# What % gambling is sports betting

what % gambling is sports betting

All urls found in this thread:

It's Roulette-tier gambling

he doesn't think sport games are rigged

Versing actuaries, mathematicians, statisticians and engineers that determine the odds of x winning the game
Versing actuaries, mathematicians, statisticians and engineers that determine the odds of x winning the game with a markup so the company makes money.

Yeah, nah, roulette with the payout even more against you.

Betting made me suicidal, in the long run you will always end up losing.

Last year of career I made probit/logit regression models to see if the house could be beaten in soccer (ordered logit/probit) and basket leagues of my country.
I used public available data, so there weren't many variables, just 6-8 depending on the model.
Procedure was simple, get estimated probabilities of the different outcomes and bet to those that were higher than the house estimations (taking into account their % of fee (around 5% in my country bc high taxes), plus a 2-5% margin of security). If my model was better than theirs, I should win money, and vice versa.
I used 5-9 (I don't remember) past seasons of data, and made the test on the last one from which I didnt gather any data.
Net result after betting to all matches (that complied the requisites) on the season was 0. So I could say that the model was good enough to compensate for the 5% house fee, but not enough to make profits.

The problem and the difficult part is getting data about the matches. I'm sure If I had access to that, the model could out-perform the house's one (that is made, theoretically (if the house doesn't have shaddy behaviors), from supply-demand of the different outcomes)

Trying to beat actuaries, mathematicians, statisticians and engineers at their own game

Idk how their odds are made, but it's either:

1. Supply-demand of bettors. They have their fee. 0 risk, they get more profit the higher the volume of bets. The classic way.

2. They make models (regression, neural networks or wathever). And compare those against the supply-demand one (the same I'm doing). And then leverage based on their predictions. Low to zero risks bc enough volume to reduce variance, and higher profits than 1.

If they do 1, you can win money against their odds since they are market based, with a good model. If they do 2, you are competing against their model, and then is when you are
Trying to beat actuaries, mathematicians, statisticians and engineers at their own game

I did this out of boredom. Not a single coin was betted

I think they do 1, so you don't have to beat any actuaries, mathematicians, etc. The reason I think this is:
- Number 2 is illegal afaik. Can't find sources now, but I remember reading somewere that betting houses weren't allowed to do financial operations incurring in risk to get bigger profits.

- The theorethical net of my model was around 0 profit while they had 5% fee. That means their models are at least 5% off. That's not possible if there were any mathematicians/staticians behind

It all depends on the sport. Some, like baseball, are notoriously hard to beat. Others, like NASCAR, MMA, korean women's volleyball, etc. are relatively easy

I simply refuse to believe that a betting company isn't allowed to make statistical models. Isn't that what the lottery, scratchy ticket things, even insurance is.

If they do 1. and only 1 i'm sure you can make some money out of it, i'd just be super suprised if they didn't do 2 legally or illegally .

I didn't say they aren't allowed to make models. They do them for sure, they need it to make the first odds before opening to the market the bets (to be stable on the first bets and avoid big jumps).
Even if illegal, they could do it with a third party betting on the statistically winning party, but to move the odds even a 2-3% would require vast sums of money and non 0 risk of the classic way.
But if they do 2, then it makes no sense than I could make the 5% of their fee as profit and achieve a net 0 after considering an entire season of bets.
Maybe they have different policies depending on the sport. My models were based on basket/soccer first/second division of spain. Numbers were better at basket, I think it was bc it's less known (less people competing, less perfect market), or bc having 2 outcomes really diminished variance compared to soccer.

Does anyone ITT have any model they follow in sports betting that have given them good return?
I'm talking medium-long term approaches rather than getting trigger happy on a random weekend..

Go on TheMatchedBettingBlog.com

Guarenteed £500 a month risk free. I made about 400 myself last month, and that wasn't with a great deal of time inputted.

Punting is for plebs, abuse their systems and profit by exploiting them how they exploit others.

You guys are all fucking clueless.

Wont you get banned? Then you will have to buy fake ids to continue which is fraud.

Most bookies nowadays just follow exchange prices or prices in asia then just balance, they don't really care if the price is accurate though it often will be because most of the money in markets is by people who know what theyre doing.

mma is very beatable if you watch for 10 years and really think about why someone lost / won

mma gets underdogs who have a really good shot or even gaurantee of winning ie miocic vs werdum, mcgregor vs diaz

but its not worth attempting unless youve been watching a really long time

other sports are probably not beatable, but idk i dont watch them

You're clueless about sports betting unfortunatley.

not an argument :)

I know, just a fact