Detroit finished last season 60-22. That was the best record in the Eastern Conference by four games, the best point differential in the conference at plus-669, and the No. 1 seed. Then the Pistons lost a seven-game second-round series to Cleveland, the New York Knicks won the title from the three seed, and the summer moved on.
Kalshi runs a market on who finishes first in the East next season, and it lists Detroit fifth at 12 percent. Ahead of the Pistons sit New York at 21, Philadelphia at 19, Boston at 14 and Toronto at 13. Philadelphia went 45-37 last season with a point differential of minus-15. Detroit was 14 and a half cents on August 26.
We gave all 15 teams to eight models with every price hidden. Every one of them made Detroit the favorite, at a median of 22.8 percent, nearly double what the market charges.
New to Kalshi event contracts? Start with our plain-English walkthrough of how to bet the NBA on Kalshi.
Stay ahead of the markets.
Daily insights and expert picks on Kalshi, Polymarket, and what's moving markets.
Free forever. Unsubscribe anytime.
Why All Eight Models Put Detroit First
The market's case against Detroit is history. The same franchise held the East's top seed in back-to-back seasons only four times in 21 chances, which is 19 percent. Eight of those 21 top seeds came from a team seeded fifth or worse the year before, including Boston in 2007-08, which finished last in the conference and then won 66 games. Ten different franchises have held the seed in 22 years.
The models' case for Detroit is the bar. Across the last 22 seasons, the East's No. 1 seed also held the conference's best record every single time, and those 22 teams averaged 58.9 wins and a plus-562 point differential. This is a market about which team wins something like 58 games, and Detroit is the only East team that cleared that bar last season.
The models were not ignoring the repeat problem. They were reading it differently. Claude Opus put the argument most directly: "The 19% repeat rate is being misused by others: it is spread over 14 rivals, so 19% would already make Detroit favorite, and the profile that did repeat (DET 05-06, CLE 09-10, MIL 19-20) is exactly a young best-differential team with continuity."
A 19 percent repeat chance is not a low number when the alternative is one of 14 other teams. It is more than double what a random draw would give.
The second argument was about whether the 60 wins were earned. A team can win 60 games by going 20-5 in close ones, and that kind of record does not repeat. Detroit's plus-669 point differential was the best in the East, and third in the league behind Oklahoma City and San Antonio, which is the opposite signal. Claude Fable: "its 60-22 was ratified by the conference's best differential (+669) and the core returns intact." Cade Cunningham is 24, Jalen Duren is 22, Ausar Thompson is 23, and all three are still there.
The case against is the one the models kept naming themselves. Cunningham played 64 games last season, Detroit missed on Austin Reaves in free agency, and there is no second star. GLM: "if one injury drops him to 50 games, there is no second star to hold the line."
The market rescales Detroit to 10.3 percent against the panel's 22.8, a gap of 12 and a half points and the widest on the page. It has grown by three points since August 26. The record on gaps like this runs the market's way: across 323 settled markets where this panel and the market disagreed by ten points or more, the market finished closer to the truth 242 times, which is 75 percent.
The Board
Prices are Kalshi's own, fetched September 21, 2026. The bid is what buyers are currently offering and the ask is what sellers want, so the gap between them is what a round trip costs. The panel column is the median of eight model estimates, meaning the middle number rather than an average, so a single outlier cannot drag it. It is taken after a second round in which every model read the other seven's reasoning with the names stripped off. The panel ran on August 26, before the Kawhi Leonard ruling covered below.
One quoting note before the table. Kalshi's page shows each team's last trade, which is why it lists Toronto fourth at 13 percent. Toronto's live quote is 5 bid and 7 ask, so treat the 6-cent midpoint as the read and the 13 as history. On live midpoints Detroit is fourth at 10.5 cents, and the table below ranks by those.
