Very silly National Lottery related question.

Very silly National Lottery related question.

Author
Discussion

ferrariF50lover

Original Poster:

1,834 posts

233 months

Saturday 18th June 2016
quotequote all
Bear with me...

Waiting for the News tonight, I caught the Lottery draw. As most of you will know, the Lottery is drawn by dropping little numbered rubber balls into a drum with a rotating arm inside. The idea is that the arm muddles the order of the balls, a job it does very effectively. My question is: why?

As I say, bear with me...

The balls always drop into the drum in the same order. The rotating arm always starts in the same position, accelerates at the same rate to the same speed. It always impacts the balls with the same force and the balls always have the same nature (weight, material etc). Why then, given that everything is the same every time, do the balls react differently?

If I take a tennis ball and drop it 1000 times from the same height, onto the same surface in air that's the same density etc, it'll bounce to the same height. That's how physics works. If nothing changes, nothing changes.

What's the craic?

Stupeo

1,343 posts

200 months

Saturday 18th June 2016
quotequote all
Don't they use different machines each week, in some sort of rota? I bet they aren't identical?

MitchT

16,237 posts

216 months

Saturday 18th June 2016
quotequote all
Here's another silly question.

If there are six balls plus a 'bonus ball', then there are seven balls in total.

If you match six balls, then you've matched six balls.
If you match five balls plus the bonus ball, then you've matched six balls.

So why is matching six seen as so much bigger a deal than matching five plus the bonus? Surely the odds are the same.

Simpo Two

87,097 posts

272 months

Saturday 18th June 2016
quotequote all
Technically you're right but I think chaos theory applies, or some other law of randomness. 0.00001% alteration at one stage gives 50% alteration further down the line. Compare a game of chess - which has a chance of being the same twice because every piece has only 64 places to be (excluding bishops), with snooker where positions are infinite.

MitchT said:
So why is matching six seen as so much bigger a deal than matching five plus the bonus? Surely the odds are the same.
Marketing I expect.

Sarkmeister

1,678 posts

225 months

Saturday 18th June 2016
quotequote all
MitchT said:
Here's another silly question.

If there are six balls plus a 'bonus ball', then there are seven balls in total.

If you match six balls, then you've matched six balls.
If you match five balls plus the bonus ball, then you've matched six balls.

So why is matching six seen as so much bigger a deal than matching five plus the bonus? Surely the odds are the same.
Surely matching 5 + bonus ball is more likely because there are 6 possible combinations (i.e. swapping one of the 6 numbers out for the bonus ball)? There is only one possible way of getting the main 6 balls.

Sarkmeister

1,678 posts

225 months

Saturday 18th June 2016
quotequote all
Also, to answer the first questions, to further make the draw more random, the "go" button is pressed at a non specified time after the machine starts spinning. This means the spinning things in the drum aren't always in the same position.

ferrariF50lover

Original Poster:

1,834 posts

233 months

Saturday 18th June 2016
quotequote all
Sarkmeister said:
Also, to answer the first questions, to further make the draw more random, the "go" button is pressed at a non specified time after the machine starts spinning. This means the spinning things in the drum aren't always in the same position.
Ah, ok, hadn't spotted this (I've seen the Lottery show perhaps twice in my life).

In order to perhaps illicit some further interesting answers, can we please pretend that your version isnt true?

Sarkmeister

1,678 posts

225 months

Sunday 19th June 2016
quotequote all
Yeah, sorry about that. Ignore my post above.... smile

V8LM

5,269 posts

216 months

Sunday 19th June 2016
quotequote all
MitchT said:
Here's another silly question.

If there are six balls plus a 'bonus ball', then there are seven balls in total.

If you match six balls, then you've matched six balls.
If you match five balls plus the bonus ball, then you've matched six balls.

So why is matching six seen as so much bigger a deal than matching five plus the bonus? Surely the odds are the same.
It isn't the same: the first is matching all six that you have chosen with odds 1 in 45 million. The odds of picking five of the six is 1 in 144 thousand. The odds of matching the bonus ball from the 52 numbers left is 1 in 52. The odds of matching five plus the bonus is therefore 1 in 7 million.

V8LM

5,269 posts

216 months

Sunday 19th June 2016
quotequote all
ferrariF50lover said:
Ah, ok, hadn't spotted this (I've seen the Lottery show perhaps twice in my life).

In order to perhaps illicit some further interesting answers, can we please pretend that your version isnt true?
As said above - chaos.

