New thread.

[btiw2]

Is there only a 5% edge using that 3-dice trick?

That was from memory, let’s work it out.
There are 216 possible results (6x6x6).
In 75 (5x5x3) you win a single
In 15 (5x3) you win double
In 1 you win treble
So in 125 (216-[75+15+1]) you lose.
You expect to gain (75x1)+(15x2)+(1x3)=108
Net loss =125-108=17
Percentage = 17/216 = 7.87%

My memory was wrong. I thought it was between 5-6%, it’s closer to 8%.

 

[btiw2]

No gold star for you matey!

This got competitive quickly. I think you deserve a house point DA.

 

[Jord86]

This got competitive quickly. I think you deserve a house point DA.

A teacher, mathematician and carpenter log on to the Screwfix Forum (just to check it's working) for an evening of Bayes Theories and....... Wait, let's not even go there, this could turn ugly faster than Dev can type the word 'rony!

 

[btiw2]

mathematician

I wish.

No, I’m just someone who was educated beyond my intellect.

 

[Jord86]

I wish.

No, I’m just someone who was educated beyond my intellect.

Well if you've honed your skills on the basics in childhood economics by plundering your classmates dinner money, I'lll assume you've done okay for yourself as an adult

 

[btiw2]

Well if you've honed your skills on the basics in childhood economics by plundering your classmates dinner money, I'lll assume you've done okay for yourself as an adult

It turned out that taking money from statistically impaired children wasn’t a terribly transferable skill.

I had to fall back on my charm, wit and general affability.

Spare any change?

 

[btiw2]

Book update for DA.

• Dogs of war. Started well, then took itself too seriously. Fun at the start. Genetically engineered humanoid dogs (with guns) who show complete loyalty to genocidal masters - it was always going to be fun. It should have been a short story. As a novel if had to be put out of its misery at the end.
• Letters to a young Muslim. Fed up with non-Muslims telling you what Muslims think? Read this. A unique insight on what a Muslim thinks about the Islamic world. Touching, honest, insightful and not a little naive. I’ll read this again. A page turner.
• Eleanor Oliphant is completely fine. TBH I picked this at random. It was everything I usually avoid. A best seller chick lit?! Perhaps I picked it to be contrary. I expected “Bridget Jones” (not that I’ve read it) - I got “wasp factory” (fantastic) mixed with “the curious case of the dog in the night-time”. The first half was engaging and funny, but for the last quarter the author was just phoning it in.

 

[Deleted member 33931]

That was from memory, let’s work it out. Let's not. 'Least not visibly.


My memory was wrong. I thought it was between 5-6%, it’s closer to 8%. Thank you

 

[dwlondon]

Okay. Here’s what is called a “moral proof”. It explains why it’s true but isn’t algebraically rigorous.

Imagine I bet on every number.

I roll the dice and get three different numbers. Say 4,5,6
I lose a quid on 1,2 and 3
But I win a quid on 4,5 and 6
No effect. I lose three quid and win three quid. Evens out.

But say I’d rolled 5,6,6
I’d lose on the 1,2,3 and 4
I’d win on the 5 and win double on the 6
I’d lose £4 and win £3. Overall loss of £1.

Say I rolled 6,6,6
I’d lose on the 1,2,3,4,5
Win treble on the 6
I’d lose £5 and win £3. Overall loss £2

If you go through the algebra, the expected payout is about 94-95% (from memory). Better than a fruit machine or a bookie, but not a good bet.

Okay. Here’s what is called a “moral proof”. It explains why it’s true but isn’t algebraically rigorous.

Imagine I bet on every number.

I roll the dice and get three different numbers. Say 4,5,6
I lose a quid on 1,2 and 3
But I win a quid on 4,5 and 6
No effect. I lose three quid and win three quid. Evens out.

But say I’d rolled 5,6,6
I’d lose on the 1,2,3 and 4
I’d win on the 5 and win double on the 6
I’d lose £4 and win £3. Overall loss of £1.

Say I rolled 6,6,6
I’d lose on the 1,2,3,4,5
Win treble on the 6
I’d lose £5 and win £3. Overall loss £2

If you go through the algebra, the expected payout is about 94-95% (from memory). Better than a fruit machine or a bookie, but not a good bet.

In this forecast the punter never wins anything.

 

[btiw2]

In this forecast the punter never wins anything.

You’re right. That’s why betting on every number would be bad.
It’s illustrates the problem by taking the chance out of it. It’s a useful shortcut for testing gambling games.

Just as you can win on roulette when you bet on a single number (although the odds are against you), but if you bet on every number you’d always lose.