Tuesday, February 3, 2015

The Rabbit Hole - Looking At Snapshots of EVE's PLEX Market

When I began this series, I wanted to maintain a more general look at the role of real money trading in Massively Multi-player Online Games (MOGs) and avoid looking specifically at my main game, EVE Online.  But with the rising popularity of the PLEX model in the industry, no look at RMT is complete without at least one look at CCP's system for exchanging game time for real world cash.  I am engaged in a long term project gathering data related to PLEX, so I figured I would share some of the results of that work today.  I hope everyone likes bar graphs.

Last year I attempted to track all of the ISK sales at a popular RMT site, Player Auctions, for the first quarter of the year.  Player Auctions has enough data on display that I can make the attempt with a great deal of confidence of tracking almost all of the transactions.  With last year's data in hand, I am attempting to track the transactions in 2015 to see the changes that occur in the secondary RMT market compared to players purchasing ISK though the approved PLEX mechanic in the in-game market.

For a bit of bad news, the ISK sellers on Player Auctions sold 20 billion ISK more last month than they did in January 2014.  The sellers were on a pace to sell less than the previous year, but managed to sell 129 billion ISK on the last day of the month.  Apparently, an alliance or two with really bad ship replacement programs managed to have a bad day.

Now, 1.1 trillion ISK in sales seems like a tremendous amount of ISK.  Compared to what is in most players wallets, that is true.  But what about the approved primary RMT market in which players purchase PLEX from either CCP or an authorized PLEX reseller and trade the in-game object for virtual currency?

Compared to the size of the primary RMT market, Player Auctions' 1.1 trillion ISK in sales barely registers.  In EVE Online's main trade hub of Jita, the amount of ISK sold increased from 55.6 trillion in 2014 to 61.8 trillion in 2015.  I should point out that I am unable to determine the value of PLEX that were converted into game or training time, just the overall value.

In percentage terms, the growth of the primary RMT market, both in Jita and in New Eden as a whole, is vastly outpacing that of Player Auctions.  In Jita, the value of PLEX sales increased by 11.3% compared to Player Auctions' 1.8% for the month of January compared to the year before.

Virtual currency is all well and good, but how much real money are the sellers making?  Compared to last year, the folks on Player Auctions are not doing so well.

In January 2014, the sellers on Player Auctions made a bit over $18,500.  This year?  Only a shade over $13,000.  Compared to the primary RMT market, the ISK sellers make a mere pittance.

Due to sales and different prices for different amounts of PLEX sold, I don't have exact figures for the value of the PLEX sold in EVE Online for the month of January.  But using the standard 2 PLEX price of $34.99 USD, the value of PLEX sold on the market in Jita dropped from $1.6 million down to $1.3 million year over year, while the total value sold (outside the contract system) dropped from $2.4 million down to $2 million.  I should add that I don't know the value of the PLEX that were actually redeemed for game or character training time, just the total traded on EVE's primary RMT market.

Since the amount of ISK sold on both the primary and secondary RMT markets increased in January 2015 compared to January 2014, we can assume that inflation, at least in the PLEX market, is in effect.  Looking at the price of one billion ISK bought in Jita on 1 January 2014 ($29.36 USD) compared to the last day of January this year ($21.87), the ISK has dropped 34.2% in value against the U.S. dollar in just 13 months.  This loss in value impacts both CCP and the illicit ISK sellers on Player Auctions.  But with the primary market seemingly gaining a bigger share of the pie according to the available data, players look like they are heeding the advice to buy ISK and PLEX from sources authorized by CCP.

One interesting fact I was not able to nail down was the effect of the increase in penalty for buying illicit ISK.  On 1 January, a first offense for purchasing illicit ISK was increased to 7 days for a first offense and a permanent ban for all accounts for a second offense.  I probably will need to wait until I have collected data for the first quarter of 2015 to have the data to make that determination.

CCP's PLEX system and in-game market for trading virtual world currency for real world money is a mature market that has an impact on the price of the secondary RMT market.  From both my observations over the past few years along with academics using Player Auctions as a source in their research for measuring RMT in other games, I feel that using Player Auctions as a measure against which to judge EVE Online's primary RMT market is valid.  Using the data I've compiled, I would conclude that CCP is doing something right in their fight against illicit RMT, even if they still have a lot of work in front of them.


  1. It's not like you to miss something as big as this but Jita traded plex isn't 1:1 with plex sold by CCP. They haven't given numbers for PLEX they sell, and have merely stated that sold plex is traded 'many times' on average before it is consumed. I don't have the source off the top of my head, maybe the economics talk from last years fanfest.

    In other words the primary RMT market is many orders of magnitude smaller than you think it is.

