True probability

[burdo77]

I have a question on how to assess the true probability, for example the 0-0 draw.

Let's say the actual probability of a 0-0 draw is 10%, however is the 10% a product of say, 15% of occurrences at odds of 10 and 5% of occurrences at odds of 20?

[Wyndon]

There is only one 'true' probability, but unless you are talking about games of chance where the odds are known (eg roulette, dice), no one can tell you what it is. You have to make your own estimate/judgement when it comes to a sporting event such as a horse race or football match. The probability of a single event can't be 15% AND 5%. To take your example over 100 occurrences. If there are 10 successes, your bets guess is 10%. 15 successes, 15% etc. They can't all be true. You seem to be saying (in your example) that with odds of 10.0 you are successful 15% of the time; and with odds of 20.0 you are successful 5% of the time. These are different initial conditions that affect the likelihood of your winning. Common sense dictates that you should only bet when the odds are 10.0, when the odds are heavily in your favour. In the 20.0 odds, 5% successes scenario, you could only expect to get your money back in the long term.

[burdo77]

Hi wyndon, sorry if I didn't explain myself correctly but I'll explain again. Let's say I'm developing a system . Statistics tell me that 0-0 draw has a prob of 0.08, this probability includes all games which is a mix of all games where the pre game odds were 10-1, 15-1,20-1,25-1,30-1,35-1. So if I was to investigate the true probability of a 0-0 draw of games where the odds started at 10-1, would it be a combination of what actually occurs (prob 0.08) and the implied probability at 10-1(prob 0.1), or something else?

[Wyndon]

Hi burdo,

I don't bet on football, but I imagine the situation that applies in horse racing is comparable. Let's say the average odds of a favourite are 4.5 (a guess). Some favourites will start at much shorter odds than the average and win much more often than the average odds imply. Similarly favourites with much longer odds will win less frequently, despite being favourite. If the odds are different from the average this is a reflection of what the market perceives to be the true chance. The first guess of a 0-0 occurring where the odds start at 10 would be 10% - in the absence of any other information or edge you might have. It contributes to what makes up the average, not the other way around.

[marko236]

It all depends on how good you are at predicting the outcome, i'm putting bets on the Sassuolo v Juventus what i do is imaging this game getting played 10 times and then work out the percentage chance of the result.

I'v went for Sassuolo 10% chance of winning, 20% the draw and 70% for Juventus win.

I will also be looking to make Dutching bets and any chance of a good trade.

[Euler]

The chance of 0-0 is ultimately a function of the draw price.

In the EPL, historically, 32% of all draws are 0-0

So say the draw is priced at 4.00, that equals 25% in IP terms.

So .32*.25 = 0.08 or 8% or 12.50 in decimal odds.

[burdo77]

Thanks for your input guys , Eulers input makes the most sense, cheers.