Basketball Predictions and Picks

What a basketball prediction from us contains

Four things, in a fixed order. The market’s line, with the book and the time it was read. Our number for the same game. The gap between them. The pick, with the price at which it stops being a bet.

The threshold comes before the pick. If the gap falls short of the threshold, the first line of the page reads no bet. Nothing on the page asks to be taken on trust: the inputs are printed, so the arithmetic can be redone by anyone with a calculator.

When basketball is on

There is a basketball league in season in every month of the year. The NBA tips off in late October, plays its regular season to mid-April, then runs the play-in, the playoffs and the Finals into June. College basketball starts in early November, settles its conferences in early March, and plays the NCAA tournament from mid-March to the national championship in early April. The WNBA covers the summer, from May into October.

The volume is uneven. The 2025-26 NBA season was 1,322 games, playoffs included. College basketball played 6,300 games involving Division I teams, almost five times as many.

Basketball odds, in three prices

A basketball game is priced on three markets: spread, total and moneyline. The spread is a handicap in points. The total is a number for both teams’ points combined, bet over or under. The moneyline asks only who wins.

The numbers are larger than in other sports because basketball scores more. In 2025-26 the average NBA total was 229.7 points and games produced 230.2. In college the average total was 148.8 and games produced 150.1. In both leagues the average total and the average score finished less than 1.5 points apart.

Every NBA spread in ESPN’s record of the 2025-26 season ended in a half point: 1,321 regular-season and playoff games with a line, none a whole number. So no NBA spread bet pushed. A whole-number line can push, and when one appears it is worth noticing.

Picks against the spread

Basketball has no key numbers. In football a field goal makes 3 the commonest margin by a distance. In basketball scoring comes in ones, twos and threes, dozens of times a game, and the final margins spread out. In the 2025-26 NBA no margin reached 6% of games; the commonest was 4 points, in 73 of 1,322 (5.5%).

The practical result: each point on a basketball spread is worth about the same as the next. A half point around 4 is worth about what a half point around 9 is worth, and no half point sits beside a margin as common as football’s 3.

Spreads run large in both leagues. The average NBA favorite gave 7.1 points in 2025-26, the average college favorite 8.5. Lines of 10 or more: 335 of 1,321 NBA games, 1,677 of 5,706 college games.

Favorites and underdogs split close to even. NBA favorites covered 651 times and underdogs 670. In college, favorites covered 2,760 times and underdogs 2,946. By size of spread, favorites covered between 44% and 51% of the time in every group but one: NBA favorites of 15 or more, who covered 56 of 95. That is too small a sample to mean anything. Against a market that splits this evenly, a side without its own figure behind it is a guess.

Home court, by league

Home teams win more often in college than in the NBA. In 2025-26 NBA home teams won 730 of 1,316 games, 55.5%. College home teams won 3,790 of 5,592, 67.8%. Neutral-site games are left out of both.

Much of the college figure is the schedule. In conference games, college home teams won 58.5% of 3,389. Outside conference play they won 82.0% of 2,203, and the lines show why: the home side was favored by 10 or more in 854 of 1,636 non-conference games with a line, against 537 of 3,387 in conference play. Another 567 non-conference home games had no line at all, and the home side won 557 of them. A college home edge read off a season total overstates what the building is worth in February.

The back-to-back

The NBA schedules teams on consecutive nights. In 2025-26 there were 301 games where one team had played the day before and its opponent had not. The tired team won 143 of them, 47.5%. Against the spread it went 153-147, 51.0%. It was the underdog in 180 of 300.

The tired team lost more often than it won, and the prices allowed for that: it was the underdog in 60% of those games, and against the spread it finished 153-147. One season of 300 games shows no edge in either direction. A back-to-back is a reason to check the price, not a bet by itself.

Betting the total

A total is pace times efficiency: how many possessions both teams will have, and how many points each possession is worth. Both shift with who plays, so late lineup news can move a total as much as a spread.

Over a full season the two sides come out close. In 2025-26 NBA games went over 655 times and under 666. College games went under 2,913 times and over 2,821.

The moneyline

The moneyline pays on the winner alone, so it tells you what the market thinks of the result rather than the margin. How often the favorite won outright in 2025-26: NBA favorites of 6 to 9.5 points won 75.3% of 356 games; favorites of 15 or more won 96.8% in the NBA and 96.9% in college. Favorites of under 3 points won 48.4% of 273 NBA games, which over a sample that size is a coin flip.

A favorite’s moneyline price already carries those rates. Laying a short price on a big favorite is not a safe bet; it is a bet that needs the favorite to win more often than the price assumes.

Basketball futures and the tournament

A futures bet settles at the end of a season: a title, an award, a bracket. The margin built into futures prices is wider than on a single game, and the money is tied up for months. Our guide to futures bets shows how the prices are built.

Our basketball futures pages:

• NBA: championship odds, MVP, Defensive Player of the Year, Rookie of the Year and the NBA Cup
• College: the national championship, the men’s Final Four and the women’s Final Four
• WNBA: championship odds and MVP

In March the college game becomes a bracket. Our March Madness betting guide covers the tournament’s bets and dates.

Basketball betting tips, with the arithmetic

• Know the break-even. At -110 the break-even rate is 52.4%. Every tip below is a way of getting above that line, not of picking more winners.
• Price the news, or wait for it. Who plays is often confirmed close to tip-off. A bet placed before the news carries that risk; a bet placed after it gets a line that already includes it.
• Use home court by league. 55.5% in the NBA, 67.8% in college, 58.5% inside a college conference. One number for all three is wrong for all three.
• Distrust short records. A team that is 8-2 against the spread in ten games has a record, not a trend. Ask for the sample before the rate.
• Pass more than you play. With favorites and underdogs splitting close to even, most games offer no gap worth betting. A no bet is a result of the method, not a failure of it.

For the mechanics of each market, our guides cover how to bet on NBA games, the NBA betting season and college basketball.

Basketball betting terms

• Pace: possessions per game. More possessions, more points, a higher total.
• Back-to-back: a team playing on consecutive nights. Common in the NBA, priced by the market.
• The number: the spread or total a bet was placed at. Two bets on the same side at different numbers are different bets.
• First-half line: a spread or total for the first half only, settled at halftime.
• Pick’em: a game with no favorite; the spread is zero.
• Line move: a change in the spread or total between the open and tip-off.
• Unit: a fixed stake, the same on every bet, so results can be compared. Our unit guide explains it.

Pick your league

Each league has a page listing its games and our calls on them. NBA predictions follow the pro season from October to the Finals. College basketball picks follow the Division I season into March. WNBA predictions cover the summer league through its playoffs.

Questions and answers