  2. I thought I addressed that with this statement:

    "I should add that I don't know the value of the PLEX that were actually redeemed for game or character training time, just the total traded on EVE's primary RMT market."

    I probably should have added that I don't know how much of the ISK purchased on Player Auctions was destined for resale on shady ISK selling sites and not sold directly to players.

    Perhaps I need to define what I consider "the market". That could even be an entire blog post in itself. See what I mean by "the rabbit hole"?

    Thanks for bringing this up!

  3. You haven't addressed it because it doesn't really matter how much is redeemed. You can assume very little is immediately redeemed, if any, as it is so much cheaper to pay with subscription. But that isn't the point I am making.

    What you are comparing ISK sold by third parties with ISK circulated that has at some point in it's life cycle been sold by CCP. You can't compare them this way because there is no way of knowing how much of the value of PLEX traded is RMT.

    If player auctions traded in marked bills you could compare the circulation of RMT'ed in-game value like this but they don't.

    It isn't a case of definition of a market. It's like going to a sports shop and asking how many balls they sold that day then going down the circus, watching the jugglers and counting how many balls change hands.

    They will have bought those balls at some point, yes, but counting how often they change hands and multiplying it by the value of each ball is no way to tell how how much the circus' ball collection is worth.

    Do you see what I mean?

  4. Now you've confused me. Or perhaps I have not been clear.

    The way PLEX works, no actual ISK is sold by CCP. Instead, CCP sells players 30 days of game time which can then be traded for ISK. The individual sale of each PLEX by CCP can differ, so I arbitrarily use a valuation of $17.495 USD per PLEX.

    So for example, if someone with two PLEX sells the PLEX to another player, he is giving the buyer $34.99 USD in exchange for approximately 1.6 billion ISK at yesterday's average price in Jita. They buyer of the PLEX can't extract actual real world money from the game, but can obtain goods and/or services from CCP.

    In the charts above, I just totaled the values of the trades, which is how I arrived at the totals. And that's how I figure I am comparing like with like in comparing the primary and secondary RMT markets.

  5. Right, but each plex can and usually is traded more than once so when you see a PLEX traded your not looking at $17.495 usd of RMT'd isk rather some fraction of it.

    Just scrolling down your page I clocked someone blogging about trading PLEX.

    This guy alone in one blog post accounts for 171 plex bought and 35 sold in a weekend. Thats 206 traded PLEX of which an absolute minimum of 35 are re-sold and thus warping your figures upwards of the primary RMT market. And to make matters worse you can be sure he isn't going to consume many, if any, of them. He just hasn't sold them yet.

    A huge number of traded PLEX are resold in this way. CCP have hinted at the numbers as I have said but nothing concrete.

    You can only compare the first sale of each PLEX with the isk sold by a secondary market RMT site to compare the size of the primary and secondary market, and sadly that just isn't data we have available.

  6. A concerned MinmatarFebruary 3, 2015 at 6:37 PM

    This seems correct, and I think that most PLEX are sold at least twice on the market: first when the primary RMT user sells it to a buy order after receiving it from CCP, the second time when the trader that was responsible for the buy order sells it on sell order and pockets his profit.

    It seems reasonable to cut the primary RMT statistic in half to account for this effect. I doubt many people that buy PLEX for real money bother to cent their own sell orders, and I doubt many traders buy from sell order to relist.

  7. "and I think that most PLEX are sold at least twice on the market:"

    That's just it, you think but you don't know.

    PLEX is the primary asset for those with a poor/limited grasp of eve economics to store their wealth when they go AFK (or even not in the case of ex T2 BPO holders). Unlike T2 BPOs which were always going to get nerfed plex 'always goes up'. Sort of like real estate is/was in real life. An accessible and obvious appreciative asset.

    PLEX price is a bubble, just one without any particular reason to burst.

  8. A concerned MinmatarFebruary 4, 2015 at 7:26 AM

    This is a valid criticism, and something I missed in my post. The net result though is that each PLEX is sold even more than two times, which means the real RMT volume is even less than half of what is reported in the graph. Therefore, I would suggest to still use half of the current data as an upper bound, because we don't have the data to produce a sharper bound.

  9. You are trying to determine how much money that CCP is making. I am describing the worth of the ISK that is purchased. As I stated in the post, those are two different figures.

  10. Okay, after reading this I definitely need to write another post explaining what I mean. I am describing the worth of the volume of ISK traded. And as I stated in the post, I don't know how much CCP actually receives because I do not know the velocity of PLEX. But the worth of the ISK does not decrease just because a PLEX is traded more than once. 30 days of game time is worth the same no matter how many times it is traded.

    Also, 1.1 trillion ISK were sold on the secondary market for real money, while 61 trillion ISK was sold on the primary market in Jita/The Forge for game time, which is worth real money.