Tag Archives: World of Tanks

Why I’ve Quit World of Tanks

I realised something today: World of Tanks is a casino with nothing to win except a momentary sense of accomplishment.

I’ve loved tanks since I was a little boy making plastic models of WWII tanks. As I grew up that turned into a love of wargaming and all its associated modelling requirements. I would lovingly select, purchase, assemble, and paint all the tanks, soldiers, terrain, and other items required to field my own little army against fellow enthusiasts. If I’d never moved house there would still be a desk piled high with paints, brushes, glues, and tools but I did and so all those things had to be tidied away. A couple of moves later and with a pressing need for money it was time to sell all my wargaming collection on eBay. I thought I was done with it.

Years later my wargaming passion has turned into an online gaming passion. I’ve played World of Warcraft of course, who hasn’t, but I’ve also played sundry other games but never found something that drew me like my teenage obsession with tanks until I chanced upon World of Tanks. World of Tanks (WoT for short) is purportedly an MMO focused around WWII tank battles and so it is at first appearance.

You play the game by entering short PVP team battles, fifteen tanks to each side on a small thousand meter square map. Now that’s quite odd for tank warfare which is mostly conducted over significantly larger ranges and areas. This is explained by the designers as the game being focused on arcade action and fun. Sounds like a great idea.

However it’s not quite that simple. Each shot has a very large random outcome as the designers have applied a strong random factor to every aspect. Assuming you have aimed correctly when you press the fire button your shell will disperse randomly within the aiming circle. It’s meant to be on a bell curve but from personal observation it’s just wild, you cannot aim and expect the same result from the same gun consistently. Assuming you actually hit the enemy tank there’s then a random penetration roll that increases or decreases your shell’s penetration by twenty five percent. After that the game works out what damage you do to the enemy tank again with a twenty five percent randomisation of your standard damage. Add this all up and each shot you fire has a wildly different outcome. Multiply that out by your fifteen team members versus the fifteen members of the enemy team and the outcome is entirely unpredictable. Except that it isn’t. Somehow the random factor seems to favour you on some days and be against you on others. Almost is if the dice were loaded.

Now it’s certainly possible to overcome all this randomness, the designers of the game have given you plenty of opportunity to spend buckets of money on microtransactions to enhance your game. Premium subscriptions to let you progress faster, premium tanks to let you grind out experience and cash for new tanks, premium ammo to let you penetrate better, premium crews to utilise your tank at one hundred percent of its ability instead of the standard fifty percent. The list goes on and on. By the time you’ve got into the game and run into the point where you have to start grinding the game to progress you’ll either have given up in frustration or be very ready to start buying your way forward.

Once you’re established in WoT you’ll begin to perceive the underlying problem and eventually realise why it’s bad for you. Other players have taken the readily available statistics from the game and done some simple analysis. The average player has a win rate (games won as compared to games lost) of forty eight percent. The worst players lie around thirty percent and the best no better than seventy percent. The shape of the bell curve for this distribution of win rates has very steep sides meaning that it is very difficult to change your win rate once you have a lot of games under your belt. What this means is that each game you play has roughly a one in three chance of being lost no matter what you do, conversely another one in three will be won without you lifting a finger. It’s only the remaining one in three games where you will stand a chance of influencing the game. Provided that you are in a tank that can sufficiently influence the outcome and that the random rolls come your way. For the average player this is maybe one game in ten where they can have an effect on the outcome. What’s worse is that thanks to the pretty much guaranteed one win in three even the most useless players, AFKers, and bots will always succeed in the long run, simply by virtue of playing a large enough number of games which means that there are always thoroughly useless players on each team and it only takes having one more useless player on your side to lose the game.

So the game is pretty much a gamble, you take your tank, place your bet (in the form of the ammo and repair costs you will spend), and spin the wheel. On a good day you guess the outcome, otherwise it’s all down to the randomness in the system. But you never know which one it is. If you do well you’ll feel god about your accomplishments, do badly, or get on a losing streak and you will be upset, blaming every other player under the sun, and having a thoroughly bad day.

This is where it gets nasty. Unless you pay money to overcome the randomness in the system you will have more bad days than good days and even the good days will have plenty of bad spots in them. Unlike most quest based MMOs where you can feel as if you’ve accomplished something each time you hand in a quest, WoT is only rewarding for the one game in ten where you actually personally accomplished something.

So now it’s possible to see why quitting World of Tanks is the only sensible course. It may not cost much to play a little, a few quid here and there, but paying for World of Tanks is like pissing into the wind – you have no chance of success. Unlike a casino where the odds are regulated World of Tanks has no odds you always lose all your money and you only stand a tiny chance of gaining a small sense of accomplishment. I’ve always been against gambling, which may be a surprising attitude for a game player, but to me gambling is just handing your money over to a bunch of criminals whereas playing a game is pitting your skills against your fellow enthusiasts. World of Tanks flatlines any skill by replacing it with loaded dice and then screws you over for a shot at the little prize.

So if you see me installing the game or talking about it again just remind me gently of why World of Tanks is an unlicensed casino with loaded dice.

Is World of Tanks Playing with Loaded Dice?

I’m getting the impression that somewhere in the guts of World of Tanks there’s some loaded dice. I’ve been in too many battles where one side or the other just disintegrates under the least amount of pressure and I think it’s down to two possibilities:

One – the random number generation for dispersion, penetration, and damage has a factor that is rolled per battle and side not per shot

I think there’s a lot of people who have experienced battles where they and their side have either blown through the opposition with little effort or been unable to dent them even with expert marksmanship.

This could be due to a factor, think of it as a bonus or penalty, which is applied randomly to each team at the beginning of the battle, which modifies the random dispersion, penetration, and damage rolls. Effectively giving one random side a massive advantage most of the time.

i.e. Team Green (TG) and Team Red (TR) both have a random factor applied to their RNG rolls at the beginning of the battle. Let’s say that the number can be either plus or minus 0, 5, or 10% and the RNG adds a further +/- 0-15% per shot.

– In the first battle Team Green gets +10% and Team Red -10% resulting in Team Green wiping out Team Red very quickly.
– In the second battle Team Green gets -5% and Team Red gets +5% resulting in a close win for team red.
– In the third battle both teams get a +0% bonus resulting in a close battle that goes down to the wire and where the skilled players actually make a difference.

I have a rule of thumb when playing arty, if the first couple of shots go wide then so will all the others. It appears that the amount you deviate by (but not the direction) is fixed for the match. Those matches where the first shot hits on the nail are going to be the ones where you rule as arty, the ones where the first shot hits the edge of the circle aren’t going to be so good…

Two – matchmaking groups stock tanks and/or 50% crews together

We’ve all seen the strange way that the matchmaking system places groups of similar tanks on one side of the battle. It isn’t properly randomised which leads to the theory that there’s other strange coding going on in the matchmaking. If one of those is to group together stock tanks (i.e. not fully upgraded) or 50% crews (i.e. any crews less than 100%) then that again gives one side an enormous advantage.

Even if the matchmaker doesn’t factor in stock tanks, crew skill, or even player ‘skill’/win rate/premium account, simply not balancing the teams evenly can result in a significant advantage for one side over the other. 3 KV-1 tanks as top tanks vs 3 Shermans is an example I’ve seen a couple of times. Allowing tier I tanks to platoon with tier X tanks screws over the side that gets the tier I.The weighting system used in the matchmaker doesn’t work to provide balanced teams in a significant number of battles. This has just the same effect and is just as bad as fixing the random rolls for the match.

Loaded Dice

Nobody wants to play a game where one side is given loaded dice. It’s obvious to many players that something strange is going on in the game, and, as far as I’m concerned, it’s up to the developers to assure us that their game design is genuinely randomised to give all players a fair chance.

Paranoid Conspiracy Theories

Please note that I don’t think there’s a conspiracy and Wargaming aren’t ‘out to get you’ by rigging your matches to lose when you creep above a 50% win rate. It’s all down to the badly coded Team Win Generator accidentally fixing 70% of matches with one sided bonuses and/or bad matchmaking. It explains absolutely why you can only have a significant effect one battle in ten despite playing the same every time. It also makes building any real skill ten times as difficult because for every time you repeat the same action you don’t get the same result. They don’t need to fix matches for individual players as fixing matches randomly accomplishes the same ends. This is Wargaming’s version of balance: If everyone gets screwed equally then the game is balanced

Testing the Hypothesis

For anyone who wants to run their own test what needs to be looked for is a correlation between winning a match and having ‘good luck’ with regards to hits/penetrations/criticals and losing a match with ‘bad luck’ on hits, penetrations, and 0 damage criticals.

i.e. record your hits and misses, your bounces, and your criticals. For arty record your deviation from the center of the circle. and record the outcome (i.e. 15-0 loss, 2-10 win, etc) and how fast it happened. Even simply recording hits/misses and wins losses would help. This is what I’ll be doing.