Even if the button was pressed at exactly the same time relative to the position of the rotating paddles and these are rotating at exactly the same constant speed, the tube that the balls fall down isn't exactly the same as the diameter of the balls, so they will fall from a slightly differently place each time. Minute imperfections in the shape of the paddles and balls and their mechanical properties etc means that the variance in direction of travel after each collision multiplies until only after a few does the uncertainty in position of each ball become larger than the size of the drum. It's similar to trying to predict the position of a cue ball in snooker after bouncing off the cushions. After something like 12 bounces the uncertainty in position is more than the length of the table.

Edited by V8LM on Sunday 19th June 08:51

FiF

45,549 posts

258 months

Sunday 19th June 2016
quotequote all
When the lottery first started there were a relatively limited number of machine and ball combinations. Based on the thought that not all are equal, there must be some manufacturing tolerances, a study was started to see if particular numbers were likely to arise for particular machine/ ball combinations.

Iirc on one combination a trend started to show, but wasn't significant that could be definitely shown as outside random chance. Shortly after the number of machines and sets of balls was increased and don't know what happened after that. No doubt someone will have studied it somewhere. Probably in Camelot or whoever runs it now.*



  • *Just looked it up, same outfit that owns Cadburys apparently. Don't give the lizards your money.

thebraketester

14,712 posts

145 months

Sunday 19th June 2016
quotequote all
Take two table tennis balls and drop them from the same height onto the same surface. They won't bounce the same.

Moonhawk

10,730 posts

226 months

Sunday 19th June 2016
quotequote all
ferrariF50lover said:
If I take a tennis ball and drop it 1000 times from the same height, onto the same surface in air that's the same density etc, it'll bounce to the same height. That's how physics works. If nothing changes, nothing changes.
It wont though as the tennis ball is not a perfect sphere.

Even tiny imperfections in the ball/surface, thickness of the ball material, slight changes in air pressure/currents etc will lead to a slightly different result each time.

Although the effect on a single tennis ball bouncing off a flat surface will be small - in a more complex system like a lottery machine with multiple balls, multiple surfaces etc - the effects can add up to a significant difference.

The basic premise of your argument that "if nothing changes" is fundamentally incorrect. Every time you run the experiment - things have always changed - even if this is every so slightly.

Google "butterfly effect".


Edited by Moonhawk on Sunday 19th June 12:15

ferrariF50lover

Original Poster:

1,834 posts

233 months

Sunday 19th June 2016
quotequote all
And there we have our answer. Thank you, chap(esse)s.

4x4Tyke

6,506 posts

139 months

Monday 20th June 2016
quotequote all
The setup is not perfectly identical each time, there are multiple machines, sets of balls and timing.

It is a chaotic system, highly sensitive to the tiniest variations in initial conditions.

Google the Butterfly effect, then consider each impact between balls and rotor is separate randomising event.

Foliage

3,861 posts

129 months

Monday 20th June 2016
quotequote all
Its irrelevant, the draws are fixed the balls are pushed through from the back of the machine, we end up with roll overs on bank holidays constantly, their are videos online of the ball having been drawn and still in the drum.

The Casino always wins.

FrankAbagnale

1,730 posts

119 months

Monday 20th June 2016
quotequote all
Do we have the technology or knowledge to be able to replicate a lottery draw? In lab conditions using computers to drop the balls with preset times etc... Or does chaos theory require such a minor imperfection that we are unable to achieve such a feat?

4x4Tyke

6,506 posts

139 months

Monday 20th June 2016
quotequote all
I don't suppose it is even theoretically possible, Quantum Physics would pretty much rule out even infinitesimal levels of tolerance necessary to make it possible.

feef

5,206 posts

190 months

Monday 20th June 2016
quotequote all
Theoretically, if you can be sure the balls are identical both in shape, size and elasticity, and everything else is perfect as well as timed to be perfectly in sync, then there's nothing to stop the result being the same

The closest, practical application is in something like this:

https://www.youtube.com/watch?v=IvUU8joBb1Q

Where the balls and mechanisms are within sufficient tolerances to ensure consistency of result

It's those tolerances of both physical properties of the balls and mechanism and timing that are crucial

944fan

4,962 posts

192 months

Monday 20th June 2016
quotequote all
The is a book by James Gleick (spelling?) called Chaos. Doesn't cover lottery per see but covers chaos theory in an easy to understand way. Really interesting.