, sellers get 24 000g and buyers spend 20 000g. This is a bargain for buyers, of course. If you can make 1 000g/hr, that's only 20 hours of farming**. Oops! That means I'm farming at a rate of $1/hour. That's okay, if it's fun, and you'd be doing it anyway. But otherwise, wouldn't I be better off buying a token and working an hour extra in overtime on Earth? Ah, but if I've got kids, the situation changes. I can use them as slaves.

Now let's look at Europe. On EU realms, sellers get about 43 000g and buyers pay about 33 000g. Great deal for sellers, right? Let's assume again that I can make 1 000g/ hour, so it takes me 33 hours of farming to earn a gold WoW game-time token. Since the token costs €20 or £15, that's an income of  61c or 45p an hour. Thank goodness for kids, eh? Finally all those six-year-old kids out playing football can be put to good use. Sadly it's illegal for them to sweep chimneys now, but there's nothing preventing me from enslaving them in the mines of Azeroth.

So I think I've shown what's behind the disparity. Europeans have more slaves.

* What an ugly word. I much prefer "Azerothean". However the Azerothian Diamond shows Blizzard's preferences.

**  Your mileage may vary. Feel free to use your own rates. 

Monday, 27 April 2015

Tokens: A tangled web

Blizzard changed the algorithm they used to calculate bid and offer prices on the WoW Token, just four days ago. Previous to that, the prices steadily climbed, until suddenly they started steadily falling. And vice versa. A steady climb in price followed by a steady drop. The graph of the North American Gold (Game-Time) Token was almost a saw wave. Blizzard social engineers didn't like that, and around the 23rd or 24th of April, they tried a new algorithm. They tried to curve those straight lines, and I think they are pretty happy with the result, the price graph looks more fluid now.

I certainly loved it. It coincided with the  launch of the WoW Token in Europe, and it made it the most predictable price graph on the planet. I'm sharing this with you now so you can benefit as I have done. the prices are so predictable that I predict Blizzard will be tinkering with the formula pretty soon now.

The "secret" is in the rate of change of prices. They are too predictable, too like a sine wave. The price rises at a steady rate until it's near its peak, then the rate of change drops.

Do you want to buy a $teel (RMT) WoW Token to sell for gold? You want to get the most gold for your money. Don't sell that token until the positive rate of change drops. That signals that the price is near its peak (there are no sharp random behaviours here). That is the time to sell, to maximise your gold.

Do you want to buy a gold (game time) WoW Token? Easy. Just follow the prices downward until the negative rate of change  drops. That signals that the price is near its trough (there are no sharp random behaviours here). That is the time to buy. Don't spend your Azerothean gold until then.

Blizzard social engineers are tinkering with the exchange rate, like the USSR of old. They don't trust the free market. They want to control it so that there are no unexpected jolts to the system. That's what makes it predictable. I've been able to use this in the few days since EU launch to predict the best time for me to buy my tokens, and I hope you have, too.

What about Blizzard? Are they happy with the result? I don't know. They may be happy that both sides of the equation get the best deal possible, at their expense. It isn't a big expense. Smart buyers of the $teel token get a lot of gold for their euro, dollar or pound. Smart buyers of the gold (game time) token, get their 30 days for the cheapest gold price possible. Both sides are happy. The piggy in the middle, who absorbs the difference in gold, is Blizzard - who can print gold coins for free. It's win-win-win (except for the inflation, of course. But these are the sort of economic sins that are paid for by the next government - or development team -  not this one).

Blizzard are experimenting. Blizzard will change this again. I don't think they meant to make their prices so predictable. Expect them to tinker with the algorithm again. Meantime, make use of this actionable information while it is still current. I have.

[Edit: I just noticed that is now including rates of change in their graphs, which makes timing your purchases a snap. When the rate-of-change curve crosses the zero-point, you are at a price max or min. It won't recross that line for several hours. The rate-of-change curve is too predictable.]

Wednesday, 25 March 2015


With further details of the WoW token released by Blizzard yesterday, it's time to review what will happen when it is released, which should happen in the Americas first. It will be rolled out there "in the weeks following Patch 6.1.2’s release". I imagine Blizzard will want to get Noblegarden over before the launch, so they can start collecting data in a "normal" period, rather than one where prices might be driven by a seasonal event. So let's say the Tuesday after Noblegarden, Tuesday 14 April. [Edit: it actually appeared the previous Tuesday, 7 April]

Blizzard have also firmed up the earthly price ($20 or the equivalent in your currency). What they haven't yet done is firmed up the Azerothean price. Gold is still selling on various "illegal" websites for about 1500g/$, so I think that 30 000g is a reasonable place to start. I imagine, though, that Blizzard will want to make this attractive to gold buyers, and put pressure on third-party gold sellers. That makes me think the initial price will be 40 000g for a token. [Edit: the initial price was 30 000g]

Let me first distinguish two tokens here. There is a WoW Token that Blizzard sells for $20. It has only one use: you can sell it in the Auction House for an Azerothian price set by Blizzard. When it appears in the Auction House, it is transformed into a WoW Token that can be bought at a price fixed by Blizzard, and it too has only one use: to add 30-days of game time to your account. As their icons are different (one looks like it's made of steel, the other is golden), I'm going to refer to them as the $teel WoW token (costs $ to buy) and the Gold WoW Token (costs gold to buy).

