Stats and Bonuses

Jim's blog on data and everything else | 🔮💰👑🤔💻📷🍁🌌🔬🏹

The Past, Present, and Future of Online Gaming according to Jim

The Past

Club Penguin (now owned by Disney), 2006

The first game I tried that required internet was Club Penguin. I was in Grade 2 at the time and introduced to this cartoonish game by some peers at Burr Elementary School. To an eight year old, communication with other players, anywhere around the world was a novelty for sure. This was the start of my understanding of MMORPGs (Massive, Multiplayer, Online Role Playing Games), but it soon grew too childish for my tastes.

Club Penguin Secret Agent Code One memorable part of this game was a mechanic called the "Secret Code" in which some missions were written, shown below. After about a week in the game, I was starting to use this alphabet in real life to write with my nerdy CP peers. I didn't think too much about it at the time, but in hindsight it sure was intellectual for a dumb game with penguins waddling around.

Club Penguin Popularity over time It looks like I quit the game long before its peak in popularity. I suppose children slightly younger than me were embraced by an even larger community of waddling penguins that tried to curse at each other. Also interesting is the significant fraction of searches for "Club Penguin" in the developing countries of Latin America. We will come back to this later with some attempts at explanation.

RuneScape (Jagex), 2007-2012

Wilderness After CP fell out of fashion among eight year olds, RuneScape (run by UK firm Jagex) took over. The overly sensitive speech censor made sure no child learned to curse, for sure, n0t 3v3n w1th r3pl@cem3nts. Since players could trade freely with each other for a time, scammers were commonplace. I learned (the hard way) at age nine that deals sounding too good to be true, are probably too good to be true. This game was innovative for having a solo and group Player-Killing feature in the northern "Wilderness" (pictured above), which was subsequently removed for several years due to various reasons.

A fun story: I was able to take complete control of one of my friends' RS accounts by submitting a "forgot password AND security question request" ticket which was evaluated by real people. I believe the key was noting the victim's travel history (moving countries, vacations, etc.) as that would be linked to login IP's. It took a week for new security questions to be set, so I handed back the password to the account after a week of biding my time. Half a week later, I changed the password using my newly-set security questions. He started training a new "pure" strength account after that point on.

This Java-based medieval clicking MMORPG survived for a long time in my list of playable games, especially after I learned of RSBot, a popular scripting engine that interfaced with the game so that you could run Java to perform mundane 'skilling' tasks to level up or generate income. This game forced me to learn a bit of Java and how to set up PATH under Environment Variables, a skill I am very thankful for. I recall spending hours on making JDK work with RSBot, every time Java had an update. I would have the script level me up and get money while I slept or did homework, and then go into the wilderness with the best arrows gold could buy.

RSBot Interface The interface for RSBot looked something like this, and could handle anti-bot measures quite well (logs in for you, does "random events", moves the mouse in a 'human' way, etc.) Applause to Timer, one of the lead developers of RSBot who is probably younger than me.

As to why RS became less popular over time, perhaps it was due to botting. I have searched separately for 'RuneScape' and 'RSBot', normalised each series of results to 100 maximum, and plotted them. Search popularity for RuneScape terms It turns out that player count had been decreasing furiously, even before botting became significant. There is strong correlation between the two series, which I can explain as "RuneScape becomes more popular one month, so more botters are interested in playing again" rather than "botters help the community grow". Additionally, if we look at the two curves several months after Feb '09, we see an anti-correlated month or two. I would bet that Jagex made a large announcement that month claiming (truthfully) to have banned a large number of botters.

I repeat, RS was a dying game even before botters grew in population, so perhaps one could explain the situation as "RuneScape was losing popularity caused at least partially by people realising that it was just a game of clicking, clicking, and more clicking. Therefore RSBot was developed by a Java-savvy population that wanted to have fun in the game without spending days of their life clicking pixels." The first overhaul of the free trade (scammer- and goldfarmer-enabling) system along with PKing in the wilderness may also have turned away many existing players.

Back to developing South American countries and search popularity. I provide two possible reasons:

  1. The games are localised for them (relevant languages provided, and independent servers for social purposes)
  2. As a country is developing, there are relatively few people who know how to use Google to search for information. This leaves just the gamers to search up 'RuneScape cheats' and dominate the search popularity count. Once a country is more acquainted with the information age and Google, old people will start searching for 'how to do laundry' and dilute the search fraction created by game-related searches.

