Giants win total odds after Week 2: The eight-win contract that traded at 64¢ the morning after New York beat Dallas sits at 34¢ bid, 35¢ ask. The 28-6 loss in Los Angeles did not just take back the six cents of belief traders added after Week 1. It took the whole Week 1 jump and 16 cents more, and the contract now sits below the 50.5¢ it held before the Giants had played a game. The ladder carries 6.22 expected wins; Tuesday morning's board had it at 7.6, and no other team lost more than 0.4 in the same day. Nine AI models that priced the season blind on September 14 put the eight-win contract at 58%. That number stays on the record, and it now sits 23 points above the market it used to trail.
Prices as of Wednesday, September 23, 2026, 7:31 a.m. ET, Kalshi bids and asks before fees. The panel numbers are from its September 14 run.
| Giants win at least | Kalshi now | September 14 | Before Week 1 | Panel (Sept 14) |
|---|---|---|---|---|
| 6 games | 57¢ bid, 58¢ ask | 76¢ bid, 93¢ ask | 69¢ | 85% |
| 7 games | 42¢ bid, 43¢ ask | 67¢ bid, 85¢ ask | 61.5¢ | 73.5% |
| 8 games | 34¢ bid, 35¢ ask | 58¢ bid, 65¢ ask | 50.5¢ | 58% |
| 10 games | 9¢ bid, 12¢ ask | 37¢ bid, 41¢ ask | 29¢ | 27.5% |
A 34.5¢ contract pays $1.00 if New York gets to eight wins and nothing if it does not; the midpoint reads as the crowd's percentage chance. New to Kalshi? How NFL trading works there.
Stay ahead of the markets.
Daily insights and expert picks on Kalshi, Polymarket, and what's moving markets.
Free forever. Unsubscribe anytime.
Hottest Prediction Markets Right Now
- 2028 Democratic presidential nominee · $217M traded
- 2027 Pro Football Champion · $106M traded
- Pro Baseball Champion · $72M traded
- 2028 U.S. Presidential Election winner? · $66M traded
- Fed decision in Sep 2026? · $66M traded
- 2028 Republican presidential nominee · $64M 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.
Betting a Kalshi combo on this game?Kalshi Parlay Simulator before you buy it: 10,000 simulated games price every combo next to the price on Kalshi.
What The Rams Loss Did To The Ladder
On September 14 this page worked out what one win was worth: from a 50.5¢ midpoint before kickoff, a win over Dallas that taught traders nothing would have put the eight-win contract near 58¢. The overnight trades averaged 64¢, and the extra six cents was a bet that Jaxson Dart's 23-of-29 night said something about the other 16 games. That six cents lasted one week.
The Giants went to Los Angeles as the lower ladder, 7.6 expected wins against the Rams' 11.0, and lost 28-6. A loss the market expects costs a ladder only the fraction of a win the underdog was priced to take. This one cost 1.4 wins between Tuesday's board and Wednesday's, the largest cut on any of the 32 ladders, and it moved the Giants from 22nd in the league to 26th. Most of that is belief leaving, not arithmetic.
Every rung from four wins up now sits below where it was before Week 1. Eight wins gave back the most: 50.5¢ before the season, 64¢ in the trades after Dallas, 34.5¢ now. Seven wins went from 61.5¢ to 42.5¢, six wins from 69¢ to 57.5¢, ten wins from 29¢ to 10.5¢. The rung that sat at 50 cents before Week 1 was eight wins; the ladder now crosses 50 cents at 6.22.
The loss also cleared up the disagreement over what the Dallas win meant. On September 14 only three thresholds were quoted less than 10 cents wide. Six, seven and eight wins are now one cent wide and nine is two. The quotes were wide because traders disagreed over what the Dallas win meant; the disagreement is over.
