The NBA season tips off Tuesday, Oct. 20, at 3 p.m. ET in Detroit, where the Pistons, the East’s top seed a year ago, host the Celtics. That night the 76ers visit the Knicks as New York raises its first championship banner since 1973, and the Thunder open in San Antonio in a rematch of last spring’s seven-game Western Conference finals.
In between came a summer of big moves. Giannis Antetokounmpo is in Miami. Jaylen Brown and LeBron James are in Philadelphia. LaMelo Ball is in Minnesota and Ja Morant is in Portland. Here is what changed, which NBA markets and props Fourth & Value tracks, and what the first prices of the season say about the books.
- We compare moneylines, spreads, totals and 11 player-prop markets across up to 10 sportsbooks, with each book’s margin removed.
- In Sunday’s snapshot the average margin (overround) on player props was 6.87%, against 4.57% on game lines. A typical prop needs about a 53.4% win rate to break even at its posted price; a typical game line, about 52.3%.
- The same prop can cost very different amounts. Cade Cunningham’s opening-day over 24.5 points ranged from +100 to −125 across five books.
- Our NBA forecasts are experimental and run in shadow mode. They have not been validated against betting prices, and we don’t make NBA picks yet.
Where last season ended
The Knicks beat the Spurs 4–1 in the Finals, closing it out 94–90 in San Antonio on June 13. Jalen Brunson was the unanimous Finals MVP after scoring 45 points in Game 5 and averaging 32.6 for the series. It was New York’s first title in 53 years, won in Mike Brown’s first season as coach. Shai Gilgeous-Alexander won a second straight MVP award, ahead of the other finalists, Nikola Jokić and Victor Wembanyama.
The regular season was top-heavy. Oklahoma City went 64–18, San Antonio 62–20 and Detroit 60–22, the best record in the East. Boston (56–26) finished second in the East but lost a seven-game first-round series to Philadelphia, which the Knicks then swept. The Spurs beat the Thunder in seven to reach the Finals.
The offseason: the East was rearranged
Giannis Antetokounmpo to Miami
On June 22, the night before the draft, Milwaukee agreed to send Antetokounmpo and Bobby Portis to the Heat for Tyler Herro, Jaime Jaquez Jr., Kel’el Ware, Kasparas Jakucionis, the No. 13 pick, unprotected first-round picks in 2031 and 2033, a 2030 pick swap and a 2033 second-rounder, according to ESPN. Antetokounmpo, a two-time MVP, played 36 regular-season games last season for a Bucks team that finished 32–50. Milwaukee also has a new coach, Taylor Jenkins, formerly of Memphis.
Philadelphia adds Jaylen Brown and LeBron James
On July 1 Boston traded Brown to the 76ers for Paul George and draft picks, ending a Celtics career that included five All-Star selections and, last season, a career-high 28.7 points a game with 6.9 rebounds and 5.1 assists. On July 24 James agreed to a two-year, $8 million contract with a player option, leaving the Lakers after eight seasons. They join Tyrese Maxey and Joel Embiid, who played 38 games last season. Philadelphia opens on banner night at Madison Square Garden.
Boston retools around Jayson Tatum
Tatum returned on March 6 from the Achilles rupture suffered in the 2025 playoffs and played 16 regular-season games. Boston received George in the Brown trade and signed Mitchell Robinson, who had spent his whole career with the Knicks, to a three-year, $47.4 million deal. Tatum, George and Robinson all had props posted for opening day.
LaMelo Ball to Minnesota, Ja Morant to Portland
Charlotte traded Ball and Josh Green to the Timberwolves on June 25 for Naz Reid, a 2033 unprotected first-round pick, three first-round swaps and three second-rounders, then agreed to a three-year, $74 million deal with Coby White. Four days later Memphis sent Morant, who played 20 games last season, to Portland for Jerami Grant and Kris Murray, part of a reset that also brought Cameron Boozer with the No. 3 pick.
Wembanyama extends, the draft and six new coaches
Wembanyama agreed on July 10 to a five-year, $252 million rookie-maximum extension with San Antonio, choosing the 25% maximum over a larger supermax. At the draft, the Wizards, last season’s worst team at 17–65, took BYU’s AJ Dybantsa first. Utah took Kansas guard Darryn Peterson, Memphis took Duke’s Boozer and Chicago took North Carolina’s Caleb Wilson.
Six teams start the season with new coaches, according to the Associated Press. Dusty May (Mavericks), Sean Sweeney (Magic) and Micah Nori (Trail Blazers) are first-time NBA head coaches; Jenkins (Bucks), Jamahl Mosley (Pelicans) and Tiago Splitter (Bulls) have been head coaches before.
