## I want to play a game

For discussing anything related to physics, biology, chemistry, mathematics, and their practical applications.

Moderator: Flannel Jesus

### Re: I want to play a game

origami wrote:Either all the data is counted, or any data that isn't counted has to be factored and cancelled somehow.

Otherwise, you may be a very good software engineer, but nobody will ever hire you for scientific research.

Unless it's a global warming outfit. Then you are just the man they are looking for.
There's no one thing that's true. It's all true.
Ernest Hemingway

origami
Philosopher

Posts: 2313
Joined: Wed Jun 15, 2022 2:47 pm

### Re: I want to play a game

I do enjoy discussing these things, I do, don't get me wrong.

But being stuck discussing the dumb elementary things is a pain in the ass. Imagine the things we could be discussing now if instead of making some personal recrimination everytime you find yourself without an argument, you reconsider your position.

Don't be angry when you read this tomorrow. Taking the L is as important in math as in any pursuit of excellence.
There's no one thing that's true. It's all true.
Ernest Hemingway

origami
Philosopher

Posts: 2313
Joined: Wed Jun 15, 2022 2:47 pm

### Re: I want to play a game

It might be me bringing some of the conflicty vibes. I might take a break for a while.

Math is beautiful.
There's no one thing that's true. It's all true.
Ernest Hemingway

origami
Philosopher

Posts: 2313
Joined: Wed Jun 15, 2022 2:47 pm

### Re: I want to play a game

origami wrote:You can't just ignore data lol.

i can
pending

Mr Reasonable
resident contrarian

Posts: 32406
Joined: Sat Mar 17, 2007 8:54 am
Location: pimping a hole straight through the stratosphere itself

### Re: I want to play a game

Mr Reasonable wrote:
origami wrote:You can't just ignore data lol.

i can

Well proven -- one of the only - not The only -- true statements you ever made.
You have been observed.
Though often tempted to encourage a dog to distinguish color I refuse to argue with him about it
It's just the same Satanism as always -
• separate the bottom from the top,
• the left from the right,
• the light from the dark, and
• blame each for the sins of the other
• - until they beg you to take charge.
• -- but "you" have been observed --

The prospect of death weighs naught upon the purpose of life - James S Saint - 2009
obsrvr524
Philosopher

Posts: 4228
Joined: Thu Jul 11, 2019 9:03 am

### Re: I want to play a game

shut up idiot
pending

Mr Reasonable
resident contrarian

Posts: 32406
Joined: Sat Mar 17, 2007 8:54 am
Location: pimping a hole straight through the stratosphere itself

### Re: I want to play a game

origami wrote:
Flannel Jesus wrote:There are only 16 different ways to flip a coin 4 times. We count them all and ignore none of them.

Now I'm listening. What you said before was that we are only counting HHHH and HTHT and ignoring the rest.

So the deal is, we're going to count every single sequence of 4 flips that we see. There are 16 different ones.

TTTT
TTTH
TTHH
...
HTHT
THTH
...
HHHH

Now my original claim was that HTHT is as likely as HHHH,

but since we're looking at all the sequences now, and ignoring nothing (no junk data, every flip counts), we can modify my claim just a bit so that it's fundamentally the same but looking at everything.

Every sequence of 4 flips, every specific sequence, will occur approximately 6.25% of the time. If we do 64000 sets of 4, we should expect approximately 4000 of each set. We can look at all sets, but pay specific attention to sets HHHH and TTTT and HTHT.

Now, since I'm saying approximately, my position needs to have some wiggle room of course. This is random after all. So this next question is important for us agreeing on wiggle room:

If my position is that HHHH and HTHT should both appear approximately equally, I need to know what you, origami, think their relative frequencies should be. Do you think HHHH should appear half as much as HTHT? 1/4 as much? 3/4 as much? If you can be specific about your belief, we can agree on appropriate wiggle room.

Flannel Jesus

Posts: 6293
Joined: Thu Mar 31, 2011 11:32 pm

### Re: I want to play a game

origami wrote:It might be me bringing some of the conflicty vibes. I might take a break for a while.

Math is beautiful.

I do believe you've taken offence at things I'm surprised you took offence at, and introduced negativity when it wasn't there.

I did the same unnecessarily in the previous thread possibly.

You don't have to do that. I don't dislike you, I don't think you're stupid, I don't even think your intuition of the gamblers fallacy is stupid. I think you're wrong, but being wrong isn't the worst thing.

So if you could try to approach it with respectful disagreement and respectful arguments, I can too.