The Board
Each row is a separate yes-or-no contract that pays if New York finishes the regular season with at least that many wins; ties and playoff games do not count. The panel column is the September 14 median of nine models.
| Giants win at least | Bid | Ask | Mid | Panel (Sept 14) |
|---|---|---|---|---|
| 1 game | settled yes | 100¢ | ||
| 2 games | 84¢ | 98¢ | 91¢ | |
| 3 games | 86¢ | 99¢ | 92.5¢ | |
| 4 games | 76¢ | 92¢ | 84¢ | 96.5% |
| 5 games | 64¢ | 77¢ | 70.5¢ | 92.0% |
| 6 games | 57¢ | 58¢ | 57.5¢ | 85.0% |
| 7 games | 42¢ | 43¢ | 42.5¢ | 73.5% |
| 8 games | 34¢ | 35¢ | 34.5¢ | 58.0% |
| 9 games | 18¢ | 20¢ | 19¢ | 43.0% |
| 10 games | 9¢ | 12¢ | 10.5¢ | 27.5% |
| 11 games | 0¢ | 19¢ | 9.5¢ | 15.8% |
| 12 games | 1¢ | 9¢ | 5¢ | 8.0% |
| 13 games | 0¢ | 6¢ | 3¢ | 3.8% |
| 14 games | 0¢ | 3¢ | 1.5¢ | 1.5% |
| 15 games | 0¢ | 1¢ | 0.5¢ | |
| 16 games | 0¢ | 1¢ | 0.5¢ | |
| 17 games | settled no | 0¢ |
The panel numbers on this page are estimates, not predictions of fact and not financial advice. Every number in this piece gets graded in public once the market settles, on the full graded scoreboard.
Add the bids and the market implies a season of at least 5.71 wins; add the asks and it is 6.73. The panel's nine ladders on September 14 ran from 7.69 to 8.59 projected wins, every one above the top of the market's range. The gap on the eight-win contract is 23.5 points and on seven wins it is 31. Our settled ledger says how gaps like these have tended to end: when this panel's median and the market sat 10 or more points apart, the market finished closer 238 times out of 354, or 67%.
More from the same panel: all 32 NFL win totals · the NFC East race · the Cowboys' own win total · the 2027 Super Bowl.
Every Seat's Number
Nine models priced the season on September 14 from one data card (settlement terms, schedule, the Week 1 box score, last season's game log, the roster as of that date) without seeing a price, then read each other anonymously and revised or held. Projected wins include the banked win. These numbers are on the record and get graded in January; the panel has not been re-run since the Rams game.
| Seat | Projected wins | Chance of 8+ | Chance of 10+ | Reason, in one line |
|---|---|---|---|---|
| Fable | 8.59 | 67.0% | 36.0% | Last year's point differential plus regression puts the middle at nine before any coaching credit |
| Opus | 8.28 | 63.0% | 31.0% | A seven-win team that gets two close games back plays near .460 against an average schedule |
| DeepSeek | 8.26 | 65.0% | 32.0% | Dart's efficiency, Nabers healthy and Harbaugh's record justify shading up |
| Kimi | 8.18 | 58.0% | 31.0% | Harbaugh and a clean debut help; seven strong opponents and a thin secondary cap it |
| Gemini | 8.05 | 59.5% | 27.5% | Sixteen games at about a 45% win rate, anchored by regression |
| Sonnet | 7.99 | 57.8% | 27.3% | Two close-game wins come back, capped by seven 11-win opponents from 2025 |
| Grok | 7.97 | 57.0% | 25.5% | The banked win and regression, tempered by the schedule |
| GLM | 7.85 | 53.0% | 24.5% | The Lawrence-less defense and the schedule cap the ceiling |
| ChatGPT | 7.69 | 53.2% | 24.6% | The banked win matters most; Week 1 is modest evidence |
| Panel Median | 8.05 | 58.0% | 27.5% | Every seat but one puts the median season at eight wins |
Every seat on this panel is graded against real market settlements. Records to date: Gemini 86% on 9,496 graded calls · Grok 84% on 7,920 graded calls · Kimi 86% on 4,894 graded calls · GLM 84% on 4,494 graded calls · DeepSeek 84% on 4,511 graded calls. Recomputed daily; the full scoreboard is public.
The panel's argument was regression, not the box score. Last season's Giants won four games with a point differential that normally produces seven, and went 1-7 in games decided by eight points or fewer; every seat started from a season near eight wins and asked how much Week 1 added. The answers ran from 0.3 to 0.6 wins, median 0.5, the same half win the overnight trades implied. The lowest seat, ChatGPT at 7.69 wins and 53.2% on eight, is still 18.7 points above the market.
Sonnet held rather than revise, and its anchor is the one the last nine days have tested: every team since 2010 that won five and a half games or fewer and then won its next opener, 35 of them, of which 18 reached eight wins. That is 51%, and the market has now gone 17 points under it.