Another observation I would like to make is that search popularity appears... seasonal? We should examine the other figures in this article to see if this is true for all games, and if so, I would first speculate that it is an inherent data-collection artifact from Google and test this idea (at some later date).

Record of Lunia War (allm), 2007-2014.01.01

I was playing RuneScape when I saw an advertisement to the right of the screen featuring this 2.5-dimensional arcade MMORPG. I clicked in, downloaded the approximately 1 GB installer and installed my first game ever. The first thing I heard when opening the game was Ariel Lin 林依晨's song for the game. Player base was significantly smaller than RS, but was large enough to sustain profitability for a while. Most memorable of this game were the difficulty in mechanics and air combos in Player vs. Player mode, the cartoon cutscenes before and after every stage, and the powerful boss monsters before the game was updated after the open beta test. The game evolved to become easier and easier from that point on, but was still fun to play.

Rodesia was a land of elves, snow girls, ice dragons, orcs, humans, slimes, trolls... Such effort was put into the development of this game, yet it was unable to attract a sufficiently large following later on. Around 2012, the game was consolidated from regional publishers to allm's global Lunia server, with no profit margin whatsoever (pretty certain it lost a lot of money). In late 2013 it was announced that the game was shutting down for good on New Year's Day of 2014. If I amass enough wealth to purchase all rights to a game and then revive it in charity to players, it would be this one.

Limey <3 Here is Lime, the fighting slime. This was one of the most efficient boss-killing characters one could play with, since Lime could transform into a variety of giant slimes.

Slime Race Begins Slime Race Payoffs Additionally, slime race betting was an added feature of the game later on. That was what got me interested in betting strategies and probability. At age 11 I learned of the "Martingale" strategy (double bet size until satisfied) from older, nerdier players, but realised that I don't have infinite gold to bet with. I was no good at slime races back then and probably am no better today, so that speaks to my qualifications for a quant trading role... :')

I had to take out 'nokia' from the searches because apparently people search for 'nokia lunia' by accident sometimes. Lunia Search Popularity It looks like Lunia's beta and post-beta period produced a large amount of hype for the game. Lunia's transfor to (around 2008/2009) also revived the player base, but it was all downhill from there. If we examine the popularity breakdown, again a South American country dominates the search popularity scale, with searches for the game in Taiwan and Hong Kong being quite popular too (the "TLunia server" was my server of choice when living in Shanghai). Lunia seems to have caught on fine in its home country of South Korea, with causes for its demise not well-known. Perhaps it got too easy, too pay-to-win, or too repetitive. Not sure.

Note that Google Trends' popularity ratings are per-country; they represent the fraction of searches in the country for the term 'Lunia', not an absolute count of the searches in each country. Thus, if half the people in the island of Guam search 'bananas', they will rank #1 in the 'bananas' list even though they don't eat that many bananas.

Maplestory 1 (Wizet), 2008-2012

Magic Claw first animation Maplestory is more of a household name than most of the above games. It is a 2-dimensional side-scrolling MMORPG featuring (not the same) slimes, mushrooms, pigs, and so on. The most iconic skill I can show is probably the Magic Claw of the wizard class. The beauty of music in this game cannot be compared to that of any other game. In fact, I still think it is worthy of listening to all day long despite the game turning into a chaotic pay-to-win festival.

I am not sure what made this game so popular for so long, but I can bet that the music helped a lot. It was a relatively brainless game that was all farming and very little skill, so maybe that is what is key to a successful game (for younger players). There was also a significant social component to it which can be helpful for retaining players too.

MS Google Trends Similar to the RuneScape vs. RSBot search, I downloaded the CSVs for the non-English searches of MapleStory and normalised each to 100 maximum. Again, immediately clear is the 'seasonality' in the noisy monthly data, regardless of search language. I would now speculate that this is due to the method of data collection rather than an intrinsic property of the online game industry (although it is not too far-fetched to posit that monthly/bimonthly patches cause temporary popularity spikes in online games).

Aside from that, the plot is rather dense. I can make two interesting observations:

  1. Korean (dark blue) search popularity of the game dropped in late 2007, despite English-speaking popularity remaining strong.
  2. What is that orange spike in mid-2015? MapleStory 2 alpha tests being announced? The orange line is for non-Mainland Chinese searches of the game (usually originating from Taiwan, Hong Kong, etc.), and I only remember participating in a Korean alpha test of the game. The Mainland Chinese beta test came much later. This spike puzzles me.

