Skip to content

Useful library to perform better on quiniela bets

License

Unknown, MIT licenses found

Licenses found

Unknown
LICENSE
MIT
LICENSE.md
Notifications You must be signed in to change notification settings

MiguelAngelTorres/quini-utils

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

95 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

quini-utils

R library to perform better on quiniela bets.

How this works

Standard quiniela's players think that the best way to earn money in quiniela is to be right in as many matches as you can. So, in order to achieve that, they usually bet that the team with more probabilities of winning will be the winner of the match.

However, the prize in quiniela is not fixed, it is calculated as the collected money in bets at this week minus 40% of government taxes. Besides, the prize is splitted between the right bets. That means that to win quiniela when so many others fail is better than win quiniela when many others do. But, as expected, if you select rare bets with low probability you will never win the game. The point is to balance between bets low voted and probably bets.

The government updates the % of people that are voting every result at every match. You can't know exactly what combinations were done in the 14 matches, but you can make a good estimation. In the other hand, there are several betting houses publishing the probability of winning for every match. In a graph would be:

explain

Now, the quini-utils library comes. It allows you to download this information and calculate the probability of being right in x matches and the probability of having other people with x matches right for every single combination. Many people bet by dividing probably by voted percentages, which is called profitability, and ordering that to select the most profitability combinations.

However, you can also calculate the mathematical expectation (EM), which calculus is a bit complex. It will take a while to calculate (about 30 minutes for EM10, EM11, EM12, EM13 and EM14 all together, 10 min for EM12, EM13 and EM14) but it is worth it.

Install

Install last version from github

install_github("MiguelAngelTorres/quini-utils")

Examples

How to select 10 best bets ordered by EM12

library(quiniutils)

Get matches information from EduardoLosilla website

matches <- get_data_eduardolosilla('2023', '35')

Eduardo

Calculate the real and voted probabilities

table_prob <- calculate_probabilities(matches)

prob_voted

Calculate the EM of each bet

table_prob <- get_em(table_prob, money = money, level = 12, optimization = TRUE)

EM

Pick the best 10 bets

selected <- table_prob[order(-em_12)][1:10]

Get the output to vote

write_bet("selected.txt", selected)

About

Useful library to perform better on quiniela bets

Topics

Resources

License

Unknown, MIT licenses found

Licenses found

Unknown
LICENSE
MIT
LICENSE.md

Stars

Watchers

Forks

Packages

No packages published

Languages