Flannel Jesus

Posts: 6293
Joined: Thu Mar 31, 2011 11:32 pm

### Re: I want to play a game

Please don't throw random insults at me about people hiring me for scientific research. You do not understand just how inadequate your participation in the experimental side of the conversation has been, so we could just insult each other all day about it, but I'd rather fucking not.

Please stop with finding little opportunities for jabs, I would like to stop too. Disagree with me on the points, and *that's it*. You can tell me why you think I'm wrong, but don't tell me some weird shit about who's going to hire me. It's completely unnecessary.

Flannel Jesus

Posts: 6293
Joined: Thu Mar 31, 2011 11:32 pm

### Re: I want to play a game

origami wrote:The game goes like this.

We will choose a trustworthy croupier to flip a coin 100 times.

For each flip, me and my opponent will have to bet $1. My opponent will give me 50 50 odds on every flip. I will give my opponent odds that shift depending on previous flips. They will shift in the following way: -------For the first flip, the odds are 50 50. -------Every time a flip gives a result, that result adds a 3% chance for the opposite result on the next flip. For example, if the first flip is heads, the second flip will pay 53 to 50 for heads and 47 to 50 for tails. If the second flip is also heads, then it's 56 44. If the third flip is tails, then it's 53 47 again. And so on. At the end of the 100 flips, the smaller pot will be subtracted from the larger pot and the loser will have to pay the difference to the winner. I didn’t read the *whole* thread. But. My intuition is that, if every play, each player bets a dollar (no more, no less), so the odds are based on payout, then the law of large numbers - since the *real* odds of a coin flip are always .5 every flip - says you’re both breaking even. Though, I’m not sure how many flips makes the bell curve… is 100 large enough? Just do it on independently designed software. But not with me. Fall semester ends 12/16/22. Apologies if I do not reply immediately. “In choosing myself, I choose the other.” - A marriage of Sartre & Levinas Ichthus77 ILP Legend Posts: 6012 Joined: Fri Sep 22, 2006 6:48 pm Location: pale blue clump of star particles ### Re: I want to play a game Ichthus77 wrote: origami wrote:The game goes like this. We will choose a trustworthy croupier to flip a coin 100 times. For each flip, me and my opponent will have to bet$1.

My opponent will give me 50 50 odds on every flip.

I will give my opponent odds that shift depending on previous flips. They will shift in the following way:
-------For the first flip, the odds are 50 50.
-------Every time a flip gives a result, that result adds a 3% chance for the opposite result on the next flip. For example, if the first flip is heads, the second flip will pay 53 to 50 for heads and 47 to 50 for tails. If the second flip is also heads, then it's 56 44. If the third flip is tails, then it's 53 47 again. And so on.

At the end of the 100 flips, the smaller pot will be subtracted from the larger pot and the loser will have to pay the difference to the winner.

I didn’t read the *whole* thread. But. My intuition is that, if every play, each player bets a dollar (no more, no less), so the odds are based on payout, then the law of large numbers - since the *real* odds of a coin flip are always .5 every flip - says you’re both breaking even. Though, I’m not sure how many flips makes the bell curve… is 100 large enough? Just do it on independently designed software. But not with me.

Then he later claims:

origami wrote:No no, lose 1.06 if you are wrong.

How can you bet $1 every flip and lose 1.06 if you are wrong? If the bet is always$1, then the most you can lose per flip is $1. Betting$1 means that is the most you lose, but greater than 1:1 odds mean you can gain more than you stand to lose.

2:1 odds means you win $2 for every dollar you bet. So you get back$3, the $1 that you bet, and the$2 that you won. If you lose you lose your $1. It's nonsense to claim you lose 1.06 if you are wrong when the bet is always$1.

Philosopher

Posts: 1692
Joined: Fri Aug 27, 2021 5:32 pm

### Re: I want to play a game

Yes, he acknowledged already that he misrepresented the odds in the op.

Flannel Jesus

Posts: 6293
Joined: Thu Mar 31, 2011 11:32 pm

Maybe subtracting losses from a pot seeded with the $1 per bet money, and seeing who empties it first? Fall semester ends 12/16/22. Apologies if I do not reply immediately. “In choosing myself, I choose the other.” - A marriage of Sartre & Levinas Ichthus77 ILP Legend Posts: 6012 Joined: Fri Sep 22, 2006 6:48 pm Location: pale blue clump of star particles ### Re: I want to play a game Flannel Jesus wrote:Yes, he acknowledged already that he misrepresented the odds in the op. So you must acknowledge that betting heads every flip would not lose more than$1 on the flips you are wrong, correct?

