How profiles thing work?

degoodegoo Posts: 49 ✭✭
edited February 14 in General Discussion
I'd like some from Hutch explain how works the assigment of cars for each player. It is clear that when creating an account, it is assigned kind of "profile", so you have more chances to get some cars rather than others, based of this crappy thing.

I compared with friends and always follow the same pattern... ie, for me, now in the last update, exhibition thing, it is almost impossible to get a Lotus.. but have lot of bentleys and astons... others players tons of lotus but none astons.. it is frustating....

Comments

  • degoodegoo Posts: 49 ✭✭
    I'm not talking about better or worst, just different cars. Profile is a sure thing, just compare with other people, it is a way to avoid people with very same collection of cars... the events could be weirds.. but it is frustating when you can not get never some cars and wow, the same car multiple times. It is the same probability for all cars??? impossible
  • kaje73kaje73 Posts: 320 ✭✭✭
    Since GE I've pulled a good number of every GE manufacturer (except Ariel which isn't surprising as there is only 1 non prize / non legendary), as I'm sure most here have. Profiles limiting which manufacturer you can pull isn't a thing.  
  • willcf15willcf15 Posts: 382 ✭✭✭✭
    edited February 27
    degoo said:
    Profile is a sure thing, just compare with other people, it is a way to avoid people with very same collection of cars...
    The beauty of long-shot gambling (read "Top Drives") is that it very rarely produces the same results time after time, if the payouts are truly random.

    I know that, when you pull your fourth Skyline R34, or 9th Renault R20 without a single Dezir when they were both C class (I'M NOT SALTY, YOU'RE SALTY), it's tempting to believe that there's a reason for this. But there isn't. It's just RNGsus. Succumb to our lord and torturer.


    Post edited by Hutch_Gsearch on
  • KarhgathKarhgath Posts: 180 ✭✭✭
    What you're describing is a common misconception about randomness, that it's evenly distributed. 

    That's not the case. Randomness produces clumping, for example lots of drops from one manufacturer and few or none from another. 
    Yes. It is simply that, as you draw more cards, the chance you pull any duplicate (same car, same manufacturer, etc.) of a card you already own becomes much higher than pulling a unique card you do not have. Often the random distribution will seem like it favors a manufacturer or something ("clumping"), but that had nothing to do with the underlying randomnes. As humans, we often try to find or assume patterns. It's called apophenia.
  • degoodegoo Posts: 49 ✭✭
    Oh sure sure, random is a thing, but, after almost three years of playing, I'm so sure each player has more chances to get a set a cars than others (not impossible, just more chances). In my case: Lotus, Caterham... just in dreams, among others. I can collect tons of some cars, while others never seen. Not one, two neither three cases that should be supposed to happen in case of true plain random.
    just some calculation, supposing we have a total of 1000 cars in Top Drives:
    the chance to get a car: 5/1000 (5 cars, I know, not the same % depending on car class)
    Chance to gain a car five times (lot of cases to me, even much more): 3125/1000000000000000. Wow.. if the lottery have the same "random" lot of people can be rich just following some numbers...

  • REALAISREALAIS Posts: 906 ✭✭✭✭
    That example is very bad. If you do examples then do them properly. 

  • degoodegoo Posts: 49 ✭✭
    edited February 27
    yeah, mixed concepts, not the same getting the same car multiple times at once, but anyway, it's weird repeating more than 5 times (*lot* of cars) and never getting others, but the same supposed chance... again, not just one case or two, that should be usual in pure random environment...
  • MoogMoog Posts: 469 ✭✭✭✭
    No disrespect to Hutch, but do you think they've coded a complicated system of profiles and car weighting system or a system that checks your hand for cars that you need. Or did they just write a one line query that picks a class based on the drop rates and then picks a random car of that class.
    Remember, this is Hutch we're talking about.
  • degoodegoo Posts: 49 ✭✭
    Moog said:
    No disrespect to Hutch, but do you think they've coded a complicated system of profiles and car weighting system or a system that checks your hand for cars that you need. Or did they just write a one line query that picks a class based on the drop rates and then picks a random car of that class.
    Remember, this is Hutch we're talking about.

    he he! nah, no complicated system, just my theory is there are kind of 3 or 4 profiles where some set of cars have more or less % to get, when creating a new user, they assign one of this (don't know if based on location, platform, a mix, ..) and that's, all your drops depends on this.

    Just a theory... but.. in great exhibition have lot of Aston Martin, Bentley's.... not a single Lotus and got the same aston martin and bentley (C class) three times each one! pfew...weird random algorithm
  • evilprofesseurevilprofesseur Posts: 1,843 ✭✭✭✭✭
    degoo said:
    Moog said:
    No disrespect to Hutch, but do you think they've coded a complicated system of profiles and car weighting system or a system that checks your hand for cars that you need. Or did they just write a one line query that picks a class based on the drop rates and then picks a random car of that class.
    Remember, this is Hutch we're talking about.

    he he! nah, no complicated system, just my theory is there are kind of 3 or 4 profiles where some set of cars have more or less % to get, when creating a new user, they assign one of this (don't know if based on location, platform, a mix, ..) and that's, all your drops depends on this.

    Just a theory... but.. in great exhibition have lot of Aston Martin, Bentley's.... not a single Lotus and got the same aston martin and bentley (C class) three times each one! pfew...weird random algorithm
    Got any worthwhile data to support that theory?
  • mikesmikes Posts: 4,372 ✭✭✭✭✭
    I concur with everyone here, this is not a thing. It's frustrating when a unicorn alludes you, but there is no profile thing.
  • Hutch_GsearchHutch_Gsearch Posts: 2,581 admin
    Not a thing.  Just saying. 
  • NinjaDesignzNinjaDesignz Posts: 2,782 ✭✭✭✭✭
    degoo said:
    I'm not talking about better or worst, just different cars. Profile is a sure thing, just compare with other people, it is a way to avoid people with very same collection of cars... the events could be weirds.. but it is frustating when you can not get never some cars and wow, the same car multiple times. It is the same probability for all cars??? impossible

  • KarhgathKarhgath Posts: 180 ✭✭✭
    degoo said:
    Moog said:
    No disrespect to Hutch, but do you think they've coded a complicated system of profiles and car weighting system or a system that checks your hand for cars that you need. Or did they just write a one line query that picks a class based on the drop rates and then picks a random car of that class.
    Remember, this is Hutch we're talking about.

    he he! nah, no complicated system, just my theory is there are kind of 3 or 4 profiles where some set of cars have more or less % to get, when creating a new user, they assign one of this (don't know if based on location, platform, a mix, ..) and that's, all your drops depends on this.

    Just a theory... but.. in great exhibition have lot of Aston Martin, Bentley's.... not a single Lotus and got the same aston martin and bentley (C class) three times each one! pfew...weird random algorithm
    Randomness is never distributed evenly. This is why it's unintuitive and statistics are hard to get properly.

    Take 23 random people. What is the chance than any 2 of them share a birthday? Well, it's 50%. Yes, even  if there are 365 days in a year. If it seems wrong to you, search for the birthday paradox.

    This is why you seem to get a lot of duplicate vs cars you want or identify as missing. The more card you draw, the more chance you have to pull a duplicate.

    Let's say they are 100 rare cards for the sake of the argument. If you have 80 of them, you have 80% chance of pulling a duplicate and 20% chance to draw one you do not have. That's why getting that damn unicorn is hard and you end up getting duplicates from packs. There is no memory in randomness.
Sign In or Register to comment.