1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

Question on probabilities and conversion (CPA and getting laid)

Discussion in 'BlackHat Lounge' started by MyspaceisstillCool, Jun 22, 2015.

  1. MyspaceisstillCool

    MyspaceisstillCool BANNED BANNED

    Joined:
    Feb 16, 2015
    Messages:
    304
    Likes Received:
    36
    Does having a couple of succeses within a small sample size make that probability more likely or not?

    For example does two conversions within a small sample mean that the conversion rate is any higher?

    Although I thought so at first I realised actually that its like if you happened to get two heads in a row isnt it? Ie getting two hits is just as likely as getting all misses its just your mind rememebrs the hits?

    Once could easily have been a fluke but twice in a short space makes it alot less likely to be fluke doesnt it or no?

    This actually relates to me getting laid. I happened to get laid twice in the space of a few weeks in this new club I started going to but Ive been going again and not getting any results so I dont like it now. Im thinking they were just wild flukes I got two in a row.

    Is the only way to know by going over a long period of time?

    But it also relates to cpa where with one network I got no hits within a small sample size then hated the network and got a fuke in another such that I thoguht it was excellent and stuck with it. But then when I was getting more traffic I tried the first one again and ended up getting more money :D

    So Im wondering what is the most effective way to find out the value of your sample? I know the larger the sample size the better but are there any short cuts to save time so I dont have to waste nights out on shitty nights where I dont get laid? Maybe I should approach more in the clubs Im at so that effectively increases sample size in the environment on the given evening. That isnt so easy though given my hatred of rejection and pathetically low self esteem.
     
    Last edited: Jun 22, 2015
  2. PinguSpy

    PinguSpy Jr. VIP Jr. VIP

    Joined:
    Dec 7, 2007
    Messages:
    1,116
    Likes Received:
    968
    Occupation:
    Internet Farmer
    Location:
    Pineal Gland
    Home Page:
    [​IMG]

    I dont get it bro.
     
  3. THUNDERELVI

    THUNDERELVI Elite Member

    Joined:
    Sep 12, 2009
    Messages:
    2,383
    Likes Received:
    2,060
    Gender:
    Male
    Location:
    W3
    What shit have you smoked ?
     
  4. MyspaceisstillCool

    MyspaceisstillCool BANNED BANNED

    Joined:
    Feb 16, 2015
    Messages:
    304
    Likes Received:
    36
    What is your problem its not hard to understand.
     
  5. MyspaceisstillCool

    MyspaceisstillCool BANNED BANNED

    Joined:
    Feb 16, 2015
    Messages:
    304
    Likes Received:
    36
    Here it is explained more plainly

    If there is only a small sample size but you get a certain eventuality twice within that sample does that make it any more likely that that eventuality will happen in future? I dont mean in terms of shaping the future [​IMG] but rather than its a good indicator that it is probable in future.

    Ie if you are trying to figure out how often a favourable result will happen. If you did something ten times and the result you wanted happened twice does that make it any more likely that the favourable result would happen in future than if it happened 0 times? Or could both those times have been flukes? Well I see that its more likely than 0 [​IMG] but Im not sure how much?

    Is it just a case that the only way to tell a good average is having a bigger sample? But do the 2 hits indicate a good chance it would happen in future or it doesnt tell you much about future eventualities given the small sample?

    If the given result is a real scenario can I test that probability by saying that so far there is a 2/10 favourable result? So I can do the same thing again 10 times and expect or hope for at least 2/10 result? Not that I definintely will but the previous succeses indicate it might make it more epxected to happen?
     
  6. HoNeYBiRD

    HoNeYBiRD Jr. VIP Jr. VIP

    Joined:
    May 1, 2009
    Messages:
    6,952
    Likes Received:
    7,982
    Gender:
    Male
    Occupation:
    Geographer, Tourism Manager
    Location:
    Ghosted
    if you're talking about email/zip submits, in that case if an offer is any good, the first 1-2 conversions usually come very fast (within the first 10 uniques), then it stabilizes on a level (1 conversion per 20-30 uniques), then it can get adjusted again, if your traffic isn't converting on the advertiser's end

    if an offer stops converting, just switch to another from a different advertiser
    you should do the same with that club, if it doesn't convert anymore :p