Philosopher

Posts: 1692
Joined: Fri Aug 27, 2021 5:32 pm

### Re: I want to play a game

I have no idea what that sentence means

Flannel Jesus

Posts: 6293
Joined: Thu Mar 31, 2011 11:32 pm

### Re: I want to play a game

Flannel Jesus wrote:I have no idea what that sentence means

Is it true that every bet will be $1? Forget the odds for a moment. Is it still true that every bet each player will bet$1?

Philosopher

Posts: 1692
Joined: Fri Aug 27, 2021 5:32 pm

Assuming you only put in a dollar, you can only lose a dollar ($1). You can’t lose more than you put in (1.06). Fall semester ends 12/16/22. Apologies if I do not reply immediately. “In choosing myself, I choose the other.” - A marriage of Sartre & Levinas Ichthus77 ILP Legend Posts: 6012 Joined: Fri Sep 22, 2006 6:48 pm Location: pale blue clump of star particles ### Re: I want to play a game Motor Daddy wrote: Flannel Jesus wrote:I have no idea what that sentence means Is it true that every bet will be$1? Forget the odds for a moment. Is it still true that every bet each player will bet $1? Not according to origami, no. Flannel Jesus For Your Health Posts: 6293 Joined: Thu Mar 31, 2011 11:32 pm ### Re: I want to play a game Ichthus77 wrote:Assuming you only put in a dollar, you can only lose a dollar ($1). You can’t lose more than you put in (1.06).

Right. The bet is how much you stand to lose. The odds are how much you win compared to how much you bet. 2:1 odds means you win $2 for every$1 you bet.

If I bet $25 with 4:1 odds, if I win I win$100 and if I lose I lose $25. Motor Daddy Philosopher Posts: 1692 Joined: Fri Aug 27, 2021 5:32 pm ### Re: I want to play a game origami is trying to finesse you, FJ Fall semester ends 12/16/22. Apologies if I do not reply immediately. “In choosing myself, I choose the other.” - A marriage of Sartre & Levinas Ichthus77 ILP Legend Posts: 6012 Joined: Fri Sep 22, 2006 6:48 pm Location: pale blue clump of star particles ### Re: I want to play a game Flannel Jesus wrote: Motor Daddy wrote: Flannel Jesus wrote:I have no idea what that sentence means Is it true that every bet will be$1? Forget the odds for a moment. Is it still true that every bet each player will bet $1? Not according to origami, no. The bet and the odds are separate issues. So he changed his original statement that every bet will be$1?

Philosopher

Posts: 1692
Joined: Fri Aug 27, 2021 5:32 pm

### Re: I want to play a game

Flannel Jesus wrote:
Is it true that every bet will be $1? Forget the odds for a moment. Is it still true that every bet each player will bet$1?

Not according to origami, no.

The bet and the odds are separate issues. So he changed his original statement that every bet will be $1? Yes, precisely. The entire betting scenario in the op is not what he meant. The bet he meant to offer is entirely different from what it says in the opening post. If it was exactly as he said in the opening post, I would have taken the bets in a heartbeat. Flannel Jesus For Your Health Posts: 6293 Joined: Thu Mar 31, 2011 11:32 pm ### Re: I want to play a game Flannel Jesus wrote:Yes, precisely. The entire betting scenario in the op is not what he meant. The bet he meant to offer is entirely different from what it says in the opening post. If it was exactly as he said in the opening post, I would have taken the bets in a heartbeat. So his change was that the bet will be according to the odds? So say the odds are 53:47. He is now saying that the bet has be .47, and the win will be .53? So with 53:47 odds, if you bet$47 you lose $47 if you lose, and win$53 if you win?

Philosopher

Posts: 1692
Joined: Fri Aug 27, 2021 5:32 pm

### Re: I want to play a game

No, he changed it so that all bets are 1:1 bets. He never offers any change of odds at all, he merely offers me different sizes of bets. So if he sees 1 heads, the next bet he offers me a 1.06 bet, where I stand to lose 1.06 or I stand to gain 1.06. The odds never change, just the size of the bet, according to what origami said he meant.

Flannel Jesus

Posts: 6293
Joined: Thu Mar 31, 2011 11:32 pm

### Re: I want to play a game

See page 3 of the conversation where he explains that, to him, 53 50 odds means "I stand to lose 1.06 and I stand to gain 1.06".

I believe he's realized since then that that's not how odds work.

Flannel Jesus