Kevin Durant Next Team Odds: Houston 85-89C, Panel At 90%

Two weeks ago a Kalshi next-team board settled on a nine-cent longshot. Philadelphia closed at 9 cents on July 24 while the favorite, Miami, closed at 43; the next day LeBron James was a 76er, Philadelphia printed 99, Miami collapsed to a penny, and every one of the other 29 contracts on that board went to zero. That result is the reason this one is worth reading carefully, because Kevin Durant's board is the same product wearing the opposite shape. Where LeBron's market spread real money across four live destinations, Durant's puts 85 to 89 cents on a single contract that says nothing happens at all, and the 29 franchises underneath it are fighting over what is left. We had a price-blind AI panel score all 30 outcomes from a fetched dossier. It came back agreeing with the expensive part of the board and disagreeing with almost everything else.

The Quick Answer

Kalshi's "Stays with Houston or Retires" contract trades at 85 cents bid, 89 asked, and our seven-model price-blind panel blends the same outcome to 89.9% after a revision round. The panel is fractionally above the market on the favorite and well below it nearly everywhere else: the 29 destination contracts carry about 35 cents of implied probability between them against a 10% panel blend. Only one destination survives contact with the panel's reasoning, and it is Detroit, whose 3.9% blend is the board's biggest gap in the models' favor: nearly double its 2-cent midpoint, though still inside the 4-cent ask. The full board, the reason the market's loudest-looking price is not a price at all, and the dated catalyst that could restart this story are below.

Durant is not a free agent, which is the fact the whole board turns on. He signed a two-year, $90 million extension with Houston in October 2025, roughly $30 million below the maximum he was eligible for, and the only route to a different jersey before this market expires is a completed trade or a waiver claim. That structural point separates this board from the two next-team markets our panel has already seen settle.

Free: The Weekly PM Market Brief — the 8-model panel's graded record, the week's biggest market-vs-model gaps, and what's spiking next. One email, Sundays. Sign up in the box at the end of this article.

New to Kalshi event markets? Start here →

The Board

Prices are the live YES bid and ask on Kalshi as of August 9, 2026, a Sunday afternoon with the league in its offseason lull. The blend is the equal-weight mean of the seven models' round-two numbers.

Outcome	Market (YES bid–ask)	Panel Blend
Stays With Houston Or Retires	85–89c	89.9%
Detroit	0–4c	3.9%
Miami	1–2c	1.3%
Golden State	1–3c	0.9%
New York	0–1c	0.9%
Cleveland	2–4c	0.1%
Philadelphia	0–5c	0.1%
San Antonio	0–12c	0.3%
Dallas	0–9c	0.3%
The Other 21 Teams	0–1c each, plus Minnesota, Denver and Washington above the floor	2.3% combined

These are model estimates, not predictions of fact and not financial advice.

More live boards from the same panel: Steph Curry's next team, the other aging-icon board, Golden State 92¢ · Kawhi Leonard's next team, the same market family pointing the other way, Toronto 66¢ against a 22¢ incumbent · Bronny James' next team, the Lakers 84¢ with a roster-crunch mechanism underneath. Prices fetched August 9, 2026.

Every Seat's Number

Each model priced all 30 outcomes in a single pass, without seeing any price, then read the other seats' anonymized reasoning and could revise. These are the round-two numbers.

Outcome	Market	Fable	Opus	Sonnet	GPT	GLM	Kimi	DeepSeek	Blend
Stays With Houston Or Retires	85–89C	91.8	93.5	81.4	84.0	89.9	91.8	97.0	89.9%
Detroit	0–4C	3.1	3.0	5.0	5.0	4.0	4.1	3.0	3.9%
Miami	1–2C	1.0	1.0	2.0	2.0	2.0	1.0	0.0	1.3%
New York	0–1C	1.0	1.0	1.0	1.0	1.0	1.0	0.0	0.9%
Golden State	1–3C	1.0	0.5	1.0	2.0	1.0	1.0	0.0	0.9%
Los Angeles Lakers	0–1C	1.0	0.5	1.0	1.0	0.0	0.0	0.0	0.5%
Oklahoma City	0–1C	0.0	0.0	0.0	1.0	1.0	1.0	0.0	0.4%
Minnesota	0–2C	1.0	0.5	1.0	0.0	0.0	0.0	0.0	0.4%
Boston	0–1C	0.0	0.0	1.0	0.0	1.0	0.0	0.0	0.3%
San Antonio	0–12C	0.0	0.0	1.0	1.0	0.0	0.0	0.0	0.3%
Dallas	0–9C	0.0	0.0	1.0	1.0	0.0	0.0	0.0	0.3%
Los Angeles Clippers	0–1C	0.0	0.0	1.0	1.0	0.0	0.0	0.0	0.3%
Cleveland	2–4C	0.0	0.0	1.0	0.0	0.0	0.0	0.0	0.1%
Philadelphia	0–5C	0.0	0.0	0.0	1.0	0.0	0.0	0.0	0.1%
Denver	1–2C	0.0	0.0	1.0	0.0	0.0	0.0	0.0	0.1%
Milwaukee	0–1C	0.0	0.0	1.0	0.0	0.0	0.0	0.0	0.1%
Phoenix	0–1C	0.0	0.0	1.0	0.0	0.0	0.0	0.0	0.1%
The Other 13 Teams	0–2C Each	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0%

