World of Tanks, rigging, and its defenders

In early February I returned to play World of Tanks. I had quit sometime late 2014 or early 2015 after playing hours on end for 6 months because I needed something to show for in my life. I had read at a later point a convincing argument that the game was rigged, which stopped me from coming back at various times, but this time I thought, if I don’t care about winning or stats, then it shouldn’t matter. I knew the itch I needed to scratch was being able to play around with pretty things in a pretty environment – but out of all the games available to me, it’d “be nice” to be “doing something” at the same time.

Pretty quickly I recognized that I wasn’t able to / it was impossible to simply disregard winning and losing and that the game really was rigged. My brow was permanently furrowed for about a week after I started playing at all hours of the day. I looked it up again and came across a different analysis as well as an actual patent for rigging, and thought I might as well collect some of my own data and make my own while I’m still playing the game. It was obviously frustrating, obviously rigged, but maybe it’d be a bit of a fun project to collect data and then play around in Excel a bit – something, anything to keep me in game so it could scratch that itch. I was collecting data, arguments, and some responses too – my plan was to write my own “Don’t Play World of Tanks!” with my own approach.

Then I played NieR:Automata[1] and I realized there was no point in bothering.

I won’t say there was never a dull moment, but N:A in light of everything else was a wild ride from beginning to end. A masterfully crafted story and experience. So much that on /v/ the contrarians marketers couldn’t help but reveal themselves by saying verifiably false things, revealing their position and intent all along. Things that anyone who actually played the game, regardless of whether they liked it or not, would know where wrong. For people who hadn’t played the game though it wouldn’t be so clear. Who’s telling the truth? Which opinions are the reliable ones? At a glance rather than an overwhelmingly positive response it’d instead look like the response to the game was “controversial”, when it’s nothing but. Both the experience of playing and of reading the feedback to N:A convinced me that it was a waste of my time to gather data and construct an analysis[2].

First of all my approach was completely wrong. All the detailed complicated list of numbers is miniscule stuff, the best arguments should always play the biggest moves first. The first question is whether or not a game is fun / worth playing. In this sense it’s obviously not unless you have a tank fetish. If you’re in WOT just for something to play in, it’s a game where you have to play for tens to hundreds of hours being nothing more than a punching bag, and then maybe you get to punch back. Maybe[3]. These things are neither up for debate nor contested at all; people actually argue that many higher tier tanks are balanced because there’s a bunch of shit tanks that come before them. But why would anyone do that when they can play N:A or any number of other games where it’s all fun all the time? Do you have tens or hundreds of hours that are not only at your disposal, but hours with which you have nothing better to do with your life? I’d get more out of scrolling down Facebook. This is even more clear for premium tanks, which cost 50+ USD. For about the same amount I’ve gotten mediocre full titles which I had 30~50 hours of fun. 30 hours in WOT is about 250 games, and there isn’t anyone whose honeymoon with a premium has lasted that long. Not unless they have a tank fetish. For everyone else the game is suffering; if you ask /wotg/ whether or not you should play this game, you will get a resounding “no”, if you ask why they still play it, they’ll say something to the effect of “stockholm syndrome”.

I don’t know what textbook stocholm syndrome looks like, but I imagine it’s pretty close to what WOT players exhibit. Most supporting evidence for the argument that WOT is rigged is not only not contested, it’s fully agreed upon, and is “funny”. German tanks with their small shells and long barrels have high accuracy, but isn’t it funny that big Soviet shells with short barrels can smack people from across the map? No, that doesn’t signify that there are hidden stats, it’s just an effect of RNG! An RNG that consistently swings one way and never contradicts the experiences of thousands of people who are all saying it works one and only one direction. Nevermind that historically people actually complained and believed in hidden stats, that WOT’s devs denied they existed, and then eventually somebody found the numbers for them and then they admitted its existence. Stats which are to this day not available in official materials and can only be found in third party sites.

The T110E5 is listed as having 203.2mm frontal turret armor. A third party 3D model inspector shows that 203.2 is true for about half the turret – and that’s if you only count how thick the armor is, rather than how effective it is, which is what really matters. The CDC is listed as having 35.29 horsepower per ton. You would think it’d run twice as fast as the FCM, which has ~18hp/t, but it doesn’t because of hidden stats dealing with how a tank’s tracks interact with different ground types. It basically goes as fast and accelerates about equally with a tank whose official stats say it should be twice as slow. There is no 3D model inspector in-game. There are no terrain resistance stats available in-game.

