Betting Tips Feb 11, 2018
Yesterday we did well, let's do it again today.
| Home | Away | Bet | EU Odds | UK Odds | US Odds |
|---|---|---|---|---|---|
| Torino | Udinese | Draw | 3.75 | 11/4 | +275 |
| Southampton | Liverpool | 2&BTTS | 3.40 | 12/5 | +240 |
| Nantes | Lille | Draw | 3.40 | 12/5 | +240 |
Good luck!
Yesterday we did well, let's do it again today.
| Home | Away | Bet | EU Odds | UK Odds | US Odds |
|---|---|---|---|---|---|
| Torino | Udinese | Draw | 3.75 | 11/4 | +275 |
| Southampton | Liverpool | 2&BTTS | 3.40 | 12/5 | +240 |
| Nantes | Lille | Draw | 3.40 | 12/5 | +240 |
Good luck!
Good luck @beat-the-bookies! What methods are you using to make these picks?
The Naive Bayes classifier and intuition.
The Naive Bayes classifier performs well enough and is simple to implement. I tried different ML algorithms but didn't see much improvement, if anything, most of them under-perform NB.
I don't have (yet) an automated system. Instead I run a bunch of scripts to scrape data(results, odds,...), run NB on it and see if there are any games where my odds differ significantly from the bookmakers. If there are some games that pique my interest I take a closer look.
The intuition part is mostly incorporating data that NB classifier has no knowledge of. The current form, new manager, new player(s), televised game, team fighting to stay in the league, team fighting to get promoted,... you know - the usual stuff.
Nothing to brag about, but still profitable: