Building a College Basketball Betting Model
Honestly I never bet on college basketball. It is a league with a lot of teams, with a lot of players and probably some “betting rules” that I am not familiar with. After all, we Europeans follow mostly major league sports and college is still little bit far away.
Basketball computer picks
But I believe that every sport can be and must be described with numbers before you bet. It is a market of numbers and if you don’t have the numbers, then you are just guessing. And if you are guessing, then it is only a matter of time, when you will see that you are wasting time.
Being selective in sports betting
I am also not a fan of jumping from sport to sport and I always recommend that if you bet, you should stick with one sport, one league and even one bet type is enough if you are good. Especially if you don’t do this as a full time job like me. I bet on my sports (MLB = main sport), but at the same time I also research betting market, learn, improve all the time and challenge myself in other sports and then I share this knowledge with our members.
My goal and my full-time job is to help sports bettors to understand market, to use numbers before they bet and become independent handicappers. So they don’t send me messages like: “Hey, what is your play today?” .
Stop following college basketball picks and parlays
I hate those questions. Because I know that those people don’t want to become better, they are just looking for “winners” and there are no winners in the sports betting world. Only different probabilities of outcomes. And the whole thing is a relative competition against others.
And this is how I decided that I will spend part of my time for CBB to help our members who already bet CBB. I researched the market, I have created simple betting model, that anyone can use and I think it is very useful especially for those who already bet on CBB.
NCAA basketball picks against the spread & totals
I wanted to describe games with simple basketball ideas that were already implemented fo NBA and European basketball. With this I wanted to project the games, where I will get my:
And then I will compare my numbers with bookmakers’ numbers.
Here is the example (on my site all projections are free every day and I show 10pts diff projections – read below for more info):
Simple College Basketball Betting Model and Lines
Because of simplicity of the model, where you basically need 10 seconds to prepare after it is already created, I started to track bets based on:
- 5pts difference (or more) between my projected spread and bookmakers spread (Exaple: my projected BYU -15 Bookmakers BYU -10; I would take BYU -10 because it is 5 pts difference or more)
- 6pts difference (or more) between my projected spread and bookmakers spread (Exaple: my projected BYU -15 Bookmakers BYU -9; I would take BYU -9 because it is 6 pts difference or more)
- 7pts difference (or more) between my projected spread and bookmakers spread (Exaple: my projected BYU -15 Bookmakers BYU -8; I would take BYU -8 because it is 7 pts difference or more)
- 8pts difference (or more) between my projected spread and bookmakers spread (Exaple: my projected BYU -15 Bookmakers BYU -7; I would take BYU -7 because it is 8 pts difference or more)
- 9pts difference (or more) between my projected spread and bookmakers spread (Exaple: my projected BYU -15 Bookmakers BYU -6; I would take BYU -6 because it is 9 pts difference or more)
- 10pts difference (or more) between my projected spread and bookmakers spread (Exaple: my projected BYU -15 Bookmakers BYU -5; I would take BYU -5 because it is 10 pts difference or more)
So far I have projected 631 games. I will make projections up to 1000 analysed games and then I will stop doing this. It takes a lot of time, because there is a lot of teams and I track a record with different pts spreads and also closing spreads. With other sports, that I already successfully bet I will not bet CBB in the future, because I simply don’t need it. But I think it is great tool for CBB bettors who love this sport. With this model they will get projected numbers before they bet on a very simple way and with this they get the picture where they have value and where they don’t have a value.
COMPUTER BASKETBALL PICKS RESULTS AFTER 631 ANALYSED GAMES
5 PTS DIFF
- Record: 260-229 +10.20 Units
- Spreads:139-118, +9.88 Units (BTL: 53.15%)
- Totals: 121-111, +0.32 Units (BTL: 45.13%)
6 PTS DIFF
- Record: 203-189 -2.24 Units
- Spreads:113-97, +6.96 Units (BTL: 53.07%)
- Totals: 90-92, -9.22 Units (BTL: 48.03%)
7 PTS DIFF
- Record: 155-140 +2.60 Units
- Spreads:85-74 +4.20 Units (BTL: 52.55%)
- Totals: 70-66, -2.60 Units (BTL: 61.22%)
8 PTS DIFF
- Record: 115-99 +6.80 Units
- Spreads:65-53, +6.80 Units (BTL: 50.00%)
- Totals: 50-46, +0.00 Units (BTL: 52.50%)
9 PTS DIFF
- Record: 89-67 +14.88 Units
- Spreads:49-38, +7.08 Units (BTL: 45.33%)
- Totals: 40-29, +7.80 Units (BTL: 51.79%)
10 PTS DIFF
- Record: 57-45 +7.44 Units
- Spreads:34-26, +5.28 Units (BTL: 49.02%)
- Totals: 23-19, +2.16 Units (BTL: 51.35%)
Record is based on -108 odds. You can probably get better or worse odds, but in general betting on the odds of more than -110 vs -110 is not recommended, because you probably don’t have such an edge against market. 5Dimes has -105 vs -105 or -105 vs -107 for most games, so the record could/should be even better. But I took -108 which makes my record worse.