If the Team Win Generator is setting a single bonus/penalty at the beginning of a match then the majority of shots will perform badly in some matches, well in others, and ‘average’ in a small number.

Is World of Tanks Playing with Loaded Dice?

I’m getting the impression that somewhere in the guts of World of Tanks there’s some loaded dice. I’ve been in too many battles where one side or the other just disintegrates under the least amount of pressure and I think it’s down to two possibilities:

One – the random number generation for dispersion, penetration, and damage has a factor that is rolled per battle and side not per shot

I think there’s a lot of people who have experienced battles where they and their side have either blown through the opposition with little effort or been unable to dent them even with expert marksmanship.

This could be due to a factor, think of it as a bonus or penalty, which is applied randomly to each team at the beginning of the battle, which modifies the random dispersion, penetration, and damage rolls. Effectively giving one random side a massive advantage most of the time.

i.e. Team Green (TG) and Team Red (TR) both have a random factor applied to their RNG rolls at the beginning of the battle. Let’s say that the number can be either plus or minus 0, 5, or 10% and the RNG adds a further +/- 0-15% per shot.

– In the first battle Team Green gets +10% and Team Red -10% resulting in Team Green wiping out Team Red very quickly.
– In the second battle Team Green gets -5% and Team Red gets +5% resulting in a close win for team red.
– In the third battle both teams get a +0% bonus resulting in a close battle that goes down to the wire and where the skilled players actually make a difference.

I have a rule of thumb when playing arty, if the first couple of shots go wide then so will all the others. It appears that the amount you deviate by (but not the direction) is fixed for the match. Those matches where the first shot hits on the nail are going to be the ones where you rule as arty, the ones where the first shot hits the edge of the circle aren’t going to be so good…

Two – matchmaking groups stock tanks and/or 50% crews together

We’ve all seen the strange way that the matchmaking system places groups of similar tanks on one side of the battle. It isn’t properly randomised which leads to the theory that there’s other strange coding going on in the matchmaking. If one of those is to group together stock tanks (i.e. not fully upgraded) or 50% crews (i.e. any crews less than 100%) then that again gives one side an enormous advantage.

Even if the matchmaker doesn’t factor in stock tanks, crew skill, or even player ‘skill’/win rate/premium account, simply not balancing the teams evenly can result in a significant advantage for one side over the other. 3 KV-1 tanks as top tanks vs 3 Shermans is an example I’ve seen a couple of times. Allowing tier I tanks to platoon with tier X tanks screws over the side that gets the tier I.The weighting system used in the matchmaker doesn’t work to provide balanced teams in a significant number of battles. This has just the same effect and is just as bad as fixing the random rolls for the match.

Loaded Dice

Nobody wants to play a game where one side is given loaded dice. It’s obvious to many players that something strange is going on in the game, and, as far as I’m concerned, it’s up to the developers to assure us that their game design is genuinely randomised to give all players a fair chance.

Paranoid Conspiracy Theories

Please note that I don’t think there’s a conspiracy and Wargaming aren’t ‘out to get you’ by rigging your matches to lose when you creep above a 50% win rate. It’s all down to the badly coded Team Win Generator accidentally fixing 70% of matches with one sided bonuses and/or bad matchmaking. It explains absolutely why you can only have a significant effect one battle in ten despite playing the same every time. It also makes building any real skill ten times as difficult because for every time you repeat the same action you don’t get the same result. They don’t need to fix matches for individual players as fixing matches randomly accomplishes the same ends. This is Wargaming’s version of balance: If everyone gets screwed equally then the game is balanced

Testing the Hypothesis

For anyone who wants to run their own test what needs to be looked for is a correlation between winning a match and having ‘good luck’ with regards to hits/penetrations/criticals and losing a match with ‘bad luck’ on hits, penetrations, and 0 damage criticals.

i.e. record your hits and misses, your bounces, and your criticals. For arty record your deviation from the center of the circle. and record the outcome (i.e. 15-0 loss, 2-10 win, etc) and how fast it happened. Even simply recording hits/misses and wins losses would help. This is what I’ll be doing.

If the Team Win Generator is setting a single bonus/penalty at the beginning of a match then the majority of shots will perform badly in some matches, well in others, and ‘average’ in a small number.