Extremely low chance on drawing 5* event heroes

More recent articles quote higher numbers, but I’m not looking for them just now.

Edit: Just from memory, the more recent daily visitors quoted were something like 2 million, but memory is unreliable and I’m a flawed vessel, so don’t quote me on that.

2 Likes

Downloads does not mean people playing. I myself download many games, and if I do not like them, just delete. Just a wander around outside alliances will find how many did exactly that. The data also coincides with a training alliance, where about 40% of newbies stop playing after a couple of days

Perhaps you didn’t bother to read the article I linked to. A million daily players. Back in February.

1 Like

i did read that, but besides looking like a standard pr notice, it got me wondering where are the 34 thousand active alliances housing all this people, and why are there no recruits, and why they choose not to participate in events

Allow me to answer your questions with some questions:

  • How many alliances do you think there are?

  • How many players do you think are in a typical alliance?

  • How many players do you think are not in alliances?

  • What do you think happens to the availability of recruits as the number of alliances increases?

  • What proportion of players do you think bother to participate in events?

  • Of those, what proportion do you think bother to participate in the ‘rare’ tier?

1 Like

I wanted to look at this another way, How I (a slim spender, or “mostly F2P”) would see it, divided by 3/4/5* cards. Based on the numbers given, the total cards are 168(+3)=171:

5* - 1(3) - 0.58% (2.33%)*
4* - 46 - 26.90%
3* - 121 - 70.76%

*including the 3 Gravemakers

So compared to the expected rate of the TC20, the 5* is low, the 4* is higher, and 3* is slightly lower. Hmmm…

2 Likes

And what are your conclusions from these numbers?

Same as my usual conclusion; I’m a fellow player with limited funds:

I play TC20 first, and single draws as I get them second. If I had moolah, I might try the 10x rolls, but I’m not rolling in dough. I’ve done the same (and advocated the same) since I started playing last April.

Someone asked about percentages. I don’t know the total percentage, but here’s SILEX’s.

1 Like

I think our (admittedly imprecise and unreliable) estimate of the chance of pulling a 5* hero from an elemental summons was 1-3%. So without the HOTM that’s a somewhat disappointing return for 5* heroes, but once you include the HOTM it starts looking pretty good. The 4* return is very good.

And yes, as ever, TC20 is just better.

Without the HOTM, his percentage was less than one percent for 5*. Ugh! But no HOTM in TC20. weighs both in mind

1 Like

You guys keep telling that numbers like they was for sure. No ones from SG has never say anything on it.

From what we know, odds may even change in different days, on different events, basing on your level, your spending hystory or even your deck.

It all goes with you. If you have faith, these are the numbers.

If you don’t have (like me) because i tend not trust people hiding something from me, that numbers means anything.

1 Like

We have no reason to suppose that odds change on different days, for different events, based on your level, your spending history or your deck. Is it possible? Yes. But it’s also possible that last night goblins replaced the moon with a giant turnip. I’d be very silly to assume that was the case, without some evidence of it.

The one large scale study we coordinated through the forum showed convergence towards a common underlying probability over a large sample size when it came to the chance of training a 5* in TC20. (We can’t say with certainty what it was, but it looked a bit like 5%.) That was exactly what we would have expected to see, if SG simply applied RNG to an underlying probability of training a hero of a particular level.

So if you have some evidence to suggest that summoning (and training) chances vary depending on a wild range of variables, please go ahead and share it. We’ll adjust our Bayesian priors accordingly. In the absence of such evidence, however, please don’t just claim that the moon is a turnip. I’m not going to believe it.

1 Like

Why it’s silly assume that without evidence… but it’s good assume the opposite without evidence the same?
That numbers are few, from only a very small part of the players (always the same people) and not surely free from some mistakes or variable.

So i don’t see anywhere a reason to believe in that numbers other then “oh, well, i don’t have any other things to believe!”

Soooo… No thanks.
Oh, and i don’t have to prove anything because i don’t want to persuade anyone in my reasoning.

If anything, prove to me that these numbers are reliable :slight_smile:

1 Like

I see. I am agree and someone talk to my opinions this is no sent.

Because:

  1. The latter requires fewer assumptions than the former. We even have a name for the application of this principle: Occam’s Razor.