Why the summer matters for prices
Every one of these moves weakens the link between last season’s box scores and this season’s games. A model that rates teams from last season’s results, ours included, starts the year rating rosters that no longer exist, while the books have had months to absorb the news. In October, expect the market to know more than any box-score model. It is one reason our early-season forecasts are labeled experimental and kept out of picks.
What the futures market says
October odds roundups from Fox Sports and Sportsbook Review list the Spurs and Thunder as co-favorites for the title, with the Knicks and 76ers next. We don’t track futures. A title market spreads the book’s margin across 30 teams and won’t settle until June. Our method removes margin by pairing the two sides of the same bet at the same book; a 30-team market doesn’t fit that method, so we don’t publish a fair price for it.
What we cover
| Market | Included | Collected |
|---|---|---|
| Game lines | Full-game moneyline, point spread and total | Games in the next 45 days |
| Player props | Points, rebounds, assists, three-pointers made, points + rebounds + assists, points + rebounds, points + assists, rebounds + assists, blocks, steals, turnovers | Up to 16 games within 48 hours of tipoff |
Not covered: futures, halves and quarters, same-game parlays and in-game odds.
Prices refresh twice a day, scheduled for about 11 a.m. and 5 p.m. ET (10 a.m. and 4 p.m. once standard time begins). Sunday’s snapshot carried quotes from DraftKings, FanDuel, BetMGM, Caesars, BetRivers, Fanatics, Bovada, BetOnline.ag, LowVig.ag and MyBookie.ag. Not every book operates in every state, and books post props on their own schedules: as of Sunday evening, Fanatics had a handful of opening-day props, while BetMGM, LowVig.ag and MyBookie.ag had none.
- Player props line up every book’s price at the same player, market and line, mark the best price, and show the no-vig probability next to the median of the other books. Selecting a player opens their last ten games against the line.
- Game lines do the same for moneylines, spreads and totals.
- Market Watch flags a price that differs from the median fair probability of at least three other books at the same line. It is a research lead, not a pick.
- Market Analytics grades every tracked prop and game line against what the books’ own no-vig prices implied, starting with opening night. It uses our last pregame snapshot, a few hours before tipoff, not the closing line.
- Bet Tracker settles NBA bets from ESPN box scores. Overtime counts, and a prop on a player who doesn’t play is a push.
Each price on the NBA board also shows an experimental forecast from our NBA v1 model. It runs in shadow mode: it has not been validated against betting prices, we don’t display an expected value, and NBA prices are not eligible for Today’s Picks. Players listed out on ESPN’s injury report get no forecast; other statuses, lineup changes and teammates’ absences are not adjusted for. NBA prices become eligible only after the model is graded against this season’s prices and shows it adds something the market lacks. The current daily process lists which sports are eligible.
Market analysis: what the books charge
The margin built into a two-sided price is easiest to see as the overround: convert both prices to implied probabilities, add them and subtract 100%. At −110 on both sides the overround is 4.76%, and you need to win 52.4% of the time to break even. (The book’s hold, the share of money it keeps when action is balanced, is slightly lower, 4.55%.) We measured the overround on every pair of quotes for both sides of the same book, game, market and line in Sunday’s 6:41 PM EDT snapshot.
| Market | Pairs | Average overround | Break-even |
|---|---|---|---|
| Moneyline | 179 | 4.37% | 52.2% |
| Point spread | 184 | 4.64% | 52.3% |
| Game total | 185 | 4.69% | 52.3% |
| Points | 187 | 6.95% | 53.5% |
| Rebounds | 98 | 6.86% | 53.4% |
| Assists | 55 | 6.87% | 53.4% |
| Three-pointers made | 51 | 6.91% | 53.5% |
| Points + rebounds + assists | 68 | 6.82% | 53.4% |
| Points + rebounds | 65 | 6.74% | 53.4% |
| Points + assists | 39 | 6.76% | 53.4% |
| Rebounds + assists | 39 | 6.77% | 53.4% |
Break-even is the win rate needed at the average overround, assuming the margin is split evenly between the two sides. Steals and blocks had too few pairs to report. Game lines cover the 46 games listed through Dec. 25; props cover the one game inside the 48-hour window on Sunday, Celtics at Pistons, with 14 players.
Props cost about 2.3 percentage points more than game lines, roughly the difference between −110 and −115 on both sides. That matches the usual pattern: books charge more on player props than on main spreads and totals, so the same opinion costs more to bet there.
Books differ, too. On opening-day props the average overround ran from 6.40% at BetOnline.ag and 6.66% at Bovada to 6.70% at DraftKings, 6.94% at FanDuel, 7.15% at Caesars and 7.67% at BetRivers. On game lines most books sat between 4.4% and 4.9%; LowVig.ag was lowest at 3.19% and BetRivers highest at 5.46%. One game is a small sample for props, so treat the book-by-book figures as a first reading, not a ranking. Our vig explainer shows how to compute the margin and remove it.
