Simulation: 10 vs 15 phone calls

Teh_DaminTeh_Damin Unconfirmed, Member Posts: 16
Hey guys,

Some time ago we got introduced to 10 phone calls which at first looks like a nice and cheap alternative to 15 calls, and also gives guaranteed 32 tokens if you happen to find the highlighted hero. But is it?

I've decided to write a short app which simulates a 10000 scenarios for both 10 and 15 phones calls.

Logic:
  • Number of tries: 10000 (both)
  • Tokens of highlighted hero to reach: 500 (both)
  • Chance to get highlighted hero: 25 (10) and 15 (15)
Flow:
Rolling a number => if meets odds for highest amount of tokens - take, if not, check for the next threshhold etc. If we roll a hero, and none of the higher numbers match, we get the minimum
Example (for 15 call):
Number = 15
Check if number is less or equal to 1 - NO
Check if number is less or equal to 16 - YES (take 256 tokens)
NOTE: I've made it 16, not 15 to keep the 15% chance.

I've also included a mechanism, which checks for the highest amount of tokens in 2 (for 10) or 3 (for 15) cards and takes the highest (so if we roll, 64, 64 and 128 of sasha, the system will take 128).

Results
For 15 calls the data is as follows:
  • On average, it takes between 225 - 230 phones to get 500 tokens of highlighted hero
  • On average, it takes 15 calls to get 500 tokens of highlighted hero
  • Chance of 256 per call is 11(which is 4% lower than advertised)
  • Chance of 512 is 0,7 (0,3 lower than advertised)
For 10 calls the data is as follows:
  • On average, it takes between 282-284 phones to get 500 tokens of highlighted hero
  • On average, it takes 28 calls to get 500 tokens of highlighted hero
  • Chance of 64 per call is 21% (which is 5% higher than advertised)
  • Chance of 128 per call is 7,5 (which is 2,5% higher than advertised)
  • Chance of 256 is 0,4 (0,1 higher than advertised)
EDIT:
I do realise that it might look strange, that the chance for higher amount of tokens looks different than advertised, and it might be the fault of my logic or just the fact of limiting the tries to 10k and pure luck, I'm still thinking about it.

In my experience this kind of data is more or less mirroring the real world situation. For me, the 10 calls have been very average, whereas, 15 double calls are more potent in general.

I hope you find this helpful or anyhow interesting.

If you have any questions, or think I've done some error, please feel free to point it out. It's the first time I've done something like that, so it's probably not flawless :)
jimmorrison369WellyLugapositivemsapaevSCBMAWahooDawgsuzequeueSg_Death

Comments

  • jimmorrison369jimmorrison369 Member Posts: 1,765
    I have a excel sheet on my desktop that will be able to calculate the exact theoretical chance. I'm not at home and don't have much time so...

    A quick calculation shows me, for the 15 call, there is a .45% chance to get the 512 tokens from the double 15 call

    For each of the 3 cards there is a 15% chance that it will be the selected hero.

    So this means there is
    1-(1-.15)^3=38.6% chance to have at least the chosen hero.

    For the 512
    You need to get the 15% chance for the hero + the 1% it is a 512.
    .15*0.01=.15% for 1 card.
    1-(1-0.0015)^3=0.45% chance.

    Hope this helps for your logic.
    Namely it seems you consider the 1% to be for all 3 cards.
    + You consider the only possible hero being the one advertised.

    When I get home I might have time to find my excel sheet.

    Great job regardless.
    SCBMA
  • Teh_DaminTeh_Damin Unconfirmed, Member Posts: 16
    @jimmorrison369
    Thanks for your answer!
    I kind of could not be bothered bo write down it all, but basically what I do is I do each card separate. So, for each card there is 1% chance for it to be 512.

    And yes, I did not consider other heroes, basically if it rolls another hero, it doesn't count in my calculations. Tbh NG keeps soooooo much details away from us it's kind of funny... Even with the newest feature - GW, there is so much me and my guild had to test out and theoritize to learn, rather than NG giving us all that in info notes. So I tried my best with what I've got.

    Anyhow, this simulation it to show the disadvantage of 10 phone calls more than be super accurate for scientific data. If my assumptions are wrong, then they are most likely wrong for both cases, therefore, both cases will be lowered/highered equally, thus, making the comparison more or less the same.
    jimmorrison369
  • jimmorrison369jimmorrison369 Member Posts: 1,765
    In general, the 15 call will be much better than other calls.
    The 40 call can be better depending on what you are after.

    Take a look here if you're interested.
    There is someone in there who simulated against our theory.

