The Quick Answer
New York Giants playoff chances in 2026 come down to the gap between a story and a price. The story is John Harbaugh taking over a team that went 4-13 in 2025; the price is a season win total of 7.5 as of July 28, and Kalshi's market on the exact question trades yes at 28 cents bid, 29 cents ask as of July 30, 2026. Our feed marks the question all but decided in the won't direction, decided by price rather than by math, and we argue the won't side too: teams that missed the previous postseason make it back just 30.5% of the time, and this column says the Giants stay in that majority.
One number frames the whole page before anything else. BetMGM publishes Harbaugh at 65% against the spread; computed from the primary 1999-2025 game log, he is 161-146-10, which is 52.4%, across 317 games, on a log that includes playoff games most published tables leave out. That correction is the reason this page exists. Below is the full case: the win cliff that makes 7.5 such a hard place to start, what the Harbaugh number does and does not mean, where our projected NFC bracket leaves New York, and why our primary pointer goes to a venue that pays us nothing.
This is the Giants column of our NFL playoff odds hub, which runs one of these for all 32 NFL teams with the same treatment: the market's real number with its date, a measured base rate, and against-the-spread records computed from the primary 1999-2025 game log. Our won't call here is an editorial one, argued as a case rather than issued as a model output, and this time the market's number leans the same direction we do. Agreeing with the price is not an achievement, so the argument below stands on football and arithmetic, with every sample size printed.
Start Where Every 4-13 Team Starts
The base rate first, because every one of these columns argues against one. Since 1999, teams that missed the playoffs made the field the following season 30.5% of the time (155 of 508, with a 95% confidence interval of 26.7% to 34.6%). That number is not a wall; it reflects real churn. An average of 5.96 playoff spots turn over every year, with a range of four to eight, and since 2020 the fourteen-team format has churned slightly faster at 6.17 spots per season. Roughly six teams that missed last January will be playing this one.
So the honest framing is not "the Giants can't get in." It is that a team coming off 4-13 starts as one of 18 misses fighting for about six open chairs, and the market's own price says this particular team is not among the favorites for one. That price is where the argument gets specific.
The Market Says 7.5, And The Playoff Line Sits Near Ten
As of July 28, 2026, the Giants' season win total sits at 7.5. There is now a second market read, and it is a direct one: Kalshi's per-team playoff market opened for trading, and as of July 30 yes on the Giants sits at 28 cents bid, 29 cents ask. An exchange quote has no bookmaker hold to strip, only a spread between traders, so read that as a 28-to-29 band with a midpoint around 28.5% rather than a probability precise to the decimal. Our feed marks this question all but decided at that price, and decided in the direction this column argues, though a 29-cent contract is a strong lean and not a settled outcome; nothing settles until late January. Both numbers describe a team that lives well below the playoff line. Here is what actual win counts have meant in the 17-game era:
| Season Wins | Made playoffs | Sample |
|---|---|---|
| 7 | 0% | 17 teams |
| 8 | 11% | 18 teams |
| 9 | 39% | 23 teams |
| 10 | 94% | 16 teams |
| 11 | 100% | 12 teams |
The rows that matter here are the bottom two. Seven wins has never made the playoffs in this sample (0 of 17), and eight wins has made it just 11% of the time (2 of 18). The market's central expectation for this roster lands the Giants exactly there. Even nine wins, a full game and a half above the price, has made the field only 39% of the time (9 of 23). Ten, the near-automatic tier at 94%, requires beating the market's number by two and a half wins.
One caveat, stated plainly because this series does not lean on win totals as proof. A preseason win projection is a weak predictor of playoff qualification: measured on weeks 1-3 closing spreads from 2002 to 2025, it grades out at an AUC of 0.733, and it is flat through the middle of the range, where teams projected at 8, 9 and 10 wins made the field 35%, 43% and 39% of the time, overlapping and non-monotonic. Teams projected around 7 wins made it 28% of the time (confidence interval 20% to 37%), so the 7.5 by itself does not doom anyone. The total is context. The argument is the distance: what the Giants would have to do to the price, not what the price says on its own.
The market prices the Giants at 7.5 wins. Nine wins has made the playoffs 39% of the time, and ten has made it 94%. The story does not have to beat the price by a little. It has to beat it by a lot.