Every seat on this panel is graded against real market settlements — records to date: GLM 81% on 2,307 graded calls · Kimi 83% on 2,285 graded calls · DeepSeek 80% on 2,332 graded calls. Recomputed daily; the full scoreboard is public.

These are model estimates, not predictions of fact and not financial advice. Seven of the panel's eight seats returned a board on this market; the eighth did not run, and the blend is the mean of the seven that did. The remaining 13 franchises (Washington, Brooklyn, Toronto, Chicago, Indiana, Atlanta, Charlotte, Orlando, Portland, Utah, Sacramento, Memphis and New Orleans) drew a zero from every seat in round two. Every number here lands in our Model Verdict ledger, and each one gets graded in public once the market settles: the full graded scoreboard keeps the panel's complete record, wins and misses alike.

The Market At A Glance

Venue	Kalshi, a CFTC-regulated event-contract exchange (18+; availability varies by state, as of August 2026)
The Event	"What will be Kevin Durant's next team?" (Kalshi event KXNEXTTEAMNBA-26KDURANT7), 30 contracts, roughly $167,000 traded and about $128,000 of open interest
The Contract	Each destination reads: "If Kevin Durant's next team is [team] before Oct 21, 2026, then the market resolves to Yes." Per the NEXTTEAM contract terms, a player "officially joins" a team through a signed contract of any duration, an official league-approved trade, or a waiver claim, and the determining moment is the transaction becoming official, not the report, not the physical, not the press conference. Sign-and-trades resolve to the final destination team, multi-team trades resolve to whoever acquires the player's rights, voided deals are disregarded, and official retirement without a new team resolves the stay-side contract Yes.
Settles	October 21, 2026, 10:00 AM ET, or earlier, since the contract expires as soon as a qualifying transaction becomes official
Prices Below As Of	August 9, 2026, live YES bid and ask

Read the strike wording closely, because it defines the question. "Stays with Houston or Retires" is a single compound contract that uses the terms' OR logic to bundle two different endings: he is still a Rocket on October 21, or he has officially retired without signing anywhere. There is no standalone Houston contract and no standalone retirement contract, so a reader who wants to price "he stays put" and a reader who wants to price "he walks away" are looking at the very same contract.

What The Models Were Given

Every fact on the panel's card was fetched at generation time, dated and sourced, and no seat saw a price. The ledger half: Durant, 37, appears on Houston's 16-man roster, coming off 78 games at 26.0 points, 5.5 rebounds and 4.8 assists on 52.0% shooting and 41.3% from three, for a team that went 52-30 and closed the regular season 9-1 before going 2-4 in the playoffs, one series lost in six. The contract half: the two-year, $90 million extension signed in October 2025, with a player option for 2027-28, taken about $30 million below his maximum. His business partner Rich Kleiman said at the time that Durant understood he was sacrificing money "so that the two sides could partner for the long term and give the franchise team-building flexibility."

Then the news half, which is where this board's whole argument lives, and it moves in one direction. On July 4, Yahoo Sports relayed ClutchPoints insider Brett Siegel reporting that Houston did not view Durant as untouchable, that Detroit was expected to be first in line if he became available, and that Durant had interest in playing alongside Cade Cunningham. Twelve days later, on July 16, Heavy's Sean Deveney wrote the counter: "There does not seem to be a market for Durant...Rumors of Pistons interest has faded as the focus there is on re-signing Jalen Duren." Deveney allowed that a deal might resurface once Duren was resolved and called that a longshot. On August 1, at a USA Basketball Foundation charity event, Durant said he hopes to be picked for the 2028 Los Angeles Olympics — "I stay on top of my game and hopefully I get that call" — which is a man planning to be playing in two years, not one contemplating an exit. And on August 8, the freshest item on the card, Hoops Rumors relayed Omari Sankofa II of the Detroit Free Press reporting that Detroit and Duren remain in a standoff that could run all the way to October 1, with no mention of Durant anywhere in it.