Note: the Simplified Chinese curve (grey line) is probably affected (in an unknown way) by the censorship of Google Search (among other services) in China starting 2009. We cannot conclude the effect is strictly downwards, since the typical MS player could be savvier or more desperate than the average Chinese citizen in terms of VPN/web proxy knowledge. Who knows.

Interesting Note: For release in Mainland China, The publisher (Shanda Interactive) renamed the game from "Maple Valley" to "Adventure Island". Sound familiar? There are more examples of this: 'Golden Arches' instead of 'McDonalds'! 'White Fruit' instead of 'Apple'! 'Coloured Windows' instead of 'Microsoft'!

Another interesting note: 'MapleStory' accounted for a slightly larger portion of searches in Canada than it did in the US. They really like their maple syrup up there, huh ;)

GunZ, the Duel (MAIET), 2009-2014

MAIET is "TEAM" backwards with an 'I' in the middle!

Of all games I have played, this third-person shooter is by far the most demanding in terms of raw keys pressed per second. Of all 3-D games, it is by far the most skill-dependent game. I would not say it is the most demanding in terms of combo skill, because that can be reserved for 2.5-D games like DnF/DFO (not discussed) or Lunia. The glitchy nature of animation cancelling, perhaps by design, caused all sorts of tricks to be possible. A typical sword-shotgun combo would look like "jump, slash, switch, shoot, reload, switch, dash, slash, switch, shoot, dash", all happening within a second and each action requiring an average of 1-2 keystrokes.

Two shots, two slashes, two dashes in one jump.

Link to video here if you cannot see the embedded one

My favourite mode was 'gladiator' mode (swords-only) because I had the uncommon skill of performing 3 slashes per jump. This game is significantly more 'skill-based' than a modern popular title like Overwatch. I am not sure why it died out, but I would at least blame it partially on the money-hungry (publisher for both GunZ and Lunia at one point) as well as the large number of private servers which decreased the profitability of the official servers.

We can see that here: GunZ Google Trends In red is searches for private servers, with some prominent ones I've tried being listed. In reality there are more. Up to late 2009, private server usage continuously increased while official server searches continuously decreased. I attribute the 'cannibalisation' effect of private servers on the official server. After that, everything just went downhill. People probably lost interest due to individual small private servers and a dying official server, soon to be retired by the publishers ijji, and later AeriaGames. Hackers were always there, according to the yellow line. They certainly hurt the game, but probably weren't too bad.

GunZ was both private server-friendly and hacker-friendly. Bans were not too common for hackers, and hackers ruined an entire game within seconds. One of the most egregious hacks was a deathmatch 'grenade' hack, which applied grenades to everyone on the map (even the hacker) after detonation. To survive this, the hacker would use godmode. Alternatively, the hacker could teleport to victims one-by-one, coming with a 'lawnmower' of slashes or an infinite chain of 'massive' attacks. I really enjoyed using hacks on the ijji server, and they were what inspired me to learn how to code. I still have no clue how they were written.

Link to video here if you cannot see the embedded one

GunZ would be second on my list of games to revive.

The Present

League of Legends (Riot Games), 2012-current

Froggen's Anivia This is where things get a little more modern. League is a MOBA game, which is quite different from genres I've tried before. Two teams of 5 players attack the opposing team's minions, turrets, and home base "Nexus". Compared to arcade and third person shooters, MOBAs are more strategy, less raw mechanics. League still requires good mechanics (esp. mouse-wise), but they are definitely slower than both Lunia and GunZ. A major frustration is that the matchmaking system often results in lopsided games, which are immediately clear at 5 minutes into the game and then force the player to stay until around 20 minutes. Somehow, such a game became so vastly popular that tickets to the season world finals at major stadiums sell out within minutes.

LoL Google Trends This plot makes me doubt the validity of non-English data on Google Trends. First, the yellow Chinese search curve is somewhat explainable if we consider that Google is censored in Mainland China, but why is the Traditional Chinese version of it also not showing up after 2013? Also, why is the red Korean curve behaving similarly? We can come to one of 2 conclusions using the blue curve:

  1. People have been searching for things other than video games recently
  2. League of Legends is on its way out the door

I can't imagine the latter to be happening right now, but maybe I am just stuck in my Caltech bubble and League's glory days are long gone. I will look into LoL's population statistics.

