Why do favourites only win 1/3 of the time?

[Brovashift]

Hi all,

Been looking at racing data recently and was wondering; why do favourites only win 1/3 of the time?
Why not more... why not less?

[jimibt]

on average... otherwise, we'd all just be plonking ££ on high priced faves and waiting for the money to roll in. it's all about market efficiency and the backroom boys setting the odds fairly accurately.

i guess that's really where value betting comes in - spotting either those runners that are rated too high or too low to their true odds.

[Derek27]

Why am I 5 feet 8 inches? Why not more, why not less?

[Trader724]

There are as many answers as there are variables in each race.

[Euler]

Here is your detailed answer: -

https://youtu.be/wzkuIsbEa1o

[ShaunWhite]

2.0 horses win 50%, 10.0 horses win 10% and favourites have an average price of X so win (100/X) % of the time. Market efficiency. If they averaged 4.0 they'd win 25% of the time, if not then we'd all be rich.

[ShaunWhite]

btw Brovashift, It's market efficiency that makes backtesting betting or tradings strategies pretty straightforward, you don't even need the results data you were looking for and actually shouldn't use it for that purpose. Quite simply if you're obtaining a net position that's better than BSP you'll be up and if you're placing bets or getting a net trading position worse than BSP you'll be down. BSP has been used as a simple benchmark for donkey's years across all markets and sports. In fact using what won or lost instead of whether you beat BSP will just give you misleading skewed results which fall into the trap of backfitting and being perpetually on the wrong side of mean reversion.

[Anbell]

Because the average price of a favorite is 3.0

[ShaunWhite]

Because the average price of a favorite is 3.0

Seeing a rational number in a data question was bad enough without you encouraging him with integer averages.

[Brovashift]

btw Brovashift, It's market efficiency that makes backtesting betting or tradings strategies pretty straightforward, you don't even need the results data you were looking for and actually shouldn't use it for that purpose. Quite simply if you're obtaining a net position that's better than BSP you'll be up and if you're placing bets or getting a net trading position worse than BSP you'll be down. BSP has been used as a simple benchmark for donkey's years across all markets and sports. In fact using what won or lost instead of whether you beat BSP will just give you misleading skewed results which fall into the trap of backfitting and being perpetually on the wrong side of mean reversion.

Hi Shaun,

Is this true of a lay strategy as well? I assume with a lay you'd want a price below BSP?
I am collecting win data on 1st, 2nd, and 3rd favs for lay purposes. I was thinking if a particular fav has more losses than wins at a particular course, or distance, etc. then this might yield a positive expectancy.
Obviously Im aware of the potential liability risks, and my concern is that 'What if' I have 30 winners in one session. Highly unlikely I know, but not impossible. I've only collected 1 weeks worth of data so far, and although positive, still early days.
Should I still be looking to beat BSP, even with a positive lay expectancy? or am I going to end up with skewed results? I want to make sure Im doing it right and not wasting my time

TIA

[Brovashift]

Here is your detailed answer: -

https://youtu.be/wzkuIsbEa1o

Thanks Peter, very informative

[Brovashift]

Why am I 5 feet 8 inches? Why not more, why not less?