Early-season college basketball lines are not wrong in one predictable direction. They are softer, because in November everyone is pricing teams on very little data, the bookmaker included. The market leans hard on preseason ratings and then overreacts to a few results against weak opponents. Your edge comes from doing those two things better: weighting last year's picture and this year's games in a sensible way, and adjusting every early result for who it came against. The catch is that your own number is also rougher in November, so you should demand a bigger gap before you bet, not a smaller one.
Why are college basketball lines softer early in the season?
There are more than 360 Division I teams, and in November most of them have played two or three games. Rosters also change a lot from one season to the next because of transfers, so last season's numbers describe a team that partly no longer exists.
A bookmaker still has to post a line on every game. For the big programs the market corrects that line quickly. For a Tuesday night game between two small-conference teams, far fewer sharp bettors are looking, and sportsbooks take smaller bets on college basketball than on the NFL or the NBA. A smaller market moves slower and stays wrong longer.
That does not mean the line is wrong in a direction you can guess. It means the error is bigger, both ways. Bigger errors are only useful to you if your number is better than the line, and that is the whole job.
How much should preseason ratings count in November?
Preseason ratings should carry most of the weight for the first few weeks, and then fade as real games come in. Three games is not enough to throw out a full off-season of information about coaching, returning players and transfers.
A simple way to do this in a spreadsheet is to give the current season a weight of games played divided by games played plus 10. The 10 is a starting value you can test, not a law.
| Games played | Weight on this season | Weight on preseason |
|---|---|---|
| 0 | 0% | 100% |
| 3 | 23% | 77% |
| 5 | 33% | 67% |
| 10 | 50% | 50% |
| 20 | 67% | 33% |
Here is how it works on one team. Say the preseason rating has a team at +8 points per 100 possessions better than an average team. After three games, its opponent-adjusted results say +12. The blend is 0.77 × 8 + 0.23 × 12 = 8.9, so you price the team at about +9, not +12.
The market's typical mistake runs the other way. One or two loud results move a line further than three games deserve.
Why do early blowout wins fool bettors?
Early blowouts fool bettors because November schedules are full of mismatches. A big program will often open against teams from the bottom of Division I, and winning those games by 25 points is what a good team is supposed to do.
The raw margin looks impressive. The opponent-adjusted margin may say almost nothing new. If a team was expected to win by 22 and won by 25, it played 3 points better than expected, not 25.
So every early result in your sheet needs one more column: what the game should have looked like. Without it, you are adding noise and calling it form. My sample size lesson shows how few games it takes for luck to look like a trend.
How do you turn ratings into a college basketball spread?
You turn ratings into a spread by taking the gap between the two teams per 100 possessions, scaling it to the number of possessions you expect in the game, and adding home court if there is one.
Using the team above:
- Team A: +9 per 100 possessions
- Team B: +2 per 100 possessions
- Expected possessions: 68
- Home court for Team A: about 3 points, a common starting value
Your spread: (9 - 2) × 68 / 100 + 3 = 7.8. So your line is Team A -7.8.
If the market has Team A -5.5, you have a 2.3 point gap. If the game is at a neutral site, as many early-season tournament games are, the home court number is 0 and your line is -4.8 instead, which flips the gap to the other side. Check the venue on every November game.
The same logic, with more detail on pace and efficiency, is in my article on how to build an NBA betting model.
How big does the gap need to be before you bet?
The gap needs to be bigger in November than later in the season, because your own number carries more error early on.
To turn a spread gap into a win probability, a common starting value is a standard deviation of about 11 points for a college basketball result against the spread. With a 2.3 point gap, that gives Team A -5.5 a cover probability of about 58.1%. In odds, that is a fair price of -139 American, or 1.72 decimal. At a standard -110 (1.91 decimal), the break-even rate is 52.38%, and the expected value is about +11% of your stake.
That number is too good to trust in November. If you assume only half of your gap is real, the cover probability drops to 54.1%, a fair price of -118 (1.85). That is still above the 52.38% break-even at -110, with an expected value of about +3.3%. That is a real but much smaller edge, and it is closer to the truth this early.
When I built my NBA model, I used to look for gaps of 5 to 7 points against the line. As NBA lines became more efficient, I came to treat 3 points or more as a good signal. Early college basketball is the opposite situation. Both the line and your number are rough, so the bar should go up, not down.