Market analysis: opening-week lines
Game lines for opening week have been posted since at least Oct. 7, so we can see how they have moved. The table compares the median line across books in our first snapshot (Wed, Oct 7, 4:20 PM EDT) with Sunday’s (Sun, Oct 18, 6:41 PM EDT). Spreads are the home team’s.
| Game | Spread Oct 7 | Spread Oct 18 | Total Oct 7 | Total Oct 18 |
|---|---|---|---|---|
| Celtics at Pistons, Oct 20 | −2 | PK | 221 | 222 |
| 76ers at Knicks, Oct 20 | −5 | −4.5 | 231.5 | 232 |
| Thunder at Spurs, Oct 20 | −2.5 | −2.5 | 230.5 | 231.5 |
| Hornets at Nets, Oct 21 | +2.5 | +1.5 | 223.5 | 221.25 |
| Jazz at Grizzlies, Oct 21 | +2.5 | +1.25 | 241.5 | 242 |
| Mavericks at Rockets, Oct 21 | −8.25 | −8 | 227.5 | 230 |
| Cavaliers at 76ers, Oct 22 | −4.5 | −2.75 | 233.5 | 235 |
| Nuggets at Thunder, Oct 22 | −7.5 | −7.5 | 232.5 | 234.5 |
Median of the books quoting each game: 6–9 books on Oct. 7 and 8–10 on Oct. 18. A median of .25 or .75 means books split between two numbers. Highlighted cells moved 1.5 points or more.
The biggest spread moves were toward the visitors: Detroit went from a two-point favorite over Boston to a pick’em, and Philadelphia’s home line against Cleveland shrank from 4.5 to 2.75. Moves like these can come from injury news, from bets or from books following each other. A median tells you the market changed, not why, and we don’t attribute these moves to any one report.
On totals, the 17 games from Oct. 20 to 23 carry a median consensus total of 234 points, from 221.25 for Hornets–Nets to 243 for Timberwolves–Heat. Last season’s regular-season games averaged 231.2 combined points in ESPN box scores, up from 227.7 in 2024–25 and 228.4 in 2023–24, and the first three weeks of 2025–26 averaged 233.6. Don’t read that as an early-season rule. In each of the two previous seasons, the first three weeks scored below the full-season average (227.1 and 226.3). Opening totals already sit above last season’s average, so a bet on early-season scoring needs a reason beyond the trend.
Two more numbers from last season worth keeping in mind: 4.4% of regular-season games went to overtime, which counts for totals and player props (see our NBA betting rules), and home teams won 55.4% of games.
Market analysis: same prop, different prices
The cheapest lesson in prop betting is that books disagree. These are the opening-day prices for Celtics at Pistons in Sunday’s snapshot (quotes from 6:40–6:41 PM EDT). They will have changed by tipoff.
| Prop | Books at this line | Over price range | Break-even range |
|---|---|---|---|
| Cade Cunningham over 24.5 points | 5 | +100 to −125 | 50.0% to 55.6% |
| Jayson Tatum over 26.5 points | 7 | −109 to −121 | 52.2% to 54.8% |
| Paul George over 15.5 points | 6 | −118 to −127 | 54.1% to 55.9% |
| Jayson Tatum over 8.5 rebounds | 5 | −104 to −131 | 51.0% to 56.7% |
Cunningham: BetRivers +100, Caesars −115, DraftKings −118, BetOnline.ag −119, Bovada −125. Tatum points: BetRivers −109 to Caesars −121. George: FanDuel −118 to BetRivers −127. Tatum rebounds: BetRivers −104 to DraftKings −131.
Three things stand out. First, the price spread is wide even on a star’s points line: backing Cunningham’s over at Bovada instead of BetRivers means needing to be right 5.6 points more often. Second, books also disagree on the line itself. FanDuel listed Cunningham at 23.5 rather than 24.5, and FanDuel, BetOnline.ag and Bovada had Tatum’s rebounds at 9.5 (over +118 to +125) while five other books used 8.5. A half point on a points prop or a full rebound on a rebounds prop is worth more than most price differences, so compare the same line before comparing prices.
Third, the best price on one side often comes with a worse price on the other. BetRivers’ +100 on the Cunningham over came with −136 on the under, so its no-vig probability for the over was 46.5%, against 49.8% to 52.0% at the other four books. A book that shades one side this way may simply be moving its line, and the price can be gone by the time you look. This is the kind of disagreement Market Watch is built to surface: something to research, not a forecast that the over will hit.
What our NBA research says
Our technical report, NBA Forecasting and Market Pricing, tested the experimental NBA v1 model on the 2025–26 season. The findings shape how we run the NBA board this year:
- The market was better. Against 979 DraftKings lines that ESPN carried in its game summaries, DraftKings’ no-vig prices beat our experimental model on moneylines, spreads and totals, and simple rules that bet the model’s disagreements lost money. The most likely reason is that the market knows who is playing tonight.