The card also carried something no rumor column has: our own settled boards. LeBron James's next-team market resolved Yes on Philadelphia and Stefon Diggs's resolved Yes on Washington, both re-verified against the exchange. But the card handed the models the cause rather than the base rate, and the cause is where those precedents break. Both of those subjects were unsigned players inside an open, publicly reported process — free to choose, with agents briefing national outlets. Durant is under contract. Every seat tested that difference and every seat concluded it does not transfer. "The settled-board cause (unsigned player in an open process) is structurally absent here," Claude Fable wrote. Kimi K3 put it more bluntly: "there is no free-agency mechanism at all."

Houston: The Panel Agrees, And Then Goes Further

The market's 85-to-89-cent quote already says this is close to settled. The panel puts the probability a shade higher still, blending the compound contract to 89.9%, a small edge on the most expensive price on the board, and a notable one given that the last board in this family burned everyone who assumed the star stayed put.

Nearly all of that 89.9% sits on the stay branch rather than the retire branch. The models handled the compound strike the same way: the August 1 Olympics comment makes retirement a rounding error, but because retirement and staying share one contract, that evidence only shuffles weight inside the strike rather than away from it. Fable put retirement at "roughly half a point inside the 90." Opus put it at "roughly a quarter of a point."

What replaces the retirement path is arithmetic. To beat this contract, some team has to take on one of the largest salaries in the league and pay real assets for a 37-year-old, and it has to get that done inside about ten weeks. Kimi K3 named the timing problem that most rumor coverage skips entirely: the only mechanism that beats the incumbent contract is a salary-matching trade executed before October 21, rather than at the in-season deadline, which falls after this market has already expired. That is the structural cruelty of this market. February is when star trades actually happen, and February is after this contract has already paid out.

Detroit: The One Place The Panel Is Braver Than The Market

Detroit carries the board's widest gap in the models' favor: a 3.9% blend against a 2-cent midpoint, still inside the 4-cent offer, and every one of the seven seats scored it as the top non-incumbent destination. It is also the second-most-traded contract on the board at roughly $23,000, which tells you the July rumor cycle left a mark even after the reporting cooled.

The panel's case is narrow and specific rather than enthusiastic. Detroit is the only franchise linked to Durant anywhere in the dated reporting the panel was given, so it inherits the entire "something we cannot see" allowance by default. Claude Opus was explicit about why it raised the number in the revision round, calling Detroit "the only ever-named acquirer" whose "stated blocker, the Duren standoff, has a dated resolution point of Oct 1 that leaves a three-week runway inside the window." That is the sharpest observation on this board. The exact thing reported to have killed Detroit's interest has a deadline, and that deadline falls three weeks before this market expires.

Sonnet and ChatGPT, the panel's two most skeptical seats on Houston at 81.4% and 84.0%, both put their extra probability here. ChatGPT held its 84 through the revision round and said why: it still thinks 90-plus "underprices the July 4 report about Houston not viewing him as untouchable" and the locker-room tension that followed the playoff exit.

San Antonio's 12-Cent Offer Is Not A 12-Cent Price

The loudest-looking number on this board is San Antonio, quoted 0 bid and 12 asked. A reader scanning for the market's second choice would stop right there, and they would be reading an offer rather than a price.

Nothing on the bid side supports that 12 cents, and the last price anyone actually paid for San Antonio is one cent. That is the entire gap: an eleven-cent spread between what a seller is asking and what the market last agreed the contract was worth. On a book this thin, an unfilled offer is a hope, not a probability, and nobody has to take it.

Dallas is the same trap wearing the opposite costume. Its last traded price is 20 cents, which looks like a live contender until you see the quote around it: 0 bid, 9 asked. Nobody will pay you 20 cents for that contract today, and nobody is bidding at all. Read a lone last price off a book this quiet and you will mistake one old fill for a market. The panel scored both teams at 0.3%.

There is a nice symmetry in how the panel arrived at that. San Antonio was the one number a model publicly took back. In round one GLM 5.2 had allocated 2% to the Spurs; in round two it removed it, writing that it had assigned that probability "with no evidentiary basis in the data card — no Source Agency report links SA to Durant." The market's most eye-catching offer and the panel's most-retracted number landed on the same team, for the same reason: there is nothing under it.

