Raids are unfair, but not for the reasons you think


Merged thread

Please read back for wisdom from other players on your topic. :slight_smile:


I’m more wondering why I have to reroll 10x in a row in order to avoid Guinevere. The fact that she’s so prevalent in the highest bracket speaks volumes as to the fact that she is not only OP, but doesn’t have a good counter measure.


Counter measures exists but the best are not always available (G. Panther, Merlin, Gretel, Hansel, Zeline) and for raise the winning chance we are likely forced to run as many dark heroes as we can.

If we could do a fusion between Kashhrek and Li Xiu then the ladder would be filled by them :persevere:


Double up on fast purple hitters and try to work the flanks to charge them. Sartana and Tibs work nicely


Im totally aware how the system works. But the main problem is tiles, they are not totally randomized, I often get 4-5 turns of wins followed by 4 to 10 turns of lost. This is basically because tiles come well and then suck not depending on the team I faced most of the time.
Hence, I advise dev team to randomize tiles, please remove all the rules you embedded in tile algorithm. Because having a biased system for tiles coming up, you basically distorting the game’s nature. If you need help or discuss more, im open to give free consultancy as being a startup founder. I really learned so much from the game that how to optimize revenue with lots of tricks but come on, tiles should be randomized in a way that nature should not be distorted. Hope you understand my concerns


Every time I hear this BS about tiles not being random, I say the same thing. If you want to claim that starting boards are not random, collect the data and prove it. Collect screenshots of your next 500 starting boards and analyze the number of each color tile. Post your results here. If you don’t want to be bothered with proving your assertion, then don’t expect anyone to believe you. Confirmation bias will get you every time.


dev team should have all the data to analyse. Further, It is not the opening tiles but also the coming tiles. So that is impractical for a user to screenshot everytime, you sound funny sorry


If you’re attacking with a double red team, for example, and you play all the red tiles, guess what? You’re going to wind up with fewer red tiles. Simple math. I’ll walk you through it.

  1. There are 5 color tiles and 35 board spaces. That means that on average, your starting boards will have 7 of each color. On average. That’s important. I’m working strictly with averages here.

  2. Move 1: you match 3 red tiles because red is your strongest color against a green tank. That means there will be 3 new tiles on the board after that move, and 4 other red tiles remaining.

  3. Each of those 3 tiles has a 20% chance to be each color. So working with averages again, you get 0.6 red tiles back for a total of 4.6 red tiles. You also have 7.6 tiles of each other color.

  4. Move 2: you match 3 yellow tiles and it starts a cascade. At the end of the cascade, you have used 3 yellow, 6 blue, 4 purple, 3 red and 3 green tiles, for a total of 19 new tiles coming on to the board. Your remaining tile count before new ones is 4.6 yellow, 4.6 green, 1.6 blue, 1.6 red, and 3.6 purple.

  5. Your new tiles come onto the board. With 19 new tiles, you’ll get 4.8 of each color. Your new totals are 9.4 yellow, 9.4 green, 6.4 blue, 6.4 red, and 8.4 purple.

  6. Move 3: you match 3 red again, trying to take out the tank. You’re now down to 3.4 red.

  7. 3 new tiles come in, or 0.6 of each color. You now have 10 yellow, 10 green, 7 blue, 9 purple, and only 4 red. OH NO! THE BOARD MUST BE BIASED! I ONLY HAVE 4 RED TILES! IT’S A CONSPIRACY!!!


Dear Anne,
As I said, Im totally aware of system. Im a data scientist and a dev too. You had not need to explain and make me comprehend the model, but anyway thanks for the walk through some people may need this, it might be a good contribution to community.
Second, thats a gut feeling that i wrote, based on empirical observations and being a data scientist.
Third, this was written because if some dev team member sees (and if they embed rules to distort randomization) they may talk about it whether to remodel it or not. other than that, If it is already randomized and this was a false feeling coming from a user, they probably do not care, and move forward. That’s all.
Hope good vibes to you,


I’m not a scientist.

Sometimes I lose raids.

More often I win them.

I don’t like when I lose 50 cups and can only win 10 back, but mostly I attack 40+ and realise you can’t have both.

Raids are fun. I love them. And there’s about three people in my alliance who we just try and have more cups than the other for bragging rights. It adds a cool layer of fun and banter to an an already fun game.


Devs have publicly stated that the boards are fair and tile drops are random. No evidence has ever been advanced on this forum that disproves their statement.


NPKY is good people. She has helped me immensely in raiding and events. Last night I made it to 2392 cups with my little 4 star team. She’s a helpful person, you’re out of line.


My opinion vs your opinion. Everyone is welcome to their own. I stated mine and did not report any contact to it. I believe the article to be completely false from what I see in the game. Sorry if your feelings are hurt.

[Edited by Rook. Please follow Forum Rules.]


True, but if they share it, I’m sure it would be given a little more credibility if it’s backed up with some research rather than thinly veiled accusations.


As the original post had proof? I see statements made by someone. I see no proof. I do see what happens every time I play. For one the AI is far from random. It targets whomever had the Mana built every time for attacks. Not sure what to do to prove that to your. Just as the original post proved nothing to me


Leave it, JB. It isn’t easy for some people to accept that they could be doing something better. It’s okay


Of course the AI is not random - no one has ever claimed that it is. It is stupid, however, and it certainly does not target “whomever had the Mana built every time for attacks”. You can’t prove that it does this, because it doesn’t. It’s very easy, however, to prove that you are wrong:

Here you’ll see I have Alby fully charged, with Leo getting close.

Now let’s charge up Gormek and see whom he chooses to hit with his special.

Oh! It seems he chose not to hit the hero with the special ready to fire, or even the hero with the second most mana. He hit the three heroes with the least mana.

I do not understand where you got this silly idea from.


My lady, as always, your lightest touch commands obedience.



The idea came from the fact that they do attack, at least in my case the one with the full Mana. In every instance it is wu that is attached the minute his Mana is full


It can seem this way much of the time, but those instances where Wu gets sniped right as his mana fills are much more memorable because they hugely affect the outcome of the battle. You don’t tend to remember the times when Wu successfully charged and used his special as expected, or when the same sniper hits Richard at full health with no effect on the overall battle because Sabina was ready to heal the damage anyway. That’s just the way the human mind works: it is biased toward remembering details that greatly affect outcomes and discarding details that least affect outcomes.

But to understand the underlying processes, you need to pay attention to all the details. The best way to do this is to record a few of your raids, then watch them back a few times. Note which heroes are targeted and what their mana/health levels were. All the evidence I’ve seen so far suggests that the AI is random, but if you find any distinct patterns feel free to let the rest of us know so that we can investigate them too.