Premium Bond Winner ?

kuratowski

If, as stated, they are sampling from a physical process, it will be random.

thegentleway

kuratowski said:

If, as stated, they are sampling from a physical process, it will be random.

You’re right. Just looked it and it is physical: neon tubes. Random number generators are usually programs and therefore not completely random. Pretty cool. 

RG2015

thegentleway said:

kuratowski said:

If, as stated, they are sampling from a physical process, it will be random.

You’re right. Just looked it and it is physical: neon tubes. Random number generators are usually programs and therefore not completely random. Pretty cool. 

Light bulb moment. (Possibly filled with an inert gas, such as neon?)

Thank you both and Eco-miser. I understand that whilst it does involve a computer chip, it is not a computer or supercomputer process which relies on software programming.

eskbanker

The issue of validating the randomness is addressed at https://nsandi-corporate.com/media-resources/ernie

But how can you prove it’s all random?

The test of proving that ERNIE’s outputs are robust has long been discussed by experts in the field of randomness.

However, from the outset, engineers working in conjunction with – but exclusive of – the ERNIE programme have proven that the numbers created by the machine are unpredictable and follow no set patterns.

The original system to test ERNIE’s randomness was called Pegasus, designed by Dame Stephanie Shirley, who went on to be one of Britain’s greatest computer engineers.

Since then, as with ERNIE’s evolution, testing randomness has developed and is now managed by the Government Actuary’s Department (GAD).

Each month after the Premium Bonds numbers have been generated, they are sent securely to GAD who run a number of tests to identify whether the outputs are truly random:

• The frequency test – whether every possible character in each position of the Bond number appears as often as it should. 
• The serial test – looking at the number of times one digit follows another (for example the number of 3s coming directly after 7s). 
• The poker test – looking at the number of times that a group of characters generated consecutively contain four identical characters, three of a kind, two pairs, one pair and all different.
• The correlation test – looking for correlation between characters in two different Bond positions over a series of Bond numbers. 

ERNIE has never failed to be anything but random in every test carried out by GAD.

No doubt this will be dismissed as propaganda by those who choose to cling to the belief that newer bonds are luckier or that it's better to have large blocks, or whatever other old wives' tale is flavour of the week among conspiracy theorists, but even if it could be demonstrated that outcomes weren't random from a purely mathematical perspective, the notion that such lack of randomness would favour specific styles of purchasing is obviously nonsense....

thegentleway

eskbanker said:

The issue of validating the randomness is addressed at https://nsandi-corporate.com/media-resources/ernie

But how can you prove it’s all random?

The test of proving that ERNIE’s outputs are robust has long been discussed by experts in the field of randomness.

However, from the outset, engineers working in conjunction with – but exclusive of – the ERNIE programme have proven that the numbers created by the machine are unpredictable and follow no set patterns.

The original system to test ERNIE’s randomness was called Pegasus, designed by Dame Stephanie Shirley, who went on to be one of Britain’s greatest computer engineers.

Since then, as with ERNIE’s evolution, testing randomness has developed and is now managed by the Government Actuary’s Department (GAD).

Each month after the Premium Bonds numbers have been generated, they are sent securely to GAD who run a number of tests to identify whether the outputs are truly random:

• The frequency test – whether every possible character in each position of the Bond number appears as often as it should. 
• The serial test – looking at the number of times one digit follows another (for example the number of 3s coming directly after 7s). 
• The poker test – looking at the number of times that a group of characters generated consecutively contain four identical characters, three of a kind, two pairs, one pair and all different.
• The correlation test – looking for correlation between characters in two different Bond positions over a series of Bond numbers. 

ERNIE has never failed to be anything but random in every test carried out by GAD.

No doubt this will be dismissed as propaganda by those who choose to cling to the belief that newer bonds are luckier or that it's better to have large blocks, or whatever other old wives' tale is flavour of the week among conspiracy theorists, but even if it could be demonstrated that outcomes weren't random from a purely mathematical perspective, the notion that such lack of randomness would favour specific styles of purchasing is obviously nonsense....

