Well my "Boffin" replied with this:
"I think that the formula you have on the site is probably as good as you can hope to achieve."
He added that with a house as a prize even it's actual relative value needs to be varied and is only truly relevant at the time of the draw.
That isn't what any of us wanted to hear I know - we wanted an easy to apply method to assess how fair a competition was but I guess it does make sense. Anyone else got any comments?
Can we create a Winning Formula?
Can we create a Winning Formula?
Group Option
Admins
Members
Group Information
Category:
Name:
Can we create a Winning Formula?
Created:
Tuesday, 14 April 2009
Creator:
Description
Whilst most of us realize we are making an emotional rather than logical decision when we enter a competition, could a logical formula be created to choose? If possible (and I believe with enough thought it is) perhaps we could more easily confirm our "gut" feelings?Announcements
No bulletin added yet
Discussions
No discussion added yet.
Photo Albums
No album created yet.
Videos
There are no videos added yet
Wall
Thanks for the input Number One, I have some statistical contacts who may be able to add some comment - will ask and see what we can get.....
Wouldn't it be great though to be able to "measure" a competition. Once we have created a tool to do it it may even be possible to apply it to other competitions too.
Wouldn't it be great though to be able to "measure" a competition. Once we have created a tool to do it it may even be possible to apply it to other competitions too.
Saturday, 18 April 2009 12:09
Also before I forget good point WinOurHome about some of these 'skill' contests actually ending up being about luck 'cause you have to assume anyone that got it correct is put into the hat along with everyone else!
Wednesday, 15 April 2009 19:06
Bit rusty on my option theory but seems to me that Win a House is on the right track... In theory you should be able to work out the present value of the payoffs at each step/scenario and work backwards to get a single formula... Although I'm not sure how to work out some of these probabilities, anyone got some spare time and good maths skills?! I'm off to make a cup of tea :)
Wednesday, 15 April 2009 19:04
A decision tree or flowchart is a possibly but what I had in mind was a formula of some sort.
it must be simple as life is way too complicated already. Or too formal - entering a competition has to be fun.
But I opened this discussion because I don't know the answer myself :0)
Could it be something like ?.....
First assessing the value for money,
Value factor = Prize / max tickets / ticket cost per entry
which for our comp = 300k / 18500 / 12.5 = 1.3
If our prize was 100k then factor would be 0.4 so, lower the number, lower the value for money?
Then that figure should increase with higher probability that you could be the winner.
With a spot the ball comp like ours there is a high level of skill combined with a significant element of chance because with thousands of pixel positions to choose from, it is unlikely that many people will get it right. This means that the winner stands out and does not need to be chosen by the organisers.
Other pure skill methods like Sudoku or crosswords mean that thousands of entrants could get through therefore placing much more reliance on the organiser's method of pulling the winners name out of a hat. Perversely, does this makes it more of a game of chance than of skill?
Now this is where I start to get a bit lost trying to measure value. Although I have not yet figured it out fully, my instinct tells me that the more the player controls vs the organiser, the better the players chances which equates to better value.
This is a work in progress as you can tell so - all input is appreciated!
it must be simple as life is way too complicated already. Or too formal - entering a competition has to be fun.
But I opened this discussion because I don't know the answer myself :0)
Could it be something like ?.....
First assessing the value for money,
Value factor = Prize / max tickets / ticket cost per entry
which for our comp = 300k / 18500 / 12.5 = 1.3
If our prize was 100k then factor would be 0.4 so, lower the number, lower the value for money?
Then that figure should increase with higher probability that you could be the winner.
With a spot the ball comp like ours there is a high level of skill combined with a significant element of chance because with thousands of pixel positions to choose from, it is unlikely that many people will get it right. This means that the winner stands out and does not need to be chosen by the organisers.
Other pure skill methods like Sudoku or crosswords mean that thousands of entrants could get through therefore placing much more reliance on the organiser's method of pulling the winners name out of a hat. Perversely, does this makes it more of a game of chance than of skill?
Now this is where I start to get a bit lost trying to measure value. Although I have not yet figured it out fully, my instinct tells me that the more the player controls vs the organiser, the better the players chances which equates to better value.
This is a work in progress as you can tell so - all input is appreciated!
Tuesday, 14 April 2009 20:12
Interesting discussion Paul - are you thinking of building a decision tree to compare the different scenarios a player might consider before buying a ticket in a house competition?
For a start, I guess this could involve defining the expected pay-offs depending on the likely probabilities assigned to each outcome.
This might also involve assigning weights to different criteria - to factor in the different personalities of individual players. While some people might assign a high value to winning cash, others may determine that the lifestyle changes resulting from winning a home would go well beyond the simple financial gain implied by the valuation of the property. After all, two people looking at the same property may have completely different reactions.
The assessment of the likelihood of winning may also be subjective depending on the type of competition. For instance, your house competition is based on Spot the Ball - two players may have differing views about how difficult (or how easy) it may be to pick the winning spot. While it may be tempting to simply look at the maximum number of entries (or more accurately, the probable number of entries), with any game of skill one can't simply focus on the 'odds' of winning in this way.
As a result, two players may come to different conclusions in terms of their chances of winning the competition - despite being faced with the same information before entering. In the same way, the results may vary quite a lot if you change one of the pieces of the jigsaw: based on a different photo but keeping all other factors the same, it is also likely that the same person may form a very different opinion.
This also suggests it may be challenging to come up with a model that allows you to compare across different types of competitions at the same time (Spot the Ball vs. quiz vs. Sudoku etc) - which is what players are required to do before buying a single ticket.
Quantifying our "gut feelings" remains one of the more challenging things to understand. However, it is likely that most players go through these thought processes and assign probabilities to different outcomes in a very intuitive way and without putting much thought to them.
Putting these ideas on paper and assigning weights and values to these different scenarios can be difficult. That said, it could serve as a useful rule of thumb for anyone tempted to enter more than one house competition and who is unsure how to make a decision.
For a start, I guess this could involve defining the expected pay-offs depending on the likely probabilities assigned to each outcome.
This might also involve assigning weights to different criteria - to factor in the different personalities of individual players. While some people might assign a high value to winning cash, others may determine that the lifestyle changes resulting from winning a home would go well beyond the simple financial gain implied by the valuation of the property. After all, two people looking at the same property may have completely different reactions.
The assessment of the likelihood of winning may also be subjective depending on the type of competition. For instance, your house competition is based on Spot the Ball - two players may have differing views about how difficult (or how easy) it may be to pick the winning spot. While it may be tempting to simply look at the maximum number of entries (or more accurately, the probable number of entries), with any game of skill one can't simply focus on the 'odds' of winning in this way.
As a result, two players may come to different conclusions in terms of their chances of winning the competition - despite being faced with the same information before entering. In the same way, the results may vary quite a lot if you change one of the pieces of the jigsaw: based on a different photo but keeping all other factors the same, it is also likely that the same person may form a very different opinion.
This also suggests it may be challenging to come up with a model that allows you to compare across different types of competitions at the same time (Spot the Ball vs. quiz vs. Sudoku etc) - which is what players are required to do before buying a single ticket.
Quantifying our "gut feelings" remains one of the more challenging things to understand. However, it is likely that most players go through these thought processes and assign probabilities to different outcomes in a very intuitive way and without putting much thought to them.
Putting these ideas on paper and assigning weights and values to these different scenarios can be difficult. That said, it could serve as a useful rule of thumb for anyone tempted to enter more than one house competition and who is unsure how to make a decision.
Tuesday, 14 April 2009 11:30
Hi Paul this sounds interesting I'll join your group to learn more about your thinking :)
Tuesday, 14 April 2009 10:42