So let's look at the buyers of the $teel WoW token. Firstly, these are our avatars on earth (the 'players') rather than ourselves in Azeroth (the 'adventurers' or 'characters'). Players buy these tokens for $20, and presumably assign them to a particular adventurer who then lists them on the AH. Why would players do this? Because their adventurers are poorer than the players - they have less gold than they want, while the players have more dollars than they want.

The wants of the adventurers can be split into one-time wants and ongoing needs. Take the typical raider: she may want some pets, mounts and toys. These are one-time wants in the sense that once she has the item, she doesn't need another. To stimulate this want, Blizzard will have to release more pets mounts and toys that she can buy for gold (expect expensive flying licences for Draenor later this year).

Her ongoing needs are mainly for flasks, potions, food and repairs. She also needs gems and enchantments as she gets some gear. Up to now, she has been struggling to meet her repair bill, eats other people's food and skimps on flasks and potions. On a typical raid night, her bill should be about 250g for repairs, and another 250g on the other stuff, but she struggles just to scrape together the money for repairs. If she raids twice a week and does LFR/dungeons another night, she could easily want 1500g/week, though in reality she keeps the bill to 500g by skimping on consumables and not repairing damaged gear before and during LFR. Selling one $teel WoW token will initially cover these ongoing costs for about 26 weeks (if the initial price is 40 000g). Or fewer weeks, but leaving spare gold for one-time wants - mounts, pets and toys. For the sake of argument, let's say the gold lasts the average buyer 20 weeks, and then they will want to buy another.

Who are the buyers of the equivalent Gold WoW Token? They are players with a stockpile of gold, with nothing else worth spending it on. They will be wanting to buy 30 days' game time. If the cost of the token is 40 000g, they need to be earning 1333g/day to keep up. Otherwise, they will run down their stockpile and eventually will have to spend dollars again.

If the price is to remain in equilibrium at 40 000g, then the number of Gold token buyers, who want to buy a gold token every month, must be about the same as the number of buyers of Iron tokens at any time. But we've seen that they need only buy once every 20 weeks or so, while the buyers of Gold tokens want to buy every 4 weeks or so. So the pool of Iron token buyers must be about five times as large as the pool of Gold token buyers for this to be possible.

Now lets look at week 1. The prices will be around 40 000g/20$ that week. Almost every $teel token buyer will spend their $20 that week to get the gold they want. Every Gold Token buyer will be spending their 40 000g to get their free game time. All $teel token buyers will receive the gold they want. Blizzard will see to that. Even if it means printing gold, I very much doubt that Blizzard will return a token unsold to a player. If the pool of $teel token buyers is indeed five times the size of the pool of gold token buyers, all gold token buyers will be satisfied.

What will happen in week 2? Nothing much. Buyers and sellers of both currencies will have dried up, as everyone did what they needed to do on week one.

Week 5? Any $teel token buyer [who as you recall is a gold token seller] bought and sold in the first few weeks. They have their gold. It'll last most of them 20 weeks or so. But gold token buyers, who want free game time, need to use a new token; and unless they stocked up on week 1, they are going to have to buy one. While sellers are still thin on the ground, buyers will be turning up in their droves, camping the AH to snap up any that might appear. Prices will rocket.

So. Whatever the initial price is, be clear that it will rise steeply in week 5 if not before, and continue to climb until $teel token buyers run out of gold again, or see the gold value of a token as too mouth-watering to ignore.

Recommendation: buy Gold WoW tokens on day 1. Buy all you can afford. There will never again be such a concentrated glut of sellers. The price will never drop to 40 000g again. Don't buy $teel WoW tokens before week 5, once the upward price momentum of gold tokens is obvious.

Addendum: I expect Blizzard to be a little surprised by this. Once they see the inexorable rise of the gold WoW token, they'll be scrambling to find things for people to spend their cash on - expect expensive new mounts, new toys, new pets, soon. Also expect repair bills to rise and gold sources (such as quest rewards and mission rewards) to dry up. See also Plexing Warcraft for the social effects of this move.