Good point about if it wasn't completely random and a pattern could eventually be observed, it would be rather difficult to exploit as you can only buy new bonds.

eskbanker

thegentleway said:

Good point about if it wasn't completely random and a pattern could eventually be observed, it would be rather difficult to exploit as you can only buy new bonds.

Indeed, but further back on this thread (I think), there have been multiple deluded people deciding that their bonds were unlucky by virtue of their age and that selling and repurchasing new ones would magically change that!

Eco_Miser

thegentleway said:

kuratowski said:

If, as stated, they are sampling from a physical process, it will be random.

You’re right. Just looked it and it is physical: neon tubes. Random number generators are usually programs and therefore not completely random. Pretty cool. 

This is why I pointed out that ERNIE is not a computer.

I suspect that while a lot of so called random number generators are actually pseudorandom number generating programs, those where the randomness is really important, such as Premium Bonds, or the National Lottery use physical events of some kind to introduce true randomness.

thegentleway

eskbanker said:

thegentleway said:

Good point about if it wasn't completely random and a pattern could eventually be observed, it would be rather difficult to exploit as you can only buy new bonds.

Indeed, but further back on this thread (I think), there have been multiple deluded people deciding that their bonds were unlucky by virtue of their age and that selling and repurchasing new ones would magically change that!

As effective as kissing a rabbit's foot...

Eco_Miser said:

thegentleway said:

kuratowski said:

If, as stated, they are sampling from a physical process, it will be random.

You’re right. Just looked it and it is physical: neon tubes. Random number generators are usually programs and therefore not completely random. Pretty cool. 

This is why I pointed out that ERNIE is not a computer.

I suspect that while a lot of so called random number generators are actually pseudorandom number generating programs, those where the randomness is really important, such as Premium Bonds, or the National Lottery use physical events of some kind to introduce true randomness.

I didn't catch you'd specified they used hardware - apologies!

pbcpdeveloper

I think people 'know', that all numbers have an equal chance but their perception can be easily influenced by their familiarity of the linear nature of the numbering system. With Premium Bonds that perception gets futher influenced as you can introduce the 'age' of numbers and how often they win, which has already been discussed above.  

You can see this with the National Lottery where people preceive that the numbers 1-6 have less chance of appearing than a scattering of 6 numbers between 1- 59, as you can apply criteria that make the chances of 1-6 look astronomical.  I once saw a program where someone in L.A. ran courses on picking lottery numbers, as you get certain types of special numbers that have influence in nature and science, etc etc.

The reality is that the for the purposes of the Lottery, the numbers 1-59 have lost their value as numbers, they are actually just easily identifiable symbols that we are used to working with.  If you replaced the numbers with 59 different squiggles and shapes scattered aross a surface they loose the criteria which makes them linear so there are no symbols 1-6 and to say these 6 symbols have more or less chance of being drawn than another set of 6 symbols becomes nonsensical.

Nebulous2

pbcpdeveloper said:

I think people 'know', that all numbers have an equal chance but their perception can be easily influenced by their familiarity of the linear nature of the numbering system. With Premium Bonds that perception gets futher influenced as you can introduce the 'age' of numbers and how often they win, which has already been discussed above.  

You can see this with the National Lottery where people preceive that the numbers 1-6 have less chance of appearing than a scattering of 6 numbers between 1- 59, as you can apply criteria that make the chances of 1-6 look astronomical.  I once saw a program where someone in L.A. ran courses on picking lottery numbers, as you get certain types of special numbers that have influence in nature and science, etc etc.

The reality is that the for the purposes of the Lottery, the numbers 1-59 have lost their value as numbers, they are actually just easily identifiable symbols that we are used to working with.  If you replaced the numbers with 59 different squiggles and shapes scattered aross a surface they loose the criteria which makes them linear so there are no symbols 1-6 and to say these 6 symbols have more or less chance of being drawn than another set of 6 symbols becomes nonsensical.

I remember reading that around 10,000 people per draw just pick numbers 1 to 6. If they ever do come up the payout will be disappointing.