Worked Example: The Two-And-A-Half-Win Problem
Run the arithmetic the cliff sets up. If the Giants land on their price at 7 or 8 wins, history says they miss almost every time: 0 of 17 at seven wins, 2 of 18 at eight. Beat the market by a game and a half to reach nine, and they are still on the wrong side of a coin flip at 39%. Only at ten wins, two and a half clear of the number, does the door swing open at 94%. So the will case is not "this team improves under Harbaugh." Improvement to 8-9 wins would be a strong season by any honest measure and still miss the playoffs most of the time. The will case requires one of the larger price-beating leaps on the 2026 board, and that is the bet this column declines. When game lines post in September, the live NFL odds screen is where to watch the market grade it week by week; the tool surfaces the best available number on every Giants game line across DraftKings, FanDuel, BetMGM and the rest of the major books, and its line-movement view will show whether the market is repricing the leap one Sunday at a time. Our guide to real-time NFL odds covers how to shop those numbers properly.
John Harbaugh Is A Headline, Not An Edge
Now the full version of the number from the top of the page. John Harbaugh is the Giants' new head coach, arriving from the same job with the Ravens, which makes him what the data calls a retread hire, and the retread rows are worth reading carefully. Teams hiring a coach with previous NFL head-coaching experience made the playoffs 32.8% of the time (20 of 61, interval 22.3% to 45.3%), against 42.6% for teams that kept their coach. That gap is selection, not causation: teams that change coaches are usually bad already. The controlled test is wins against the market's implied total, and it is null. Retread hires missed their market number by 0.29 wins on average (t=-1.30) and first-time hires beat theirs by 0.19 (t=1.17), neither statistically significant. The market prices coaching changes correctly, which means Harbaugh's arrival is already inside that 7.5. The hire is a story. It is not an edge, and this series does not pretend otherwise.
His record against the spread, computed from the primary 1999-2025 game log with playoff games included:
| John Harbaugh ATS | Record | Cover rate | Sample |
|---|---|---|---|
| Career | 161-146-10 | 52.4% | 317 games |
| As Underdog | 56-39-5 | 58.9% | 100 games |
| Away | 90-66-7 | 57.7% | 163 games |
| Home | 71-80-3 | 47.0% | 154 games |
| Playoffs | 15-9-0 | 62.5% | 24 games |
| Last 3 Seasons | 30-25-0 | 54.5% | 55 games |
Two corrections to the widely published numbers before anyone uses that table. First, the correction from the top of the page: BetMGM's published 65% does not survive the primary game log, which has Harbaugh at 52.4% on the 317-game sample in the table above. Our figures include playoff games, which most published tables leave out and which is why our game counts run higher on long-tenured coaches, but that explains a counting difference, not a twelve-point one. Be clear about what the correction does not say, though: 52.4% is still above even money on the largest coaching sample on our board. Deflating a myth is not the same as indicting the coach, and this column is not trying to overshoot the other way. Second, a career ATS record describes history and forecasts nothing, and the measured year-one effect for a new coach is null, so none of this is a reason to expect covers. The row I keep coming back to anyway is the underdog line: 58.9% on 100 games (interval 48.9% to 68.3%). A team priced at 7.5 wins will spend most Sundays as an underdog, and Harbaugh's teams have historically been at their best against the number in exactly that role. Hold that thought for the yes case, because it belongs there, and notice what it is not: covering spreads is how a cheap team earns respect, and it pays bettors, but the playoff race is settled in wins, not covers.
Jaxson Dart, Twelve Games In
The quarterback sample is the shortest one on the page. Games with Jaxson Dart at quarterback are 6-6-0 against the spread, 50.0% on 12 games, and the confidence interval on that (25.4% to 74.6%) is the statistical way of saying 12 games admit nearly any conclusion. Ten of the 12 came as an underdog (5-5-0). A quarterback's ATS record also conflates the player with the entire team around him, so it grades the situation, not the passer. The won't case here is not built on Dart being bad; 12 games cannot support that claim any more than they could support a breakout claim. The point cuts the other way: the will case has to assume a leap from a quarterback the data cannot yet vouch for, on a roster the market prices at 7.5. Our call does not need Dart to fail. It just declines to pay for the assumption that he transforms.
The Bracket Has No Room In The NFC East
Here is the argument that turns a lean into a call. These 32 columns have to describe one possible season between them: seven playoff teams per conference, every division sending a winner. Our projected NFC seven is Philadelphia, Dallas, Seattle, the Rams, Chicago, Green Bay and Tampa Bay, and the first two names are the problem, because both of them share a division with the Giants. Our bracket has the Philadelphia Eagles in and the Dallas Cowboys in, Dallas despite coming off a losing season, which takes both the division's automatic spot and a wild card between them, in whichever order the season sorts it. We argue the Washington Commanders miss too, so nothing about our East read rests on New York merely beating out Washington. Callback to the top: about six spots open every year, but a team priced at 7.5 wins and sitting behind two bracket teams in its own division is not positioned to claim one. Every rival argument is laid out in the hub if you want to stress-test the seven, and our free expert picks run every day if you want action long before any of this settles.
