Raids are unfair, but not for the reasons you think


#85

I average 4 to 5 mats per month from platinum chests. That’s pretty regularly compared to 0 from gold.


#86

Another really good post with overlong replies which give Me cross eyes. But the essentials are all at the top anyway so my observations don’t really matter as I can’t count but use a feel your way and a little bit of study.
Raiding in early days is practice. It’s when you should be reading the wikis, forum and peers support. Then armed with what you have gained go out hit a few win some and wake the next day to fund all your gains gone. After a bit of this I gave up but gradually they drew me back in. Then I started to learn about combos. And as my four stars rolled in I learned the value of mana eaters , hit all, snipers, healers and boosters. Long term luck gave me a pretty good 4 star team but as I grew my team just wasn’t returning the same as when it was built. Fine I had others ready to try. .
So I get to that point when I find that I get attacked 5or 10 times a night. Cool revenges are so much sweeter than been revenged.
Then I realized it was just a computer chess game which scales to your experience. Some sort of mix of cups and hero strength vs what you use.
As cups don’t get you anything except raid and I will always be pitched against foes who I have a theoretical chance against. Then why worry?
Check whose raided you and if there is a profit in it …raid them back. Good fun if you find a bully. So I have stacked my defence. Hoping to catch bullies and find its taking most 2 shots to win. They are usually valueless to revenge but not good when I try.
So I am sort of stuck. Oh well it’s always practice so look at the heroes, strengths and combos . Get my win or lose and hope for a decent hero chest.
Now that I should gripe about except for that one time I got an epic token that gave me something and Gravedigger.
So I am got. I will probably never see a token like that again. Maybe I will get a run.
Nothing in raids actually hurts me. And the chest is a wildcard.
As I said a computer chess game with a matching system which works even though it pretends not to. A practice learning tool with a chest that can change your game life.
So what’s the point of arguing with it? Raid or don’t. But if you do. your play will get better than if you don’t.


#87

A simple observation: remember how a joke is never funny if needs to be explained? Likewise, when rules have to be explained, before people can understand that they’re not unfair (for arguments sake, I don’t agree), they’re badly flawed. And there are WAY to many people who have problems with raids. I hear it everywhere: it’s a nice game, but I hate the raids system.
If it was a game on itself, it would sink into oblivion in no time.


#88

The system is definitely flawed, but that doesn’t make it unfair. I would prefer to have trophies be more permanent


#89

Raids are unfair, because defending heroes have their ultimates dealing 20% more damage than attacking heroes, with equal atk and def.
This is not described anywhere, so just unfair.


#90

Your heroes also deal 20% more damage when you are defending. Same rules apply to everyone. That’s fair.


#91

That reasoning is actually not completely valid. Yes the same rules apply, but this rule favours the inactive players over the active ones…


#92

That rule is in place to attempt to level the playing field, since the attacker has all the advantages, from whether to attack or reroll to hero selection to move selection to timing of special skills to target selection.


#93

Except that, as per usual, if you double up, as if by magic, that colour tile miraculously disappears from the heavily manipulated board.

You can decide to do all those wonderful things you suggest, double on tank, attack one side, ghost etc. You’re forgetting the board; it decides.

Many a time have I lost an initial raid only to hit rematch and win convincingly.

Raids are skewed neither way. Get a crap board, lose, get a good board, win.


#94

Get a crap board and play it well, you can sometimes pull out a win. Spray and pray will guarantee you a loss


#95

If you can play it well and pull out a win then it’s hardly a crap board is it …?

A crap board is one that no matter how hard you try to play it you can’t pull out anything. That’s what makes it crap.


#96

Based on your definition crappy boards are not that frequent. That is, if crappy boards happen with probability p, then you lose with probability p+q, since you lose all the crappy boards plus some other not optimally played boards.
I know many players that win at least 70% of matches, which means that crappy boards are at least less frequent than 30%, and this is just an upper bound, the real probability will be pretty lower.
I think that when you observe lack of tiles for doubled/tripled heroes you’re only noticing more things that strike the most your attention. Just like when you notice a street lamp switching off exactly when you pass, but you don’t notice the countless lamps that stay switched on. Nobody is turning off the light just for you


#97

30%, near enough 1 in 3 is pretty frequent.

