# Is it really random? Why always same S1 5* hereos? Confused

My first 5* was Azlar. Twice in a row. Never got Justice or Horghall. For me it’s random.

Llevo más de 300 entrenamientos y obtuve 6 quintus 5 1.01 y el otro 4.80, saque 5 John, 5 leónidas, sin embargo obakan no me salió. Isarnia y vivica una sola vez y hace menos de 1 mes, hago entrenamientos desde noviembre de 2018 y siempre con una TC mínimo, por momentos 2 y en algunos casos 3 en simultáneo. Y pueden pasar varios entrenamientos que no me otorga heroes legendarios. Este juego es paciencia y a veces eso te vuelve loco

1 Like

Yes, this is random. Like if you throw a dice you’re not guaranteed to get different number every time. It is unpleasant, especially taking in account that you don’t know what to do with precious yet useless duplicates, but it is random.

I understand what you mean, but something is strange.
In case of a dice, the odd of same number appears is 1/6 * 1/6.
In this case is 1.3% * 1.3%.

And I think it is more difficult to get same 1.3% heroes than two different 1.3% heroes.
Why is this happening? Not just one case. 3 Quintus, 3 Horghall, 3 Isarnia…
1.3% * 1.3* * 1.3% * (the odd of the same one appearing among those many heroes.)
I don’t know. It is strange.

1 Like

Like everyone said, another way to see it is like when a 30% emblemed warrior will ress 6/7/8 times back to back. It doesnt mean is broken is just the odds are really low but possible.

My first 3 5* heroes from my TC20 camps were Isarnia (at the time I only had Justice). It seriously pissed me off. Nowadays my Tc20 have produced 12 5* heroes (5 of them Isarnia) but I also got others 5* now, so getting duplicates is a much more normal occurance.

2 Likes

Australian lawmakers consulted with experts who suggested a different way of showing customers summons possibilities

The short answer is: this is how an algorithm of random number generation works, but you also may call it luck.
My guess (just a guess based on experience in software development) is that portal works in 3 steps:

1. A random number is generated for a color (1 to 5)
2. Another number to determine a hero rarity. This random number is defined by declared portal probabilities.
3. Finally a 3d number is generated to pick a hero from a pool of rarity defined at step 2.

In a step 3, if you, say, have 4 different purple 5*, it is very likely that you’ll get the same one.

1 Like

Hey @Gryphonknight, you have been granted a very cool and fitting title!

4 Likes

The results are random. They would be suspiciously less random if you didn’t get dupes of one and none of another.

From TC20 I have over the almost 2 years of running pulled 3x Joon, 3X Vivica, 3X Domitia – and zero of Magni, Leonidas or Thorne (or Boldtusk for that matter)

1 Like

We know from red 1* troops bug fix, and red heroes data, that training camps select

rarity
color
individual hero

I expect micro transactions to use the same program code.

See, what happened to that person is mathematically much more unlikely than what happened to you (I suppose you didn’t pull all of them in a row but among many other pulls).
In this case of course the outcome was more in favour of the person who did the summons than in your case.

That’s how it is with randomness. Even very unlikely things can happen… though usually rather rarely. Just because the unlikely event isn’t in your favour, doesn’t mean it’s not random.
If you would have pulled lets say five different five stars in a ten pull, I doubt you would question if it was random or not… you would just consider yourself being extremely lucky.

I wish you more luck in future with your summons. But you’re results really don’t prove anything…

1 Like

So let’s explain 3 Horghalls. (My alliance members like to call it Threeghall.
Many examples are there. Fobakan, Thsarnia, Twostice, Fourne, Folena, etc.)

In a step 3, always only Horghall was picked. Yeah, that was just a bad luck.
No Elkanen, No Kadilen, No Lianna. Only Horghall three times!!!

No offense to you. I’m just struggling to accept algorithm you explained.

Anyway, thanks again.

May be. I made my assumption based on the fact that color is shown first and then a hero comes. But, in fact, order of those 3 steps doesn’t really matter. What does matter is that for each step an independent random number is produced.

1 Like

Unfortunately, you are right. In a step 3 computer says: Random Number generator, give me an integer between 0 and 3 (4 distinct values).
0 - Lianna,
2 - Elkanen,
3 - Horghall

There is nothing strange in getting a value of 3 three times in a row. Other person could get 3 of other kind (say, Kadilen).

Another example is loot. You can get 3 compasses in the same day, or you may not see any in weeks.

There are other experiences for me.

I have 2 accounts. Main one has 20 Orbs (used many. 20 left.)
Strangely, Orbs are very common there and the other 3 star rare AMs are not that common.

And my 2nd account has 18 warm capes and it’s funny that there is no orb there. (used some orbs
already, but it was not enough always.) Warm capes are very common in this 2nd account.
All kinds of chests shows warm capes very frequently.

And another one.
About a month ago, I used 9 troop tokens and I got five 4* troops. (4 4* troops in a row, and 1 3*, and then again 4* troop.) After that, about 20 summons show no 4* troops.

The question I’ve got here is like that.
Sometimes there’s a tendency that some specific color, or stars of AMs, troops, heroes are more often than others. I feel like it.

So I’m struggling to call this algorithm random.
If there’s some kind of tendency or something like that, can we call it “purely” random?

Random, to be purely random, includes such tendencies. Otherwise there is an overall pattern and it’s not random.

1 Like

I would like to share that the situation you (OP) describe does not even represent bad luck.
Having a few heroes with three duplicates and many with two duplicates while there are still many heroes who didn’t show up has significantly higher probability than having an almost perfect roll where you got almost all heroes with almost no duplicates.

Look at it this way: as soon as you say “the last three pulls were X, no way the next can also be X” you are putting a limit on the pull and thus reducing the randomness. Each pull is an independent event. For true random results there can be no influence from previous results.

Hi all…My question is, ok for random summon, but random for each player numbers or for total players on game?..maybe that’s the key if I’m not wrong.
And other side, we have two ways to change %, change number of pulls and change number of results/heroes, so, heroes are the same and result will be the same, more if you put another coefficient, randomness. That’s impossible change numbers in a sure summon results.