    I'll come back to this topic. You've sparked my eagerness to define this stuff.
  • Teh_DaminTeh_Damin Unconfirmed, Member Posts: 16
    In my experience, you don't need Maths brain to say 40 call is not worth it at all, unless you get really lucky with the drop and keep all 3. Most of the time for me it's 32s for 40 phones -.-
  • jimmorrison369jimmorrison369 Member Posts: 1,765
    I beg to differ, as can be seen in the post. When you get your priorities straight, the token per phone ratio is really good.

    Of course the double calls are much better, but that is an unfair comparison.

    And you always get to keep 3 in the 40 call. In normal you get to keep only 1.
    If you keep all above 32, and reroll the 32's than your token output will be significantly higher than the regular 15 call.
  • Teh_DaminTeh_Damin Unconfirmed, Member Posts: 16
    @jimmorrison369
    You're right. That depends on your priorities. Taking into consideration that there is only handful of characters REALLY useful in the game and most survivors have much better hero counterpart, you end up with a situation that you do not want to spend 120 phones for 3 calls, where chances for your fav hero are 1/5 out of 20% (or whatever it is for 40), but you'd rather spend 8 specific calls for that hero for 15% chance of double tokens.

    That's at least my perspective and I know a lot of people in my guild share it. However, it does not mean they neglect 40 calls. Generally, if survivor is not ideal, it's probably not worth taking anyways (aside from shooter and hunter), but again, that's my opinion and it's probably debatable :)
  • jimmorrison369jimmorrison369 Member Posts: 1,765
    Agreed mostly, especially because regular survivors aren't as valuable as they where back then.
  • jimmorrison369jimmorrison369 Member Posts: 1,765
    edited March 24
    Remember, This is about the double hero calls. It is about the odds at the selected double hero.
    Inevitably you'll get other tokens too. Those are not counted in this post,
    Also realize, the odds to get low tokens are higher than in this post. But they are lowered because sometimes you get a higher toke choice. Then, even though you did get 64 tokens for instance, it is not taken because there is 128 tokens.
    Teh_Damin said:


    On average, it takes 15 calls to get 500 tokens of highlighted hero
    Chance of 256 per call is 11(which is 4% lower than advertised)
    Chance of 512 is 0,7 (0,3 lower than advertised)

    For the 15 call
    Theoretically, it takes 12.7 calls to get to 500 tokens. 190.6 radios
    The chance of 64 is 31.6%
    The chance of 256 is 6.58%
    The chance of 512 is 0.449%
    Teh_Damin said:


    On average, it takes between 282-284 phones to get 500 tokens of highlighted hero
    On average, it takes 28 calls to get 500 tokens of highlighted hero
    Chance of 64 per call is 21% (which is 5% higher than advertised)
    Chance of 128 per call is 7,5 (which is 2,5% higher than advertised)
    Chance of 256 is 0,4 (0,1 higher than advertised)

    For the 10 call

    There isn't currently one on. From the numbers in your post.
    I'm going to assume the given percentages per card are:
    (These are the odds PER CARD given by the game. Not your odds per call!)
    25% selected hero
    0.3% for 256
    5.0% for 128
    15.0% for 64
    79.7% for 32

    If the above is true, and I'm really not so sure. But maybe someone can correct.

    Theoretically, it takes 26.3 calls to get to 500 tokens. 262.6 radios
    The chance of 32 is 33.9%
    The chance of 64 is 7.26%
    The chance of 128 is 2.48%
    The chance of 256 is 0.150%




    To finish up.
    If you are looking for your specific hero.
    The 10 call will cost you 37.7% more radios when opposed to the 15 call.
    Or you can say. The 15 call saves you 27.4% radios when opposed to the 10 call.
  • jimmorrison369jimmorrison369 Member Posts: 1,765
    @Teh_Damin

    As can be seen from the post above.
    You figured 10 calls where 24% more expensive.
    Theoretically it seems to be 38% more expensive.

    We agree on the direction, but disagree about the size of the direction.
  • UrgoldthUrgoldth Member Posts: 77
    Are 40 calls more rewarding if you only care about hero tokens regardless of hero, which is my case?
  • jimmorrison369jimmorrison369 Member Posts: 1,765
    Yes, at least when 40 and 15 are equal. Double tokens 15 call will win.

    But if you only care about hero's, and are keeping any hero tokens.
    Then the 40 call is a lot better.
  • UrbanjesusUrbanjesus Member Posts: 68
    Anyone done any analysis of the (albeit rare) 5 radio calls like the ones done when Riot Glenn was introduced?
Sign In or Register to comment.