Dark Horses The Panel Won't Dismiss

Below Detroit the numbers get small, but a handful drew a real reason from at least one seat rather than partition filler:

• Miami (1.3% Blend). The panel's second choice behind Detroit, and one of the few destinations the models rate above the price: six of the seven seats gave it a point or more against a 1-to-2-cent quote. It is the standing star-hunting front office, priced as a habit rather than a report.
• New York (0.9%). Six of seven seats scored it, despite the contract sitting at the exchange's 1-cent floor with barely $1,500 traded. This is the clearest case on the board of the panel finding a team more plausible than the market bothered to price.
• Golden State (0.9%) And Oklahoma City (0.4%). Asset-math longshots: the two rosters seats named when asked which contenders could construct a package at all, rather than which ones want to.
• Cleveland And Philadelphia (0.1% Each). Worth naming because the market disagrees hardest here: Cleveland trades 2 to 4 cents and Philadelphia offers at 5, with about $10,500 and $14,200 traded respectively, and each drew a number from exactly one seat out of seven: Sonnet on Cleveland, ChatGPT on Philadelphia. Philadelphia's volume is the third-largest on the board, which is what a market looks like when a name is fresh in everyone's memory for an entirely different reason.

Where The Panel Changed Its Mind

The revision round on this board was unusually quiet, and the quiet is the finding: six of seven seats held their headline number after reading everyone else's reasoning, and the blend moved six-tenths of a point. When a panel is looking at four dated facts that all point the same way, there is nothing left to persuade anyone with. Three arcs are worth quoting.

"In round 1 I allocated 2% to San Antonio with no evidentiary basis in the data card — no Source Agency report links SA to Durant. I redistributed that 2%: 1% into HOU (raising to 89%, reflecting that the unjustified speculative probability most properly returns to incumbency given no reported suitor) and 1% spread across Boston and the Lakers." — GLM 5.2, the only seat to materially move a destination

"Held the incumbent at 90 — no peer surfaced evidence outside the card, and both the 96 and 82-84 positions trade on emphasis, not new facts. Only change: shifted 1 point of challenger mass toward Detroit (2→3) since it is the sole sourced suitor and my round-1 allocation gave it barely more than never-named teams." — Claude Fable

And the seat that would not move, which matters more than the ones that did:

"I held at 84%. The strongest opposing argument from peers is that the short August-to-October window, Houston's stable contending roster, and Durant's below-max extension justify pushing the incumbent strike closer to 90-96%. I had already weighed those factors in round 1. I still think 90%+ underprices the July 4 report about Houston not viewing him as untouchable, the reported locker-room tension..." — ChatGPT, the panel's second-lowest seat on Houston

Fable's summary of the split is the fairest reading of the whole exercise: a high camp at 93 to 96 leaning on the complete absence of active talks, and a low camp at 82 to 84 leaning on the unpriceable tail of a quiet Houston pivot after a disappointing spring. Neither camp found a new fact. They weighted the same four the market can also see.

What Would Change The Panel's Mind

Four concrete, checkable triggers sit between now and settlement, and each one pushes the number in a stated direction:

• October 1, Duren's Qualifying-Offer Deadline. The dated event Opus built its Detroit number around. If Duren signs an extension or accepts the $9.6 million qualifying offer well before that date, the blocker reported on July 16 disappears with three weeks left on the clock, and Detroit's 3.9% becomes the panel's most likely upgrade. If the standoff runs to the deadline itself, Detroit's window closes with it.
• A Sourced Report That Houston Is Actively Shopping Him. Five of seven seats named this as the single most price-moving item, and none of them have seen it. The July 16 and August 8 items both describe a market that is not forming, rather than a team taking calls. This one moves the incumbent contract down hard.
• Late September, Training Camps Open. The point in the calendar where veteran discontent historically becomes reporting. A camp Durant attends quietly pushes the stay contract toward its ceiling; anything else is the first real fuel the go side has had.
• A Second Named Suitor, Anywhere. Detroit currently carries the entire "something we cannot see" allowance because it is the only franchise ever linked. A second name would not just add a destination, it would break the structural argument every seat leaned on — that the absence of a market is itself the evidence.