Since these third-party sites are known to most people the people who play and are vocal though, it’s okay. It’s okay that stats were hidden before because they are known now, it’s not like it means there’s more stats that are also still hidden, you can’t show they exist, so how can you say they do? Sure that tank is OP, but you could just get it and be OP yourself, I don’t see the problem here???

3nawxqi
This is not an employee of the company that makes this game.

zgzwyfi
It couldn’t possibly have any sensical meaning to it, it’s a contradiction!
Who cares if it’s a contradiction, it’s funny! There’s no deeper meaning to it, NERD.

upq9cdk
“2” is the standard hotkey assigned to premium ammunition, which almost always has strictly better listed performance.

My favorite WOT player is known for being critical, calling out things the things he sees wrong, and being both generally correct in what he calls out and not missing in what he doesn’t. Doesn’t fit any traditional definition of “shill”. But in the end he streams the game for a living, and thus, can’t say anything actually bad about it. He can ramble and rant all he wants about how this decision is bad or that decision is bad, but he can’t say – or rather, frame – something in such a way that will make people want to stop playing. The closest he gets is saying that a new premium tank shouldn’t be bought because it and the past several premiums have been powercreep, which means that the next yet-to-be-released premium tank is probably also going to be powercreep, and thus not worth your money. He will never conclude “you should stop playing this game”. And neither will anyone else who plays this. Or, in obverse: so long as you don’t explicitly conclude that the game is rigged and people should stop playing it, you can continue to provide evidence that it is rigged.

The moment you do conclude it’s rigged, they throw a riot.

All the usual stuff is said – git gud, mad cuz bad, looooserrrr, you’re just paranoid – which isn’t so big a deal and also makes sense why it’s so common among certain types of people if you remember reasoning is basically always ad hoc. Yes my reasoning is ad-hoc too, I believed the game was rigged first before I went looking for the evidence, no I don’t see a problem with that[4] because I am right and they are wrong. Whether I’m bad or good has nothing to do with whether the game is or isn’t rigged. Certainly it would be better in every sense if I was the best player ever or a billionaire or whatever, but I can’t really fix that. It also doesn’t matter if it’s true that it’s possible for me to improve my performance. I have improved in the past so I know it’s possible, but that doesn’t change that the game is rigged – that is to say, it runs by rules other than the ones which are explicitly stated. It doesn’t matter that it also runs by rules that are explicitly stated, it matters that it doesn’t only run by that set.

It is possible to show, but not demonstrate, that the game is rigged without any detailed stats.

The most dreaded time for a WOT player are 3x/5x bonus exp weekends. At these times, the “tomatoes”, referring to the red color of these players’ ratings on a third party site, come crawling out of the woodwork, the “weekend warriors”, the “casuals”. Suddenly all the games are stupid and hard to win, damn these tomatoes giving me a loss streak. During every other moment though it’s approximately the same sentiment, just towards “potatoes” instead, about how the playerbase in this game is so stupid they don’t know that they shouldn’t do this or should do that, even at tier 10 (remember, tens to hundreds of hours!), haha look at all this free damage I’m farming oh my god my teammates are so stupid. You would think with all this haughtiness that they win basically every game they play, except those really bad ones you hear about?

foska34

The top 1% players win 56%~59% of their games. This is from the most reputable third-party source, so reputable it’s first party in all but name. It’s a little more difficult to say where the 99th percentile is but let’s just say it’s 46%.

This means, in other words, absolute gods at this game perform better than absolute imbeciles at a staggering rate of 1 out of 10 games.

If you ran 100m’s against cripples you’d expect to win every time, but people lose 4 out of 10 games against cripples in WOT and posture around like they’re world class. Sure, by definition they’re world class, but by that definition Bioshock Infinite is the best game ever because it got 10/10, and no one’s about to defend that[5]. Keep in mind the official claim for the matchmaking is that it only takes into account tier and type of tank – the player is not in consideration, i.e. it’s not like Chess where you play people of higher skill the better you get. In WOT you are always going to face the same people. If it is true that there’s a bunch of idiots running around, then every game should be an absolute slaughter with a few people doing basically all of the damage and taking all of the kills. Which is what you do actually see:

I saved the results of 60 battles and found that:

Why then is the distribution of player winrate a bell curve around 50%[6]? Every dog has its day, and they get that day every other day? I thought the only common factor throughout all a person’s games was their skill and their skill alone?

I once was a tomato at 47% winrate, and I honestly didn’t think much about the game at all. If a tank appeared I shot, if not I continued driving. Essentially a bot. Then I learned about the basic game mechanics of angling armor and detection/camouflage, and the mental skills of keeping watch on the minimap so I could go to where I was needed, playing tanks which would theoretically have higher game impact in ways which would give them higher impact, and now I’m at… 52%. What kind of game is it where someone who doesn’t get the game at all only wins 3% less than a coin flip? What does it mean when someone who does have a basic grasp of all those things only wins 2% more than a coin flip? Then we take actual bots into account, bots which don’t do anything but turn their turrets this and that way or various other non-contributing actions that only exist so the AFK detection doesn’t tag them, and they have winrates of… 38~45%. You can win 4 out of 10 games by doing nothing! Learning how the game works gets you 1 out of 10 more. Getting to world class 0.1% gets you another 1 out of 10 more.

How many 100m’s do you think you’d win against someone sitting still?

As for detailed stats, the two analyses I linked earlier from Lightquick and Greedy Goblin should be more than sufficient to convince anyone who hasn’t played WOT that it’s rigged. They were the first two links in this post; please read each of them fully if you have the time.

There are some speculations as to how the rigging works. We know it is rigged thanks to the existence of the patent, but which ones are used at any given time is not something we can show for sure because unlike real life where crime scenes and actions leave physical evidence, we can’t access server code. Other than manipulation of teammates – giving you lots of teammates with historically poor winrates and giving the other team people lots of historically high winrates has a conclusion which I hope I don’t need to spell out[7] – the other obvious routes are accuracy, module damage, penetration, and regular damage. People find it “funny” when their low-accuracy high-damage gun, when firing on the move (accuracy lowered) and turning their turret (accuracy lowered), is able to hit someone across the map. Officially the aiming circle is a normal distribution with some sigma value of where your shot is going to land: “usually near the center, but always inside the circle“. Even every bit of your circle is filled with an enemy tank, where you hit is of tantamount importance. If you hit their tracks but not their hull, you do no damage. If you hit their front or rear wheels, you do damage and they lose the ability to move. If you hit their ammorack, it can be a 1HKO.

How often does this happen? Enough that there’s a whole developer series on it.

Note that in more recent iterations, they no longer show the player’s point of view, and thus, we don’t know how big the aim circle is at time of firing.

Penetration is a value on shells to determine whether or not a shot will do damage, and both pen and damage are +/-25%. If you have 100 pen you could roll 75, 100 damage you could roll 125. This discrepancy grows as you go higher in tier because both pen and damage on higher tanks is larger. Suffice to say that this means a tank could be 125/75=66% more of a tank for one player than another… or for one game than another.

“But that doesn’t happen” then how do people go on loss streaks where they only win 25~30% of their games? Or worse? Remember, 40% is the “sit and do nothing” score, and you can’t kill your teammates every game either, and the game stops you from playing if you do that.

As for people who have played WOT and don’t believe that it’s rigged, there’s nothing to say to them. In the end, it comes down to a stranger’s word versus their experience. I’ve collected about 400 games’ worth of data, but that wouldn’t be enough for anyone. 1000 wouldn’t be enough, even 10,000 would likely not cut it, because most people who are vocal who still play this game have played it for 20,000+ battles, and if they don’t believe that there’s some funny shit going on after 2000+ hours, no one’s word short of maybe their mother’s can contradict what they’ve seen firsthand.

bsyajcaNot even their own.

uxf6otn

“Seal clubbing” is a phrase about experienced players who play low tiers to play against new players, and therefore raising their winrates. It is generally true that low tiers have newer players, newer players generally don’t know the game’s mechanics, and that it’s easier to get a higher winrate than in other tiers. Now if all this is true… how is it possible for low tiers to also be “extremely variable”? If it’s “extremely variable”, how does one “seal club” there? I have to presume it’s called “seal clubbing” and not “coin flipping” for a reason.