Thing is, when I go rainbow, I very rarely get 1 in 3 crappy boards. In fact it’s not even 1 in 10. I might not win every one, but the reason for loss is not crappy boards.

Yes, we see what we look for, so I look for tile distribution EVERY board. Why, so I can note how many crappy boards I get per team type: rainbow, 221,2111,32, 41, 5 etc. Do it yourself and see how often as soon as you double, treble or more the tile distribution biases.


#98

I said at least 70% (i.e. greater than 70%), and from the remaining probability (lower than 30%) you have to subtract the number of fights you lose due to not optimal playing, so in the end crappy boards will be pretty lower than 30%, but I’m repeating myself.

If you lose whenever you stack 3 blue heroes against Ares and you haven’t at least 8/9 blue tiles over him, considering that a crappy board, then I agree, crappy boards are very frequent. But on average you should have 7 tiles per color and if you look carefully you’ll notice that boards that a first sight seem very bad, actually have more or less 7 tiles of the color you are looking for, they are just scattered away. But that’s not biased, that’s exactly what random board means.

And even when the board is below these average values it doesn’t mean you can’t play it smart or take some risks to change your luck, here’s an example of one fight during one of my last AW.

Color stack 3g heroes + 2p heroes against y tank.

Starting board 6 green (< 7), 4 purple (< 7), 10 yellow (> 7) tiles:

epilogue:

Consider also that I’m fighting with a team much weaker than the defending one (2 5* + 3 4* against 5 5*) and the d-team has the war aid.
And even if in this case I have stacked 2 purple against a yellow tank and I have only 4 purples and 10 yellows, I still don’t think the boards are biased, this was just an unlucky board (notice, not a crappy one cause I won!).

One way you have to convince yourself is doing what I suggested above in this message: count the tiles. More often than not a board you consider bad has 7+ of the tiles you are looking for, they are just spread (== random)


#99

Okay, let me give you an example, stacked 5 Red against a Green Goblin King Titan. Opening board, 1 red tile in the bottom rhs corner, one in the bottom lhs corner.

I’m well aware of the averages, 35 tiles, 5 heroes, in a perfect World 7 of each every time. I look for 6-8 as an acceptable distribution. But the number of times I have had 2 or less in far greater than RNG, probability or odds would/should dictate.

But it can happen with rainbow as well, as shown below, no green tiles…Funny as they’re strong against blue.


#100

That’s not the worst board possible by a long chalk, instant Yellow diamond clears Yellows out and Reds are pretty kindly placed to clear.


#101

Again I think you are concentrating on single examples, and probability doesn’t absolutely suggest that only the range 6-8 is acceptable if the average is 7: assuming all boards can show up (which is not true, since about 80% of them already have a match within and are discarded, but I want to keep the discussion not too technical) the probability of having 0 tiles of 1 color is 4 in 10000 fights (not even close to 0) and the probability of having only 1 tile of a given color is 0.35%, which means it should show up about once every 280 fights! Write “0.2 * 35 * (0.8^34)” in your google search bar if you don’t believe me. It goes without saying that the probability of having 2, 3, 4 and 5 tiles are farther from a rare event than you think.

In a day I play at least 16 levels to fill monster chests, 16 raids, 6 titan fights (not counting AW, events, further farming of regular stages), for a total of 38+ fights which adds up to 266 fights in a week. It means I can well expect 1 initial board with only 1 tile of a color each week (and again this is an upper bound, I play many more fights in a week).
But if I screenshot that 1 board and post it here I’m not really fair, I should screenshot also the other 400ish initial boards and count the tiles on each of them.

EDIT: Just to be clear. I gave the probability formula for boards with 1 tile of a specific color, i.e. the unlucky one. The probability for a board with 1 tile of ANY color is about 5 times higher


#102

Really great summary.


#103

Yes, that’s all very well and agreeable but not on point.

My point here is the correlation between stacked colour heroes and skewed boards. I am saying that such teams attract a far higher frequency of skewed boards than rainbow teams do.

I sometimes wonder if number of tiles is inversely proportional to number of heroes of that colour.


#104

Actually I think it is. I’m saying that

may well be due to a false perception, because as human beings we tend to focus on events that strike our attention and overlook the whole picture, and because things that we think should never happen are well within the realm of likely events, so I suggest to actually make a statistic counting the tiles over a large set of fights to have an answer to your question.