What Two Games Say
The Giants have played one game at home and one on the road, and the ladder treats the second as the truer one. They beat Dallas 28-20 as a 39¢ underdog on a night the offense scored on all four trips inside the 20. Eight days later they scored six points in Los Angeles against a Rams team that had scored seven against San Francisco the week before. Two games, 34 points for and 48 against, a one-score win and a three-score loss.
The opponents' second games keep the Week 1 result honest. Dallas beat Washington 37-20 and carries 9.1 expected wins, so the team New York beat is still priced as a nine-win team. The Rams carry 10.9, the sixth-highest ladder in the league, and 13 of 16 Week 2 games went to the higher-projected side. Losing in Los Angeles was the expected result. Losing by 22 is what cost the belief.
That is the whole distance between the panel and the market. The panel's 58% on eight wins is what the banked Dallas win alone produced from the pregame price; it never credited the half win of belief, so it had nothing to give back. The market credited it, then took it back and more. One side is treating the Rams game as a road loss to a 10.9-win team. The other is treating it as the season's real read.
Week 3: Tennessee At The Giants
The Titans visit MetLife Stadium on Sunday, September 27, at 1:00 p.m. ET at 0-2, having lost to the Jets 23-10 and to Philadelphia 24-20, both at home; their ladder carries 4.5 expected wins, second-lowest in the league. New York is the favorite at home, 57¢ bid, 58¢ ask to win, down from 63¢, with Tennessee at 42¢ bid, 43¢ ask. The Giants to win by more than 2.5 points sit at 52¢ bid, 53¢ ask; Tennessee to win by more than 2.5 is 37¢ bid, 38¢ ask. On the total, over 40.5 points is 45¢ bid, 46¢ ask, and over 37.5 is 57¢ bid, 58¢ ask. DraftKings opened the game at Giants -175 with a 45.5 total and has it at -135 and 39.5.
At 1-1 with 15 games left, eight wins needs seven more and six wins needs five. Win Sunday and eight needs six of the last 14, and the eight-win rung climbs from 34.5¢ on the read that a 22-point road loss was one bad night. Lose at home to an 0-2 team and eight needs seven of 14, and the seven-win rung, 42.5¢ now, becomes the ladder's coin flip. At 57.5¢ a Giants win adds 0.43 of a win to the ladder before any belief and a loss takes away 0.58, so the downside is already the bigger number.
What Moves The Ladder Next
- The three soft games. Tennessee, Arizona and Washington, Weeks 3 through 5, all won five games or fewer in 2025 and all three lost in Week 2; their ladders sit at 4.5, 5.7 and 6.0 expected wins. Sweep them and the Giants are 4-1 with eight wins needing four of the last 12, and the panel's 58% is the right number again. Split them and the ladder stays near six.
- The seven-win rung. It is 42¢ bid, 43¢ ask, the first threshold under 50 cents. If it crosses back above 50 before the Washington game, the Rams loss was priced as one night; if six wins is the coin flip by then, the market has settled on a season 1.5 wins under the panel's lowest seat.
Settlement Timeline
| Date | What happens |
|---|---|
| Monday, September 21, 2026 | Week 2 at the Los Angeles Rams: lost 28-6, ladder cut 1.4 wins |
| Sunday, September 27, 2026 | Week 3 vs. Tennessee, New York 57¢ bid, 58¢ ask to win |
| Sunday, October 11, 2026 | Week 5 at Washington, the last of three straight games against 2025's weaker teams |
| Sunday, January 10, 2027 | Week 18 vs. Philadelphia, the last game that counts toward any contract |
| Monday, January 18, 2027 | Listed expiration of every KXNFLWINS-27NYG contract |
An eighth Giants win in December settles the eight-win contract early. This page is re-scored when the story moves, as of September 23, 2026, and gets a graded look back on the same URL after Week 18.
More on this: Browns Win Total 2026: The Panic Lasted One Week · Cardinals Win Total 2026: The Ladder Kept The Chargers Win · Dolphins Win Total 2026: Even Harsher After Two Losses · Patriots Win Total: The Books Agree On The Number. The Machines Don't · Saints vs Giants Picks: Why Are The Saints Favored?