Even supposing it’s “extremely variable”, who cares? Isn’t the argument that the only common factor throughout all a person’s games was their skill and their skill alone?  IF it’s balanced and IF it’s totally random who gets put on whose team and IF this guy REALLY IS “seal clubbing” then he knows what he is doing and he’s up against people who don’t. How does he go from winning at a rate equal to the top 0.1% over hundreds of games to a rate equal to the bottom 1%? Note here that there’s no argument that Lightquick deliberately sabotaged himself. They don’t bring that up, and won’t ever bring that up, even though to anyone who hasn’t played WOT it’s the most obvious solution.

Why? Because they’ve had it happen to them too. It’s a dirty little secret. The reason why 3x/5x weekends are such a big deal is because they know it isn’t fair. If they actually believed what they said, that bonus weekends bring all the “shitter casuals” out, then it should be something to enjoy. It’s time to sealclub and statpad on tanks that they actually like, without the social stigma that it’s sealclubbing or statpadding! But instead they dread it because it doesn’t pad their stats. It does the opposite. They lose on bonus weekends. They can’t seem to stop losing. Shots don’t hit, if they hit they don’t pen, and oh boy they just got instagibbed by a low-accuracy tank while driving at full speed. They simultaneously believe that this will be the case, that it’s a near certainty, but the moment you confirm it explicitly and say “yeah it’s rigged”, they’ll just say it was bad RNG. They will never believe they were meant to lose.

There’s more doublespeak going on here; there’s more every time you look at any game which is obviously garbage but has vocal defenders anyways. I could explain how they’re wrong but this post is getting a little long so I won’t this time. Fortunately for me, Greedy Goblin’s analyses were written several years ago, and these commenters in particular have continued to play in the four years since they’ve made their point that “if the system was rigged like you say, i should be dropping to a 50% winrate!”. At the time of posting, Tiger313 had 22840 battles and Freshmeatiest had about 4500. They’ve played about 16,000 and 14,000 battles respectively since then, or approximately 1750 hours each. What do their winrates look like now?

Looks like they’ve been busy losing their faculties.

____________________________________

Bonus round! This game has been out since 2010~2011, the below videos are from ~2015.

Let’s not consider rigging or all the other complicated stuff.

Simplified: Would you play this game?

____________________________________

[7] You win!

[6] It’s 48% due to ties, not important; important is that it’s bell curved and not right-skewed, which is what the kill and damage data would suggest, as well as the general understanding that in PvP games it’s generally a few players that dominate the rest.

[5] Videogame journalists ranted and raved about it back then, but these days they wouldn’t. Isn’t that funny? Isn’t it funny how they’re giving Mass Effect Andromeda 7/10 now and defending how it’s a fair score? Everyone knew that 8/10 meant “garbage” and 10/10 was “meh” but now “7/10” is game of the year, and isn’t it great that we’re returning to a balance where 5/10 is the true average? Oh but Zelda still gets 10/10 no matter what. Isn’t that funny. Wait, “funny”? Hmm….

[4] Because at some point I believed it wasn’t rigged. Due to events and experience which don’t count as evidence under scientific definitions, I therefore magically and irrationally transformed from “unbiased” into “biased”.

[3] 1) You don’t know if it’s actually good or not. There’s a lot of stats about tanks on this game, but you don’t actually know what “250 penetration at tier 10 with x.xx reload, x.xx damage, x.xx accuracy” means until you play it. German tanks have better accuracy stats almost across the board; you won’t find anyone who agrees they mean anything significant. 2) It could be completely different by the time you play it, thanks to “balance patches” and “updates”. 3) Rigging.

[2] Yes I had my conclusion in mind beforehand. No I didn’t approach it with an unbiased mind, no this isn’t “scientific”, no I don’t care. Do you really think any “science” is done without significant confidence in results beforehand? Do you really believe they don’t skew results for their benefit i.e. prestige, employment, and grant (literally: Free) money? Even priests are associated with sin; somehow scientists aren’t?

[1] I’ll get to writing it. I had planned this one for a while though so I needed to get it out of the way first.

Advertisements

2 thoughts on “World of Tanks, rigging, and its defenders

  1. Pingback: [Reviews] (various) – All Else Is Halation

  2. Pingback: On Free Marketers’ Free Marketing of Free Markets – All Else Is Halation

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s