| Team | 2025-26 record | Point diff | Bid / ask | Market | Panel |
|---|---|---|---|---|---|
| New York | 53-29 | +519 | 20 / 21 | 20.5¢ | 19.0% |
| Philadelphia | 45-37 | -15 | 19 / 21 | 20.0¢ | 10.8% |
| Boston | 56-26 | +631 | 14 / 15 | 14.5¢ | 10.8% |
| Detroit | 60-22 | +669 | 9 / 12 | 10.5¢ | 22.8% |
| Cleveland | 52-30 | +336 | 9 / 10 | 9.5¢ | 12.0% |
| Miami | 43-39 | +191 | 7 / 8 | 7.5¢ | 8.5% |
| Toronto | 46-36 | +232 | 5 / 7 | 6.0¢ | 2.1% |
| Indiana | 19-63 | -655 | 3 / 5 | 4.0¢ | 1.5% |
| Orlando | 45-37 | +52 | 3 / 4 | 3.5¢ | 4.3% |
| Atlanta | 46-36 | +198 | 1 / 3 | 2.0¢ | 2.5% |
| Charlotte | 44-38 | +396 | 1 / 2 | 1.5¢ | 3.5% |
| Washington | 17-65 | -982 | 0 / 2 | 1.0¢ | 0.3% |
| Brooklyn | 20-62 | -819 | 0 / 1 | 0.5¢ | 0.3% |
| Chicago | 31-51 | -427 | 0 / 1 | 0.5¢ | 0.3% |
| Milwaukee | 32-50 | -509 | 0 / 1 | 0.5¢ | 0.5% |
These are model estimates, not predictions of fact and not financial advice.
More live boards from the same panel: the Eastern Conference champion board where Philadelphia leads at 22.5¢ and Detroit costs 7.5¢ · the 2027 NBA title board where San Antonio and Oklahoma City share the top at 21.5¢ · the NFC 1 seed board where the same question in football turns on winning a division first. Prices fetched September 21, 2026.
Philadelphia 76ers: The Ceiling And The 38 Games
Philadelphia is the market's second choice at 20 cents and sits tied for fourth on the panel's board at 10.8 percent. This is the second-widest gap on the page and it runs the other way.
Nobody on the panel disputed the talent. Jaylen Brown and LeBron James are the most production any East roster added this summer, and they land next to Tyrese Maxey and Joel Embiid. The objection was arithmetic. Embiid played 38 games last season. LeBron played 60 and is 41. A market that pays only for 82 games of accumulated wins treats those two numbers as the whole story. Gemini: "The Sixers pairing LeBron, Jaylen Brown, Tyrese Maxey, and Joel Embiid creates the highest theoretical ceiling, but Embiid (38 games) and LeBron (age 41) cannot be trusted to sustain a 60-win regular-season pace."
The 76ers' own front office is more optimistic. President of basketball operations Mike Gansey said on August 7 that "Joel is great. I think this is the first summer in a long time where he has been healthy," and that Embiid is lifting and training normally. That report was in front of every model, and it is why GLM priced Philadelphia at 14, the highest number any seat gave them. It did not change the median.
Toronto Raptors: The Trade Went Through
When the panel ran, Toronto's case was frozen. The Raptors agreed on June 30 to acquire Kawhi Leonard from the LA Clippers for Brandon Ingram, Gradey Dick, unprotected first-round picks in 2031 and 2033, a 2027 swap and two seconds, and the deal was put on hold nine days later while the NBA investigated whether the Clippers routed money to Leonard through a $28 million endorsement deal with the now-bankrupt company Aspiration. The panel priced Toronto at 2.1 percent against a market price of 9. Kimi called the Raptors "frozen by the uncompleted, penalty-encumbered Kawhi deal."
That changed on September 2. The league closed the investigation with a $30 million fine for the Clippers, five forfeited first-round picks and a one-year suspension for owner Steve Ballmer, and it did not suspend Leonard or void his contract. Leonard was fined $700,000. The trade was finalized on September 14.
That is the clean outcome the panel named in advance, and it said what it would do: a completion with no suspension moves Toronto to somewhere between 5 and 8 percent. The market got there from the other side. Toronto's live quote has fallen from 9 cents to 6, which sits inside the panel's own range. Fable's caution still applies: Toronto gave up Ingram and Dick to get him, both of whom played real minutes last season.
Dark Horses The Panel Won't Dismiss
Three teams sit under ten percent on the panel's board and got real arguments anyway.
Charlotte, 3.5 percent against a 1.5-cent market price. The Hornets went 44-38 and finished ninth, but their point differential was plus-396, better than Toronto's and Atlanta's and roughly what a 52-win team produces. Opus called it "the board's biggest mispricing signal," meaning the largest gap between what a team's results say it is worth and what it costs. Every seat then applied the same brake: Charlotte traded LaMelo Ball to Minnesota the day after the draft, and Sonnet put it plainly, "that differential was built with LaMelo Ball on the floor, and he's now in Minnesota."
Orlando, 4.3 percent. A 45-37 team with a plus-52 differential and, per the roster data the panel received, 20,001 minutes of last season's NBA production on hand. Not a contender, but the panel gave it more than the market's 3.5 cents.
