Wisdom of crowds

Hi -just entered mine. I think you will find that the bookmakers prices will most accurately reflect the crowd already.

✔

The bookmakers prices will reflect their mathematical view of the chances of each outcome, but the crowd opinion of the result will be different with the very short priced picks almost unanimous. GL.

On betfair it's sometimes a good play to go against the crowd's wisdom ......

www.midasoracle.org This guy likes his prediction markets.

crowd intelligence works well on who wants to be a millionaire

thanks this is an interesting idea.

i'm not a statistician but i'm not sure that offering 3 selections on 5 events offers enough potential for the crowd's selection to be at variance with the bookie's selections as reflected in their offered odds.

it might be more interesting to ask crowd to predict number of goals?

maybe it would work better in cricket or golf?

* crowd intelligence works well on who wants to be a millionaire

Except when they get it wrong

topkat, find the goals takes alot more indepth study than 1X2.

People see certain teams as giants and cannot see them scoring atleast "3 or 4". When doing U/O, I find that not only do you have to look at the teams but then look indepth on how the underdog will defend and if they will be able to attack.

The average person looks at the players, not the match.

Agree with Kenilworth, you will find if you bet on the "wisdom" then you would be lumping on Chelsea at home at anything above 1.05 :)

crowd intelligence works well on who wants to be a millionaire

They are not trying to tell the future.

It's the wisdom of money, not crowds. 1000 people with one opinion can be "out wised" by one person with a large bank.

"The market can remain irrational longer than I can remain solvent."

Agree with Topkat. Most will probably pick the fav in each match.A better test of this may be to select a premier league game and have 'the crowd' accurately guess the attendance.

Interesting idea. Will have a look.

The theory behind this makes perfect sense, its just the mathematics that may let you down.

Surely the wisdom of the crowd is the percentages implied in the Match Odds market? A sub-set of this larger crowd will be less likely to be accurate than the full set.

Hi everyone. The results of the experiment are now online

Goto www.simonwalters.net to see the results.

Thanks to the 53 people who took part.

The wisdom of the crowd ...
Bolton 25/1
Blackburn draw 8/11
Burnley 25/1
Wolves 50/1
. :)

If you had backed each favourite AND the draw in each match, using multiples you would have made a nice profit as each 2-way forecast came up. When do you anticipate repeating this event as I like the idea?

Hi all. So it's crunch time. If you would like to see this experiment progress please goto www.simonwalters.net and vote

I dont want to flog a dead horse so if your not interested then thats more then fair enough.

If you are interested, however, then please vote so I will no if there is interest to carry on or not.

Cheers.

zero-zero in san siro 07 Nov 20:04


If you had backed each favourite AND the draw in each match, zero,that is called back fitting.

Kenilworth quite right , you will simply end up backing every favourite regardless of the odds.

Also , what this was originally done for was the weight of a cow at a fayre i think. There was only one possible correct outcome to that question, as it was known before the question was asked.With this it is impossible for the crowd to 'know' who will win , as it is unknowable , so the question they are answering really is ''Who is most likely to win?'' , which anyone with the faintest clue about football will know anyway.

As for the Millionaire Crowd....

They are good at ''Who owns the Pub in Eastenders?'

not so clever at say ''Who painted 'The Raft Of The Medusa'? ''

kenilworth 07 Nov 12:18
The wisdom of the crowd ...
Bolton 25/1
Blackburn draw 8/11
Burnley 25/1
Wolves 50/1
. :)

Not very realistic odds.

I tried something like this on another forum years ago. Instead of a straight pick of 1X2 I asked how many each of 1X2 there would be if two teams played 10 times, i.e. Liverpool-Birmingham 1=7, X=2, 2=1.

I agree and the betting would agree, but for one match it becomes 90+ percent for the favourite it seems.

Sent a comment to you....try UNDER OVERS on select matches.

I've got data coming into me from all angles regarding where punters are putting their money, those above who say the favs will be most heavily backed are correct, Liverpool took almost 88% of the money tonight and not a bookmaker in the land shifted the price.

Betting against your %'s once they are posted will reap rewards, but only if you can get enough people involved, I havent looked back since I started. Mind you it may not reap rewards, the data I'm getting includes the stakes of each punter, Liverpool and Millwall might both have been gambled to the same % tonight but then there was 10 times as much money on Liverpool as there was on Millwall, which makes a huge difference to merely betting against opinions.

This round of games in the UK were close to being my first losing week, Liverpool brought us comfortably over the profit line in the end.

Keep going though, the 'Wisdom of crowds' aint shutting down bookmakers anytime soon

clacherholiday2, agree with what you said. I spend alot of time watching $$$$ on the U.S. market for baseball. What I discovered is that certain teams are PERCEIVED to win all the time (Yankees, Detroit, Red Sox) because they are who they are. The amount of money that goes on perceived teams on a weekly basis has a much higher average than others.

From there, I found two triggers (bearing in mind, value would be key):

If a (best wording I come up with ;) ) NON-Perceived team was playing perceived team and the support on the Non caused the money down to be tipped their way, it was a strong signal that the "real money" was down.

If a NON was playing a NON, then if the money was exponentially greater on one team, it usually played dividends.

A matter of keying in teams (baseball has 30 teams/182 games a season so easily overwhelming) and building the flags accordingly.


When I discussed this with someone else, he took it further on two points:
a) money during the week was more significant since the pros were playing
b) People in the states get paid every two weeks, so, again, a factor that causes obvious flux.
c) Best to ignore money on the weekend for baseball
d) if the average on a PERCEIVED TEAM is way down at any time and the money on a NON is up, that means that the "real money" has shifted that way.

The headline swaying of the betting public is not limited to the X-FACTOR.

Premierpicks on facebook does this with over 120,000 people.

Like most have already stated you will be basically backing the favourite so you will have to compare the prediction percentage to the actual percentage from the match odds.
Man u 80% win from group
Man u 70% win from match odds
You would then back Man u in the above example as the group % is higher, or do the opposite, at the end of the day you will end up with similar results, most probably showing losses.
It would be better to focus on goals scored or corners in the day, something people can put a number to without as much knowledge on the subject so it's more like guess-work.

good thread

lol.  im always coming across threads from the past ive contributed too and totally forgotten about.

yes good to bring them up sometimes clacher

theres one i posted on years ago that i still get pm's about, ill see if i can find it

thanks will look forward to reading it

cant find it, normally im pretty good at dredging up old crap from here