If you want to check the conversion yourself, the fair odds calculator and the expected value lesson let you change the numbers and see the result.
Where is the early-season market most likely to be off?
The market is most likely to be off where the least attention goes. In practice, that means four places:
- Small-conference games. Fewer bettors, smaller limits, slower corrections.
- Teams with heavy roster turnover. The preseason rating is a guess built on players who have not played together.
- Neutral-site tournament games. A line that quietly assumes home court when there is none.
- Teams coming off one loud result. A blowout of a weak team or a close loss to a strong one, both of which move a line more than they should.
None of these is a bet on its own. They are places to look first, so your model does the deciding instead of the headline.
How do you know if your early-season bets are any good?
You know by measuring closing line value, not your win rate. After 20 November bets, your record says almost nothing. Whether you beat the closing line on most of them says a lot more.
When I measured my NFL model's bets against Pinnacle's closing line, the closing line value came out at +3.81% one season and +4.25% on the 2025/26 snapshot. That is the kind of number that tells you a process works long before the results are clear. The closing line value lesson shows how to calculate it bet by bet.
If you would rather not build the comparison yourself every night, the AI Betting Assistant applies the same logic for you, faster.
FAQ
Is college basketball easier to bet than the NBA?
College basketball lines are generally less efficient than NBA lines, mainly because there are far more teams and much less attention on most games. That does not make it easy. Most games still have a line close to fair, and the softer ones are softer because the information is thin for everyone, including you. The advantage goes to the bettor with a consistent method: a rating for every team, an adjustment for opponent strength, and a rule for when a gap is big enough. Without that, more teams just means more ways to bet on noise.
When is the best time to bet early-season college basketball lines?
The best time is usually soon after the line opens, when it is least tested. Openers on low-profile games have had the fewest sharp bets against them. The trade-off is that limits are also smallest at that point, so you cannot bet much. As game time gets closer, the line absorbs more information and becomes harder to beat. If your number disagrees with an opener by a meaningful amount, that is the moment to act, and then check later whether the line moved toward you. If it keeps moving away, your rating is probably the thing that is off.
How many games before you can trust this season's stats?
There is no exact cutoff, but this season's numbers usually need around 10 games before they deserve as much weight as the preseason rating. With the simple weighting in this article, a team that has played 10 games gets a 50/50 blend. Before that, preseason ratings should do most of the work. After about 20 games, this season's opponent-adjusted results carry two thirds of the weight. The exact constant matters less than using one consistently and testing it on your own results.
What is home court advantage worth in college basketball?
About 3 points is a common starting value for home court advantage in college basketball, though it varies by venue and has drifted over the years. Some arenas are worth more, and many early-season games are on neutral courts, where it is worth zero. The biggest mistake is not picking 2.5 instead of 3. It is applying home court to a neutral-site tournament game. Check the venue on every November and December game before you trust your spread.
Should I bet college basketball totals or spreads early in the season?
Spreads are the simpler place to start, because the rating gap between two teams is the most direct output of your model. Totals depend heavily on pace, and early-season pace numbers are built on very few games against uneven opponents. If you do bet totals, use the same blend of preseason and current-season data for pace, not just the raw numbers from three games. In both cases, the rule is the same: price it first, compare it to the line, and only bet when the gap is clearly bigger than your uncertainty.
Where can I get college basketball ratings for my model?
You can start with public team ratings such as Bart Torvik's T-Rank, which is free, or KenPom, which is a paid subscription. Both give adjusted offensive and defensive efficiency and tempo, which is enough to build the spread formula in this article. Use them as your preseason base and as a check on your own numbers. If your rating and theirs disagree by a lot on a team, find out why before you bet, because often it is a data error on your side.
Want a model that does this math for you?
The Ultimate Modern Bettor's Toolkit gives you the spreadsheet to turn your own number into a fair price and compare it with the line.
$27, one payment. It includes the +EV betting model spreadsheet, a 45-minute video training that builds your first model with you, and the four-bucket staking system. 30-day money-back guarantee.
Get the ToolkitIf you want a ready-built model for a specific sport, you can see all of them on the betting models page.
MB built Underdog Chance around one idea: make your own number first, then bet only when the price beats it. Over the last 20+ years his simple, probability-based Google Sheets models have helped thousands of bettors stop following picks and start pricing bets for themselves.
Last updated: November 11, 2026