I think the “closing prices efficiency” is the favorite subject for many punters for so many years. There are two basic questions regarding closing prices:
- Are the closing prices the “real prices” of a betting market?
- Can we use these prices to evaluate our predictions?
I will answer the questions above using my favorite tools: a large data pool and mathematics.
There will be four articles in this series to warm up before I start posting my analysis and predictions:
- What happens if the home win % drops or rises (2 articles)
- What happens if the away win % rises or rises (2 articles)
First of all, I want to thank the people of www.football-data.co.uk for their great work. I took the data for my analysis from the Historical Data section under the URL https://www.football-data.co.uk/data.php.
My dataset has 62.917 records; each represents a unique football game from 2012-2013 to 2020-2021for 22 European football divisions; the odds are from Pinnacle.
The starting odds are gathered either Friday afternoon for weekend games or Tuesday afternoon for midweek games.
I created ten odds groups, basically the same I used in MATH to WIN: Football, to conduct a more detailed and reliable data analysis.
So, let’s get to it! The first table describes the full dataset and presents the winning % of each result combined with the capital return.
As you can see (and expect), there are losses for us almost for every situation. So we can suppose fairly accurately that there is a market efficiency in the closing prices; the bookies profit in every situation, mainly because of the overround.
Now, let’s see what happens if the probability of the home team winning the game drops more than 1% at the closing line (meaning the odds for the home win odds rising 1%).
We can see some “life” for the draw when the home team wasn’t strong and some earnings for the very strong guests.
Now, let’s see what happens if the probability of the home team winning the game drops more than 2% at the closing line (meaning the odds for the home win odds rising 2%).
You get the idea, but the next tables are more informative and useful. I will present three tables with the home win, draw and guest win probabilities and capital return when the probability of home win drops from -1% to -4% at closing time.
It is an obvious remark, for the games where the guest team was a very small favorite and the odds for home win were bigger at closing time (meaning the % of home team winning was smaller). The next two tables presents the capital return for DRAW and for the AWAY WIN.
Well, another good one. When the home team starts as clear favorite but the odds rising then the draw is a good selection.