Overwatch (Blizzard Games), 2016-current

Widowmaker Sniping Interface A first-person shooter blended with the typical MMORPG 'hotkey skills'. Quite innovative in its ability to blend those two genres and still be enjoyable. Some call it a knock-off of TF2, but I believe it is novel enough to be a worthy game in its own category. It is very team-based (6 vs 6), just like LoL, and matchmaking still feels 'lopsided' at times. It is somewhat less spectatable than LoL due to its 3-D nature (whose view do we spectate from during a competition?) and pay-to-play, which is why I believe it will not be as popular as LoL anytime soon.

OW Google Trends Looks like Overwatch is a stagnating, if not dying sport as well! I hope the decrease in search fraction is due to other searches being more common, and not due to a lack of player interest. Also, small islands/small countries really seem to like searching for Overwatch. In any of those five places, I would probably be happy retiring and playing Overwatch, every day for the rest of my life :)

The grand scheme of things

Big Games, Google Trends Here we have plotted a bunch of popular games I have tried, not individually normalised to 100 this time.

  1. We see that RuneScape (green) holds the record for highest search popularity in a given month. This is reasonable, given that RuneScape once had over 250 worlds with thousands in the most popular one at any given moment. It's dead now, though :)
  2. Club penguin (purple) observes the same timescale of popularity as RuneScape, even the same decline rate. It doesn't peak as high, however.
  3. League of Legends (red) looks about the same shape as the above two, and peaks even lower.
  4. Overwatch shows the same rate of decline in search popularity, but spikes really quickly in popularity at the beginning.
  5. MapleStory... well, it never got too popular in search, and MS is the most popular of the remaining games I've spent decent time on.

What can I say?

  • Games that are in-browser, such as RS and CP, seem to do pretty well in terms of popularity. I explain this by accessibility: downloading 1 GB is a severe pain that drives people away.
  • I suspect that, over time, people are searching for more than just online games on Google. Back then, Google was a relatively new item, not too well-known outside the frequent computing community. As a result, people searching for 'RuneScape cheats' or 'Club Penguin Agent Code' probably accounted for a lot more of the queries than they would today. Today, as people have smartphones and old people have learned to use the internet, 'how do I pay my taxes' is probably diluting the search fraction for 'Overwatch tricks'. This discredits the previous generalisation and explanation of in-browser games being more popular, but I still think accessibility is good for popularity. Fortnite/PUBG mobile do well, since they are just a tap away.
  • Games exhibit linear decay in popularity of searches. A half-life would not be appropriate to describe this. I would imagine that an exponential decay is more relevant at some lower critical population, since fewer players means less social aspects of a game and long waiting times, which lead to fewer players, and so on.
  • Based on eyeballing the green curve above, the linear decay rate of popularity is something like 14 normalised points per year relative to the maximum of RuneScape (100 points)
  • As game search popularity decreases over the years, the lifetime of a game in terms of search popularity is getting shorter and shorter. These popularities ideally should be normalised by the search frequency of video games in general, but I have no clue where to get that data. It would be silly to conclude that League of Legends is going to die in a few years based on the plot above alone, but the probability is not zero. I've seen games that I enjoy die on me out of the blue.
  • I really should investigate the curves for PUBG and Fornite, but I only included games I care about.

The Future

Of course, it is easy to look back on the past and craft a theory to fit the narrative. I hope that my points remain valid in the future, though :)

I believe the key features to League of Legends' success are:

  1. free-to-play nature
  2. spectator mode
  3. team-building effect for groups of players: kill other players
  4. MOBA/FPS specific: all players start each game as equals

The first two are nothing special, but the third... It is fun to kill bosses and farm stages with your buddies, but what could be more fun than killing and farming other players? GunZ may have all of these properties, but it doesn't have the last one: starting games on a equal footing. In GunZ, you could have a better shotgun than someone else just because you played the game for longer. All MMORPGs inherently do not have #4 (except outside of some PvP), since people level up by killing monsters. MMORPGs diminish the effect of #3 by having monsters in their games. It is difficult for a game to be televised or made popular if most of the players just kill monsters.

Let's pretend GunZ did have all four properties. It probably wouldn't be as addictive since it initially had no matchmaking system to pair up players with equal skill. If you were matched up against those people in the above video in your first game, you would probably quit GunZ immediately! The good news is, such matchmaking systems are not too difficult to implement these days, and GunZ would probably do just fine if it had enough players.