The Honest Case For Yes
A column that only argues one side is advocacy, so here is the best of the other one, printed with the same samples. The churn is real: 30.5% is nearly one in three, and six chairs open every season for somebody. The weak-signal finding cuts for New York as much as against it, since teams projected around 7 wins still made the field 28% of the time, and an AUC of 0.733 means the preseason market misses on plenty of individual teams. And the underdog thread from the Harbaugh table is the most flattering fact in the log: his teams covered 58.9% of the time in 100 games as a dog and 54.5% over his last 55 games, which at minimum is not the profile of a coach whose teams go quietly. If Harbaugh's floor-raising is real and Dart's 12 games were the start of something, New York would not be the first team to make the market's September number look silly by January: 28% of the teams priced where they are did exactly that.
Our answer is that "would not be the first team" is doing all the work in that paragraph. The yes case needs a two-and-a-half-win beat of the market, from a team that just won four games, in a division where our bracket already seats two rivals, behind a quarterback with 12 games of evidence, and the one measured effect the hire brings (the new-coach bump) grades out at null. Any one of those objections can be argued away. The won't call is that all of them together will not be.
Where To Trade It, And Who Pays US
Now the strange fact promised at the top of the series, and it doubles as our disclosure. Kalshi carries a market on this exact question: KXNFLPLAYOFF-27-NYG, "Will New York G be one of the 2026-27 Pro Football playoff qualifiers?", closing January 25, 2027. It is the only venue carrying a per-team playoff market, and as of July 30 the book is live: yes trades at 28 cents bid, 29 cents ask. That band is the market's own answer to this page's question, and it leans the same way we do. If you read the case above and think the market and this column are both wrong, that is where the disagreement gets a price; a yes position pays a dollar per contract if the Giants make the field, and the position can lose its full value if they miss. Agreeing with us costs more: buying the no side runs roughly 71 to 72 cents by exchange arithmetic, priced to win the remaining 28 to 29. Those quotes come from our July 30 feed, and quotes move, so check the live board before acting on any number printed here.
The venue that carries this exact market pays us nothing. The venue that pays us does not carry this market. Our primary pointer goes where the market is, not where the commission is.
Full transparency on the incentives, because it is the most credible thing this series has to offer: Kalshi does not pay us anything. We have no affiliate or commercial relationship with them, and they still get the primary mention because they carry the market. In the same spirit, know that Stokastic trades prediction markets and may hold positions in markets it writes about, this one included now that the book is open. Polymarket, which does pay us a commission on sign-ups, does not carry a per-team playoff market at all. What Polymarket US offers is championship futures, a different bet entirely: the Pro Football 2027 Champion market has traded about $42 million across 33 team outcomes, and the 2027 NFC Champion market about $6.3 million across 17, where a Giants conference title trades around two and a half cents on the dollar as of July 30. A conference crown and a playoff berth are different questions, so never treat one market as a substitute for the other. If prediction markets are new territory, our explainer on how prediction markets work covers the mechanics.
If you do want to trade championship futures on the Polymarket US App, code OS4 gets new users the current sign-up offer: Deposit $10, get a $50 trading bonus. If the platform itself is unfamiliar, our guide on how to play on Polymarket walks through the mechanics before you deposit anything. We are a Polymarket partner and may earn a commission if you sign up through our link.
Claim the Polymarket US App offer with code OS4
18+ Only. Restrictions and eligibility requirements apply. Not available in all jurisdictions. Trading is risky. 100% loss can occur. See polymarket.com/tos for more information. The Polymarket US App serves as an independent software provider and affiliate of Polymarket US and Polymarket Clearing, the CFTC-regulated exchange and clearing organization. Event contracts are CFTC-regulated derivatives, not sportsbook wagers, and a position can lose its full value.
More on this: What Will Trump Say? Kalshi Mention Market Odds · New York Governor Odds 2026: Hochul Vs. Blakeman · Bengals Playoff Chances: 3-Seed Call at 10.5 Wins · Washington Commanders Playoff Chances: Just 30.5% · Los Angeles Rams Playoff Chances: Why They Get In
The Record Keeps Score From Here
We have taken the won't side on the record, so here is what would make us wrong, in the order I would watch for it: Dart outgrowing his twelve-game sample, Harbaugh's floor-raising showing up in wins rather than covers, and one of Philadelphia or Dallas falling out of the East picture they currently own. Those three things decide this page, not the preseason number. If they break New York's way, this permanent page will say so in plain words, because it gets rebuilt as the season answers the question, and the record is the point.