Indiana, 1.5 percent. The Pacers won 19 games. Tyrese Haliburton missed all of it after tearing his right Achilles in Game 7 of the 2025 Finals, and he is expected back for the October 21 opener. Kimi noted the roster he returns to now includes Pascal Siakam, Andrew Nembhard and Ivica Zubac. Every seat still marked Indiana below the market's 4 cents, because the distance from 19 wins to 58 is too far for one player to carry in one season.
Every Seat's Number
These are the second-round numbers, seat by seat, from August 26. The six teams every seat put below one percent are left out of this table to keep it readable on a phone; their blends are in the board above.
| Team | Fable | Opus | Sonnet | ChatGPT | Gemini | GLM | Kimi | DeepSeek | Panel |
|---|---|---|---|---|---|---|---|---|---|
| Detroit | 23.0 | 25.5 | 22.0 | 23.0 | 22.5 | 23.0 | 22.5 | 22.0 | 22.8 |
| New York | 19.0 | 17.5 | 17.0 | 20.0 | 19.5 | 20.0 | 18.0 | 19.0 | 19.0 |
| Cleveland | 12.5 | 12.0 | 11.0 | 12.0 | 12.5 | 12.0 | 12.5 | 12.0 | 12.0 |
| Boston | 10.5 | 8.5 | 16.0 | 13.0 | 14.5 | 8.0 | 10.0 | 11.0 | 10.8 |
| Philadelphia | 10.0 | 10.0 | 11.0 | 10.0 | 10.5 | 14.0 | 11.0 | 12.0 | 10.8 |
| Miami | 8.5 | 8.5 | 9.0 | 8.0 | 8.5 | 8.0 | 9.0 | 9.0 | 8.5 |
| Orlando | 3.5 | 4.5 | 3.0 | 4.5 | 4.5 | 3.0 | 4.5 | 4.0 | 4.3 |
| Charlotte | 4.0 | 3.5 | 3.0 | 3.5 | 2.5 | 5.0 | 3.5 | 4.5 | 3.5 |
| Toronto | 3.0 | 3.0 | 1.5 | 2.5 | 1.0 | 2.0 | 2.0 | 2.2 | 2.1 |
Every seat on this panel is graded against real market settlements — records to date: Claude Fable 84% on 761 graded calls · Claude Opus 84% on 831 graded calls · Claude Sonnet 82% on 808 graded calls · Gemini 86% on 4,952 graded calls · GLM 81% on 2,544 graded calls · Kimi 83% on 2,534 graded calls · DeepSeek 80% on 2,583 graded calls. Recomputed daily; every number on this page gets graded in public once the season decides it, on the full graded scoreboard.
Where The Panel Changed Its Mind
The second round moved real numbers, and the largest single move belonged to the seat that had disagreed most in round one.
Gemini opened with New York first at 25 percent and Detroit at 15, on the argument that the Knicks' durability was the safest thing on the board. After reading the other seven, it moved Detroit up 7.5 points, cut Philadelphia by 8.5, and rewrote its case around the bar rather than the roster: "Detroit is the clearest favorite: they are the only team that cleared those historical benchmarks last season (60 wins, +669) and maintained their core... While the 19% repeat rate gives me pause, the teams that historically beat that trend looked exactly like these Pistons."
DeepSeek made the same move from a different starting point, lifting Detroit from 14 to 22 while cutting Cleveland from 18 to 12, and gave the sharpest version of the reason Cleveland came down: its plus-336 differential "is the weakest of the top group."
The seat that did not move toward anyone was Opus, which opened with Detroit at 24 and raised it to 25.5, the most confident single number anyone put on the board in either round. It also cut Boston from 14 to 8.5 and said exactly why: "the +631 was earned with Jaylen Brown playing 71 games, and swapping him for a 36-year-old George (37 GP) plus a post-Achilles Tatum on an 18,171-minute roster is a real step down, not a wash."
Boston is where the panel stayed genuinely split, and it is the only team with an eight-point spread across the eight seats. Sonnet finished at 16, Gemini at 14.5, GLM at 8. All three were looking at the same two facts: Jayson Tatum played 16 games last season and Boston still won 56, and the player they traded to Philadelphia played 71.
Kalshi's Two NBA Boards Have Stopped Disagreeing About Detroit
Kalshi lists a second market on the same 15 teams: which one wins the Eastern Conference and reaches the Finals. On August 26, Detroit was the only team on either board priced twice as likely to win the regular season as to win the conference, a ratio of 2.06. That gap has mostly closed, and it closed because the top-seed price fell toward the conference price, not the other way around.
