View original (French)

Sports betting project

Fab38 -  
 Fab38 -
Hello,

I'm 17 years old and for purely educational purposes (I'm not an adult anyway), I've been wondering if it's really possible to profit from sports betting.

I've tested a few things based on expectation and I'm at +44% over 20 matches, which is encouraging even though I think it won't last.

I came across someone who created an algorithm in R that predicts which matches to bet on, and it performed quite well.

I'd like to get it running to see, but the problem is that I'm at level 0 when it comes to programming. I'm looking for help to run this algorithm, the code is freely available here:
https://github.com/tuangauss/DataScienceProjects/tree/master/R/EPL

So, if anyone is knowledgeable and would like to help me, that would be nice. I think there’s something to explore with betting, even though I believe it’s impossible to achieve a high percentage in the long term.

1 réponse

Réponse 1 / 1
trifou
 
Hello,

You know very well that if it were possible to win in the long term at certain gambling games, those games would no longer exist; gambling is like alcohol and tobacco, one can quickly become dependent on them, and besides the financial drain, there are collateral damages, health, relationships with those around you, etc. It can quickly become a descent into hell. You're young, find yourself some other interests ;)

To install R, you need to follow the instructions as described on the website page.
https://cran.r-project.org/doc/manuals/r-release/R-admin.html#Obtaining-R
Choose a mirror from this page.
https://cran.r-project.org/mirrors.html
Then follow the links to download an executable.
0
Fab38
 
Obviously, collectively, gambling games are a loss for the group; looking at it from a distance, betting is a transfer of money between losers and winners, and bookmakers take a share of that transfer. However, on an individual level, I think one can be a winner. Of course, living off betting is impossible and not very exciting, in my opinion... However, don't worry, I have other passions, I don't bet, and I know the risks; it was just a test like others I do (woodworking, etc...).
0
Fab38 > Fab38
 
I would add in my current first test with +44% that I can see that most matches have a negative expectation.
I've done some stats, so here are some figures from my experience over 20 matches (considered a small sample):
_The expectation is positive in only 1/3 of the matches (the matches played are chosen only if the expectation is positive)
_The average odds for these matches is 3
_It's better to bet on the opponent of the favorite
_The expectation of the won matches is lower than that of the lost matches (I think this comes from the over-evaluation of the winning percentage from the site providing this information)
_The trend curve of gains is about 1.6x + 6.45 where x is the number of matches and for a strict bet of €10, which shows that the total gains are increasingly significant (no stagnation) but I think this coefficient will decrease
_I hardly ever play Draw Matches (I only play 1x2)
_The most consecutive losses: 8 (-€80) and victories: 3 (+€71)
_Winning percentage is between 1/3 and 40%
_Considering that I increase the stake as my gains progress, the gains curve is exponential (possibly incorrect info)

I'll see with time; anyway, I don't bet anything as I'm not of age, I simply take results and see if my thinking is correct (which is that by betting on matches with positive expectation, we make money). In my opinion, the trend described so far seems a bit excessive; I think it will calm down.
1
Fab38 > Fab38
 
Small update: after 36 matches

the trend curve has reversed, it is now equal to: -1.12x +34.02 which shows that we make money at the beginning but lose it in the long term, I am down about -66€ and I’m waiting to drop to -100 (hypothetical starting budget of 100€)

It’s still early to draw conclusions but I would say that my technique worked a little, in the sense that I would have lost more quickly just by predicting based on my opinion, if the bookmakers had a lower margin I think it would work, however, it’s clear that it doesn’t work now.
0