Settlement Timeline

When	What
August 9, 2026	Prices and panel numbers on this page, as of today
Mid-August 2026	The full 2026-27 NBA schedule is expected, which fixes opening night
Late September 2026	NBA training camps and media days open
October 1, 2026	Jalen Duren's deadline to accept Detroit's qualifying offer, the dated resolution point behind the panel's Detroit number
October 20, 2026	Projected opening night for the 2026-27 season, not yet officially confirmed
October 21, 2026, 10:00 AM ET	Settlement, or earlier, if a qualifying transaction becomes official first

This board re-scores when the story moves. Every number above is stamped August 9, 2026, and if a report lands that fires any of the four triggers, the panel runs again and this page carries the new numbers alongside these ones.

How To Read These Prices

A Kalshi contract pays $1 if the outcome happens and $0 if it does not, so the cent price reads as an implied probability. Three cautions before any number above does work for you.

The board does not add up to 100, and most of the excess is a floor. All 30 contracts sum to about 122 cents at their midpoints. The 29 destinations account for 35 of those cents against a 10% panel blend, but 19 of them are quoted 0 bid / 1 ask, where a half-cent midpoint is a rounding convention rather than anyone's opinion. Strip those out and the real destination market is ten contracts, not 29.

Thin books cut both ways. The whole board has traded roughly $167,000, which is small enough that a single order moves a quote. That is why the San Antonio offer looks like news and is not, and it is also why entry and exit outside the Houston, Detroit, Philadelphia and Cleveland books may not happen at the price on the screen.

A heavy favorite is a slow, asymmetric hold. Buying the stay contract at 89 cents risks 89 to make 11, minus fees, over ten weeks: being wrong once costs eight times what being right pays. A small edge on a heavy favorite only means anything spread across many independent markets, never concentrated in one.

Kalshi is a CFTC-regulated event-contract exchange, not a sportsbook; you must be 18 or older and located in an eligible state, and availability varies by state. If event contracts are new to you, start with how to bet NBA on Kalshi, and our Kalshi fees guide covers what a round trip actually costs. For nightly NBA prices once the season tips, the OddsShopper live odds screen and free expert picks today do the line-shopping work. These are model estimates, not predictions of fact and not financial advice.

FAQ

Is Kevin Durant getting traded before the 2026-27 season? Nothing has been reported. As of August 9, 2026 no Source Agency has reported a Durant trade, and he appears on Houston's roster under the two-year extension he signed in October 2025. The last substantive reporting ran the other way: on July 16 Heavy's Sean Deveney wrote that there did not seem to be a market for him and that Detroit's interest had faded. The Kalshi board prices the possibility at roughly 11 to 15 cents spread across 29 destinations.

Does retirement count as a next team in this market? Yes, on the stay side. Kalshi's NEXTTEAM contract terms say that if a player officially retires without joining a new team, the no-new-team outcome resolves Yes. On this board that outcome is bundled with staying into one contract, "Stays with Houston or Retires," so there is no separate retirement price to read.

When does the Kevin Durant next team market settle? October 21, 2026 at 10:00 AM ET, the morning after the projected start of the 2026-27 NBA season. It can also expire early: the contract closes as soon as a qualifying transaction becomes official. Because the clock stops before the season is a day old, an in-season trade at the February deadline would come far too late to settle any destination contract.

Hottest Prediction Markets Right Now

• 2028 Democratic presidential nominee · $168M traded
• 2028 U.S. Presidential Election winner? · $57M traded
• 2027 Pro Football Champion · $54M traded
• Pro Baseball Champion · $45M traded
• More tech layoffs in 20​26 than in 2025? · $31M traded
• Bitcoin price at the end of 2026 · $30M traded

Every market above links to our full AI model verdict; browse them all on the OddsShopper prediction markets hub, and see how every settled call actually scored on the full graded scoreboard.

The Bottom Line

There is no Durant trade saga right now. There is a July rumor that a July counter-report retired, a board still carrying 35 cents of destination probability against a panel blend of 10, and a market whose flashiest quote turns out to be one unfilled offer above a book that trades at a penny. The panel's read is that the crowd has the favorite roughly right and is fractionally cheap on it, that Detroit is the only destination the models rate anywhere near its price, and that the interesting date is not opening night but October 1, when the one reported obstacle to the one reported suitor either clears or expires. LeBron's board is the reason to hold that view loosely: two weeks ago this same panel watched a next-team market resolve on a contract that had closed the previous day at nine cents. The difference is a signed extension, and whether that difference holds is exactly what this page will be graded on. To be explicit one final time: these are model estimates, not predictions of fact and not financial advice.