BTL = beating the line and the tracking of BTL here is flawed, because I just check closing spreads (picks against the spread) and totals without adding the odds. I also use pinnacle closing odds and take those lines from early other bookmakers that are on the market, so it is definitely not perfect tracking. I should track actual odds, not just spreads. -7 can be at -115 or at +102. Secondly, it is also not the same if you bet the line 60% of times with CLV +6% and 60% of times with CLV +0.01%. But if you have time and if you bet only CBB it would be great if you compare your taken spreads with odds and then compare with actual pinnacle closing odds. Some lines change quickly and it looks like that there are games where lines can be easily moved. Still BTL number is not good so far. Those numbers should be better.
What have I learned after 631 analyzed ncaa basketball betting lines?
Bet only CBB if you decide to bet CBB
There is a lot of games every day. With 353 teams we also have a lot of games every day. I believe that if you go really wide with this league I think this is the only league you should bet at a time. Lines also move very quickly and the odds can open later or not at the same time for all games.
In first games I struggles little bit with team names. Now I am little bit more familiar. I used google to check every single team. For example somewhere we have Virginia Military, somewhere we have VMI and there is couple of such examples. Of course this is new for me and I believe that if you bet CBB for long time you don’t struggle with this. Definitely, I would recommend new bettors to take first year to get familiar with the league and bet tiny amount just to feel the bookmakers, numbers and how much time they need to do this on a daily basis.
Home Field Advantage
A lot of games are played on neutral field and because if this I need to check every single game. Some games are played in Jamaica, Bahamas and then I need to adjust this without home field advantage where I use 3 pts advantage for all teams. This takes extra time.
Small vs Big/important games
Our members who bet on CBB told me, that some games are not that important and the limits are much lower. But still I projected the lines for every single game. My goal was to create a simple model, where you can project your own spreads and totals and then make better decisions. I simply tracked my projections versus bookmakers spreads.
The best results
The best results so far are on games, where I look for 9 pts difference between my projections and bookmakers numbers. Both for totals and spreads. With bigger difference I am more selective and we exclude games that are “too close” with bookmakers numbers. It also makes sense that when we are more selective we also have better results.
- 9pts diff ROI (Yield) = 9.5%
- 10pts diff ROI (Yield) = 7.3%
I would definitely stick with 9-10 pts difference.
Around 10% of all analyzed games are qualified bets based on my college basketball betting model
I also track percentage of qualified bets based on my model. We have analysed 631 games so far and if we look at the best results (9pts and 10pts) we have:
- 9pts: 156 qualified bets (out of 631 analysed and every game has 2 options – spread and totals this is 1262 “opportunities”), which is 12.36%. In other words cca 12 out of 100 analysed games is qualified for betting.
- 10pts: 102 qualified bets (out of 631 analysed and every game has 2 options – spread and totals this is 1262 “opportunities”), which is 8.08%. In other words 8 out of 100 analysed games is qualified for betting.
CBB is definitely one of the main sports for betting in USA and it can be very attractive for bettors from USA, especially if you have access to low juice odds.
College basketball picks and predictions to help members
With betting model I wanted to help our members and I will keep doing all projections until I reach 1000 analysed games.
For me it takes a lot of time to complete these projections because I am not familiar with team names and many times I have to use google to check if I used the right names.
For example, we have 7 or 8 different Florida teams. Florida State, North Florida, South Florida,… and all kind of other Floridas. I also need to check every single game if they are playing in their home field or if they are playing on Bahamas. I am not sure why students travel so much, because I think they should study, but it looks like this is now huge business as well. And this makes for me job even harder. I need to check every arena and google it if this is their home field, so I can use home field advantage or not. BTL is not really good, but also use flawed tracking system for this, because I simply don’t have time to go with every closing odds and percentage. It already takes a lot of time for me to complete those projections checking team names and home fields.
But definitely, it is very useful for those who already bet on CBB.