Unfair Coin Flips

[burritosarepurple]

After seeing a community post about someone suspecting that the coin flipping mechanic is biased, I've started taking notes on the results of my coin flips. So far I've only monitored on my last 42 competitive games. I've been using a deck that relies on going first, so I chose that any time I got heads. Within these 42 games, I got 17 heads and 25 tails. That's a 40% rate of heads. While not rigged without a doubt, still a decent margin below the expected 50%. Anyone else feel the same way?

I will be updating this with more data as I play more.

[Daguza]

Are you P2P or F2P?

[burritosarepurple]

Originally posted by Daguza:

Are you P2P or F2P?

Lmao I haven't spent any money.

[HeraldOfOpera]

Somebody run the numbers on how unlikely this is. I'm going to guess that it comes up as "more likely than any given starting hand, and starting hands happen all the time".

[Papa Shekels]

Originally posted by HeraldOfOpera:

Somebody run the numbers on how unlikely this is. I'm going to guess that it comes up as "more likely than any given starting hand, and starting hands happen all the time".

I could dig up my old college notebooks and remind myself how to properly do this, but if the past threads are anything to go by, it's not worth the effort as it won't change anything. People who want to believe the world is out to get them will continue to do so not matter how many numbers are thrown at them because the numbers are all made up by the evil in the shadows to discredit their secret knowledge

[Daguza]

Originally posted by burritosarepurple:

Originally posted by Daguza:

Are you P2P or F2P?

Lmao I haven't spent any money.

Everytime I run my Archfiend's, I always draw my lvl 8s 7/10 of time in opening hand. But when I try putting in trade ins, I all the sudden don't draw lvl 8s no more. Just trade in with no lvl 8. F2p too so just feels sus...

[Filipino Enjoyer]

42 games is not a big enough sample size kek

[Vento]

Originally posted by Barbatos:

42 games is not a big enough sample size kek

42 is a big enough sample, central limit theorem, lmao

[Filipino Enjoyer]

Originally posted by Vento:

Originally posted by Barbatos:

42 games is not a big enough sample size kek

42 is a big enough sample, central limit theorem, lmao

it's not, you need hundreds if not thousands to actually show any statistical bias

[Papa Smurf]

Confirmation Bias. Think - if you're getting lots of tails, then that means everyone else is getting lots of heads, which throws your theory out of the window.

Either Konami has it out for you or you;re wrong. Which is more likely?

[JKArtorias]

Let me put on my tinfoil hat to stop the Konami spies from using electro magnetic radio waves to interfere with my message. I agree it is rigged, only give me heads Mr. Konami.

[HeraldOfOpera]

Originally posted by Papa Smurf:

Either Konami has it out for you or you;re wrong. Which is more likely?

Worth repeating that Konami cannot actually perceive humans, only wallets.

[burritosarepurple]

Originally posted by Barbatos:

Originally posted by Vento:

42 is a big enough sample, central limit theorem, lmao

it's not, you need hundreds if not thousands to actually show any statistical bias

You're completely right. This was only a theory I had based off a paranoid community post, and I wanted to see if other people were experiencing the same trend. I won't claim I have any evidence until a much more solid sample size.

[burritosarepurple]

Originally posted by Papa Smurf:

Confirmation Bias. Think - if you're getting lots of tails, then that means everyone else is getting lots of heads, which throws your theory out of the window.

Either Konami has it out for you or you;re wrong. Which is more likely?

That's a good point. Right now I'm only testing on whether or not I get 50/50 odds. My theory isn't that it's always rigged to give you tails, just that the odds aren't exactly 50/50.

[Papa Shekels]

Originally posted by burritosarepurple:

That's a good point. Right now I'm only testing on whether or not I get 50/50 odds. My theory isn't that it's always rigged to give you tails, just that the odds aren't exactly 50/50.

With the way statistics work, you won't ever get a real 50/50. There will probably be a specific point where you hit 50/50 for a moment, and then it will tip in one direction again and go there for a bit before swinging back over. It will not *stay* at 50/50. And if you do this for the hundreds/thousands of trials needed to get any meaningful data, it will be close to 50/50 but not exact. And even when you're at 2500 coin flips, you will get a lucky or unlucky streak that pushes it oddly far in one direction and it might take hundreds more for it to stabilize back to the expected ratios. That's the thing people don't seem to intuitively grasp about numbers

[burritosarepurple]

Originally posted by Papa Shekels:

Originally posted by burritosarepurple:

That's a good point. Right now I'm only testing on whether or not I get 50/50 odds. My theory isn't that it's always rigged to give you tails, just that the odds aren't exactly 50/50.

With the way statistics work, you won't ever get a real 50/50. There will probably be a specific point where you hit 50/50 for a moment, and then it will tip in one direction again and go there for a bit before swinging back over. It will not *stay* at 50/50. And if you do this for the hundreds/thousands of trials needed to get any meaningful data, it will be close to 50/50 but not exact. And even when you're at 2500 coin flips, you will get a lucky or unlucky streak that pushes it oddly far in one direction and it might take hundreds more for it to stabilize back to the expected ratios. That's the thing people don't seem to intuitively grasp about numbers

The more you flip a coin, the closer you'll get to 50/50. While you may have some outlying streaks, if you flip it enough, it'll balance out. However, if the odds are skewed towards one, the difference will gradually increase until it's noticeable. I'm just trying to see if it'll increase.