  2. We have evidence of the latter. Not great evidence and not a lot of it, but it’s the best evidence we have.

These are valid criticisms. The appropriate response to them is not to ignore the evidence, it’s to collect better evidence.

Here’s a reason: that’s how science works. We use Occam’s Razor to develop our hypothesis, collect data to try to disprove that hypothesis, but continue to treat it as fact until we have a better alternative, itself supported with data. We have a name for this process, too: the Scientific Method.

The chances are that the moon still isn’t a turnip, but I can’t prove it to you and you will believe what you choose to. If you tell me it’s a turnip, though, I’ll tell you that you’re being silly.

2 Likes

Great thing to see the numbers, and I thank the statistics that it was not even worse.

I would like to give you one more information.
I was still spending and throwing gems just because I did not received a 5* hero after so many times. Except the HOTM which I already had, I draw no 5* in 170 pulls. I received Marjana on the last 10x try and I think that saved me from driving me crazy and it did stop me spending more moneys.

So 170 pulls with no 5* hero looks quite unusual.

All the best,
Silexu

PS If you want evidence I can provide the records of the draws.

1 Like

To give an alternate view from the “spend very little crowd”, I did 4x single draws this event and got the following:

Greymane, Boldtusk, Kashrek, Gretel (Squeeeeee!)

So, 1 Event 4*, 2 4*, 1 3*, or:

5* - 0 - 0%
4* - 3 - 75%
3* - 1 - 25%

Obviously these numbers are a very small sample. I quote the TC20 percentages because they came from a sample of 1000 pulls—the records of which were publicly posted; we can see who got what—and I think they are much more reliable than mine above.

1 Like

I have vip but otherwise spend nothing on the game and have been active daily for 5 months. (I consider my status as C2P, cheap to play). As such I have enough gems to make 3-4 summons a month along with small 1-2 gem chest accelerations to hit 2 monster chests a day or a jump start on a hero chest.

Last month I ended up with 2 HOTM in 2 days from one off elemental pulls. This shows nothing more or less than rng can cut both ways. The reality is I haven’t enough to ascend 1 aeron let alone 2 but as they were my first 2 5* I was happy nonetheless.

In the last 2 weeks I hit tc20 and have daily pulls. So far trending a 5* a week. So rng has been kind to me so far and fully expect to have weeks or even months of nothing and that’s fine. That’s statistics and thus hard to get emotional about.

By contrast I have had dozens of troop tokens and yet to have a single 4*. C’est la vie.

I feel sympathy for those spending $$$$$ but at the end of the day whilst it’s a lot to spend on gems even 100 pulls is a pretty thin sample size. To those people I mostly say thank you for funding the game for the rest of us. It’s appreciated and I hope your luck turns for the better. Don’t let confirmation bias ruin your enjoyment and don’t spend money if bad pulls are liable to make you angry.

6 Likes

C2P “cheap2play”. Love it! :grin:

9 Likes

I giggled reading this. I am a trained scientist and i can tell you that this NOT how science works. When we scientist have poor data or no data we say: we don’t know. We don’t derive conclusions from speculations and anecdotal evidence (which contrary to its name is not considered evidence at all). At the moment we got nothing official so we have got nothing. Temporal variable is completely unaccounted for and in online games when constant adjustments of balance are a norm its essential to know its influence on trends. All these assumptions about how much % TC gives us are completely useless because we don’t know if rates varied over the data collection period or even if they are constant in all circumstances. We just don’t know and the reason this information is not published is because it would probably (most likely) upset a lot of people here. Casinos don’t publish odds for a reason.

I think you put too much faith into OR principle, its a mistake. Have a read why:

“In the scientific method, Occam’s razor is not considered an irrefutable principle of logic or a scientific result; the preference for simplicity in the scientific method is based on the falsifiability criterion. For each accepted explanation of a phenomenon, there may be an extremely large, perhaps even incomprehensible, number of possible and more complex alternatives. Since one can always burden failing explanations with ad hoc hypotheses to prevent them from being falsified, simpler theories are preferable to more complex ones because they are more testable.”

In science we have this unwritten rule called SISO: The S is for pile of brown, smelly substance which i cannot name for obvious reasons and I is for IN, while O is for OUT. It means that if you put low quality input to your model your output will be of equally low quality aka useless. This is something we don’t do in science.

So in short Elpis is right and you are wrong.

10 Likes