Prices below are mid-quotes fetched September 21, 2026, rescaled so each board's 15 contracts add to 100. Rescaling matters because both boards trade with an overround, meaning the raw prices sum to more than a dollar.
| Team | Top-seed price, rescaled | Conference-title price, rescaled | Ratio |
|---|---|---|---|
| New York | 20.1 | 20.3 | 0.99 |
| Philadelphia | 19.6 | 22.3 | 0.88 |
| Boston | 14.2 | 12.4 | 1.15 |
| Detroit | 10.3 | 7.4 | 1.39 |
| Cleveland | 9.3 | 6.4 | 1.45 |
| Miami | 7.4 | 7.9 | 0.93 |
| Toronto | 5.9 | 7.4 | 0.79 |
| Indiana | 3.9 | 5.4 | 0.72 |
Indiana is still priced as a team that is dangerous in a seven-game series and unremarkable over 82. Detroit's conference price barely moved. What moved is the market's faith in the 60-win regular season, which is the one thing this contract pays for.
What The No. 1 Seed Actually Buys
The contract language is one sentence, repeated 15 times with the team name swapped: "If Detroit is the #1 seed in the Eastern Conference at the conclusion of the 2026-27 Pro Basketball regular season, then the market resolves to Yes."
It settles on the regular season, and only the regular season. April decides it, not June. There is no division requirement: the NBA seeds one through eight by conference record, with league tiebreak procedure settling ties, so winning the Central or the Atlantic buys a team nothing here. Football is different, because a top seed there has to win its division first. The lowest win total that ever took the East's top seed was 49, in the shortened 2020-21 season.
What Would Change The Panel's Mind
Every seat was asked what would move its number. Four triggers came back repeatedly, all of them checkable. One has already fired.
Joel Embiid's actual workload. This is the largest available upward move on the board. Opus put a threshold on it: Embiid "actually playing 15 of Philadelphia's first 18 games at 30+ minutes" pushes Philadelphia toward 15 percent. A minutes restriction or an early shutdown drops it under 7, and Detroit and New York absorb most of it.
The Kawhi Leonard investigation resolving. Fired. The panel said a clean completion with no suspension pushes Toronto to somewhere between 5 and 8 percent, taken from Detroit and New York. The league cleared Leonard on September 2 and the trade closed September 14.
Cade Cunningham's availability. He played 64 games last season. Fable said a Cunningham or Duren injury costing meaningful October-to-December games "moves 8-10 points from Detroit mostly to New York and Cleveland." Every seat that made Detroit the favorite named this as the way it stops being one.
Jayson Tatum's ramp-up. Full explosiveness in preseason moves Boston up several points at New York's expense; a minutes restriction or a delayed opener moves it toward the back of the top tier. This is the trigger that resolves the panel's own widest internal split.
Settlement Timeline
Kalshi lists an expected expiration of June 1, 2027, with the contract closing June 15 and an early-close condition once the seed is decided. The dates in between that matter: the regular season opens October 20, 2026, Haliburton's Indiana opener is October 21, and the regular season's final scheduled date is April 11, 2027. Panel numbers on this page are from August 26, 2026; prices are stamped as of September 21, 2026.
The Bottom Line
Kalshi and this panel agree about most of the East. New York belongs near the top, and Milwaukee, Brooklyn and Chicago do not: all three sit at half a cent on the market and at or below half a percent on the panel. After the Kawhi ruling, they now agree about Toronto too.
They disagree about one team, and the disagreement has grown. The market looks at Detroit and sees a 60-win season that ended in the second round, a franchise that did not add a second star, and a base rate that says top seeds rarely repeat. The panel looks at Detroit and sees the only roster in the conference that already cleared the bar this market actually pays for, returning almost entirely intact, with a 24-year-old at the center of it.
The market's version has last season's playoffs behind it. The panel's version has the twenty-two-year rule that this market has never once gone to a team that did not have the East's best record. April will pick one.
Neither is a prediction of fact or financial advice. Kalshi's event contracts are for adults (18+) on a CFTC-regulated exchange, and availability varies by state. What this page is, is a record: 15 teams, eight models, every price hidden, dated August 26, and graded in public when the regular season ends.
Hottest Prediction Markets Right Now
- 2028 Democratic presidential nominee · $194M traded
- 2027 Pro Football Champion · $72M traded
- 2028 U.S. Presidential Election winner? · $62M traded
- Pro Baseball Champion · $60M traded
- 2028 Republican presidential nominee · $60M traded
- Bitcoin price at the end of 2026 · $33M traded
Every market above links to our full model verdict; browse them all on the OddsShopper prediction markets hub, and see how every settled call actually scored on the full graded scoreboard.