- Rest is already in the price. Rest days and back-to-backs added nothing measurable to forecasts once team pace and efficiency were known. Our low-scoring games study, covering NBA closing totals since 2007–08, found schedule situations like these largely priced in. The one NBA pattern that replicated there was overs in the final two weeks of the regular season (54.1% from 2016 to 2025, but 50.0% before), something to revisit in April, not October.
- Minutes come first. Projecting minutes and then production per minute beat per-game averages in every player statistic. Our minutes and rest explainer covers why.
- Old splits don’t help. A player’s record against tonight’s opponent, at home or away, or on the second night of a back-to-back did not improve forecasts.
None of this says NBA prices can’t be beaten. It says that a model without injury and lineup information shouldn’t expect to beat them, and that the clearer, measurable costs are the ones in the tables above: margin, line differences and price differences.
How to use the board this season
- Match the line first. Compare prices only at the same player, market and line.
- Know what you’re paying. On a prop, assume you need about 53.4% at a typical price; more at a bad one. Our odds calculator shows the break-even for any price.
- Check availability late. The NBA injury report is due by 5 p.m. local time the day before a game, and statuses often change on game day. A prop on a player who doesn’t play is void; a teammate’s prop or a game total is not.
- Be wary of October numbers. Rosters changed a lot this summer. Early-season forecasts, ours included, know the least they will all year.
- Record the price you actually got. Bet Tracker settles NBA bets automatically, and closing-line value tells you whether your prices beat the market over time.
Frequently asked questions
Which NBA props does Fourth & Value track?
Points, rebounds, assists, three-pointers made, points + rebounds + assists, points + rebounds, points + assists, rebounds + assists, blocks, steals and turnovers, for up to 16 games within 48 hours of tipoff. Game lines (moneyline, spread and total) are tracked for games in the next 45 days.
How much do sportsbooks charge on NBA player props?
In our Oct. 18, 2026 snapshot the average margin (overround) on NBA player props was 6.87%, against 4.57% on moneylines, spreads and totals. At those margins a typical prop needs about a 53.4% win rate to break even, a typical game line about 52.3%.
Does Fourth & Value make NBA picks?
Not yet. Our NBA forecasts are experimental and run in shadow mode. They have not been validated against betting prices, and NBA prices are not eligible for Today’s Picks.
Sources
- Fourth & Value NBA odds snapshots, Oct. 7–18, 2026, and ESPN box scores for the 2023–24 to 2025–26 seasons (records, series results, scoring, overtime, games played and season averages).
- ESPN, Knicks’ Brunson seals Finals MVP honors with 45 points in Game 5, June 13, 2026.
- Eurohoops, Shai Gilgeous-Alexander wins 2025–26 NBA MVP award, May 17, 2026.
- ESPN, Bucks trading Giannis Antetokounmpo to Heat for haul of players, picks, June 22, 2026.
- Associated Press via WTVY, Jaylen Brown traded by Celtics to 76ers for Paul George, draft picks, July 1, 2026.
- FOX 4, LeBron James signs with the Philadelphia 76ers, July 2026.
- Associated Press via Fox Sports, Mitchell Robinson agrees to a 3-year, $47.4 million deal with the Celtics, July 1, 2026.
- NBC New York, Jayson Tatum’s return from Achilles injury, March 6, 2026.
- Associated Press via Fox Sports, Hornets trade LaMelo Ball to Timberwolves for Naz Reid, draft picks, June 25, 2026.
- WDAM, Grizzlies trade Ja Morant to Portland, June 29, 2026.
- Associated Press via Las Vegas Sun, Victor Wembanyama, San Antonio Spurs agree on 5-year extension, July 10, 2026.
- NBC Sports, Top four picks go to chalk in NBA draft, June 23, 2026.
- Associated Press via ABC News, NBA season to begin with 6 coaches in new places, including 3 first-time sideline bosses, Oct. 15, 2026.
- NBC New York, NBA opening night schedule, October 2026.
- Fox Sports, 2026–27 NBA Finals odds, and Sportsbook Review, NBA championship odds, October 2026.
- Fourth & Value Research, NBA Forecasting and Market Pricing (FV-2026-05), Oct. 18, 2026.
Related reading
- NBA Player Props: How Minutes, Rest and Injury News Move the Number
- NBA Betting Rules: Overtime, Player Props and the Player Who Sits
- What Is Vig? How Sportsbooks Make Money, and How to Remove It
- NBA methods
Prices are from our snapshot of Sun, Oct 18, 2026, 6:41 PM EDT, and will change. Reporting was checked on Oct. 19, 2026. Several transactions above were reported as agreements before league approval; contract figures are as reported.
← Back to Blog