Those four properties implemented in a game effectively and relatively fairly, are what I believe to drive success and popularity of games. There are probably several other points that are necessary for popularity:

  • Appeal to players ages 16+
  • One game lasts long enough to spectate and televise

The first condition is rather subjective, and the second one just sounds arbitrary. Know that I am not saying these are what make a game fun or good: they are what make a game popular. In order to appeal to the non-gaming majority of the world, a game needs to be spectatable. If nobody watches it, then how will people learn of it and play it themselves? It is also why some real-life sports are more popular than others: nobody likes to watch sabre fencing, because they have no clue what is going on and points are scored while the spectator blinks their eyes (faster than GunZ).

So what about Overwatch? Why is it doing well?

Overwatch lacks only #1: free-to-play. This would be a significant problem if the game were targeted towards younger kids, but is generally fine since it targets ages 16+. People are willing to shell out $30 for a game that they see played often and looks fun.

What can Lunia do, if it were revived?

Almost Nothing. Pure PvP is not an option - Lunia would lose most of its flavour without its Player vs. Environment stages and bosses. It can, however, be more like DnF/DFO (Dungeon Fighter Online) and implement a 100% equal, fair PvP mode that is purely skill-based. There are some mid-high profile DFO tournaments in Asia and Lunia could follow this path.

Alternatively, it could implement a MOBA-style, large map mode that takes 20 minutes+ per game, but have the player be controlled by arrow keys instead of the mouse. This would make it both spectatable and team-building (most PvP in Lunia is at most 2v2), and probably fit in all of the categories for success.

PUBG? Fortnite?

Aside from free-to-play, both games fit all 6 of the conditions listed above. There is just something special about the battle royale genre that spikes your heartbeat rate, not too commonly experienced in other games. I don't foresee it being too popular in the E-Sports realm for similar reasons to Overwatch: it is hard to spectate a 3-D game during tournaments.

VR Games...

Will not catch on unless something like Sword Art Online is developed. With brain-computer interfaces around, we will have more serious fears to be concerned about.

The future of MOBAs

As new characters get added in to the big games (LoL, DotA), balancing becomes harder and harder. People will probably get sick of these MOBAs at some point for various reasons (changes they don't like, matchmaking frustration, plain lack of hype, etc.), but there is currently a healthy professional player population that feeds back into the play-watch-play-watch loop for the average player.

The future of FPS?

I believe traditional grenade + AK-47 FPS games are doomed to fail in terms of popularity due to a lack of appeal to most women (certainly true for most of the years I've been alive). That is a severe handicap to any game developer, and can explain why games like Maplestory survived for so long despite having no 'competitive' PvP scene: it had support from both genders. Overwatch seems to be faring better in terms of gender ratio than the traditional FPS due to its mixed MMORPG + FPS approach. My generalisation fails for PUBG, a game that gets as close to traditional FPS as one could imagine, yet has a decent (better than FPS average) gender ratio. Maybe the heart-pounding excitement of battle royale simply tramples over the bias against guns and grenades; or maybe women are starting to be more accepting of that genre in this year; or, more likely, maybe the neural network in my head is just overfitting to the noise at this point ;)

Third-person games?

There are plenty of third-person MMORPGs such as Dragon Nest, Runes of Magic, Continent of the Ninth Seal, Blade & Soul, etc. These all failed to catch on, probably because they are not fun in a PvP sense. They are much too slow, which I find is common to most 3-D games less GunZ. As people grow up, they stop 'grinding levels' and have real work to do, so the typical grindfest style games lose traction as the audience ages. ALLM (makers of Lunia) has made Kritika Online, a third person PvP game which looks like a blend of Lunia, GunZ, and Blade&Soul. I am not sure how it is doing, but it certainly looks fun and somewhat mechanical. MAIET (makers of GunZ) has come up with RaiderZ, which looks like a complete disappointment in terms of mechanical skill.

Why do you play games with such poor graphics? RTX 2080ti! Raytracing!

I don't really care for the titles that require good graphics cards and look super-realistic. A game's entertainment comes not from its graphics, but rather its design and mechanics. I prefer to be playing with or against real humans.

2.5-D games?

I've never seen a game quite like Lunia, but I imagine that a similar game could be somewhat successful if it focused on group PvP. It could be spectatable, televisable, have competitive player-killing, 'fair' initial conditions, and be free-to-play. It might attract ages 16+, but that would take some innovative game design.