Let's have an honest conversation about something you might have done in Week 1.
It's Sunday night. You've been watching football since infinity o'clock, the couch has accepted your shape, and there's just one game left to potentially win some coin on: Cowboys at Giants.
So you pull up your phone and start scrolling through your betting apps when something catches your eye: CeeDee Lamb anytime touchdown +120 at BetMGM. Plus money on one of the best wide receivers in the sport against the Giants' secondary? Well, that feels too good to be true, so you do the responsible thing and check another sportsbook. DraftKings, +130. Wow, even better! You click submit on the bet over there, put the phone face-down, and strap in for the show.
Second quarter starts, and it's just like you drew it up — Dak takes the snap, tosses right to CeeDee in open space, and there's never a doubt about it: TOUCHDOWN. Not a highlight by any means, but that's irrelevant. Your ticket cashes before halftime, and you sit back to enjoy the second half with a big grin on your face like the oracle you are. What a hit! Congratulations!
Now let me ruin it for you.
The best return on that exact touchdown was not at DraftKings. In fact, it wasn't at a sportsbook at all. It was on an exchange. Kalshi, to be exact, at +134 — and nobody in your group chat was even looking there.
That's the Week 1 mistake. Not the touchdown you picked, but the counter you chose to buy it at. Heck, you did the hard part. You saw it all play out like the time traveler you are. But then you screwed up the easy part by not getting the best price. So let's not make that mistake again on Thursday, okay?
Or any day, for that matter. To the exchanges we go.
The Week 1 Receipt Nobody Checks
I went back through every anytime touchdown prop from Week 1 — 155 of them across 14 games, every book and every exchange our screen tracks. Here is the part that should annoy you.
| Prop | Worst book | Best book | Kalshi | Gap, in implied probability |
|---|---|---|---|---|
| Amon-Ra St. Brown | -110 (FanDuel) | +115 (BetMGM) | +101 | 5.9 points |
| Puka Nacua | +105 (BetMGM) | +138 (Pinnacle) | +140 | 6.8 points |
| Kyren Williams | -145 (Bet365) | -125 (Pinnacle) | -116 | 3.6 points |
| Justin Jefferson | +120 (BetMGM) | +180 (Bet365) | +173 | 9.7 points |
| Saquon Barkley | -150 (Fanatics) | -125 (FanDuel) | -126 | 4.4 points |
Anytime touchdown prices across eight books and four exchanges, captured from the OddsShopper board on September 9, 2026. Exchange prices are fee-adjusted — what you actually collect after the trading fee — so they compare like for like against a book's number. Gap is the implied-probability difference between the worst and best book price on the same touchdown.
The Nacua and Kyren Williams rows are the part nobody expects: on short-priced guys, not 12-to-1 lottery tickets, the exchange was the best number in the entire market. Kyren Williams is the sharper example of the two, because on a minus-money scorer -116 against the field's -125 is the difference between laying 1.45 and laying 1.16 to win the same dollar. Slate-wide, the exchange beat 59% of individual book prices, was the single best number on 10% of them, and was the worst number on 1%.
So no, it is not a magic window where every price is the best price — if you shop nine books every week and always find the top number, the books won their share last week. Almost nobody is that person. Half of all Week 1 touchdown props had a book-to-book gap north of 5 points of implied probability, and against the book you actually use, three times out of five, the exchange was the better number. Before you count the part where you can also sell it.
Stay ahead of the markets.
Daily insights and expert picks on Kalshi, Polymarket, and what's moving markets.
Free forever. Unsubscribe anytime.
Touchdowns Are The Only Prop Market That Actually Trades
Now the Week 2 half, where the case stops being about shopping and starts being about the market itself.
I pulled the full NFL prop board on the exchange and added up where the money actually is:
| Prop Family | Share of NFL prop volume | Median spread |
|---|---|---|
| Touchdowns Scored | 43.2% | 1¢ |
| First Touchdown | 25.0% | 1¢ |
| Receiving Yards | 12.3% | 8¢ |
| Passing Touchdowns | 7.4% | 5.5¢ |
| Passing Yards | 5.3% | 4¢ |
| Rushing Yards | 3.5% | 5¢ |
| Receptions | 2.3% | 8.5¢ |
| Rushing Attempts | 0.5% | 9¢ |
| Longest Rush | under 0.1% | 31¢ |
Touchdowns and first touchdown are 68% of every prop contract traded. The next family down is receiving yards at 12.3%, and it trades eight cents wide.
That is liquidity concentration doing what it always does, and it is the most useful pattern on this exchange: the crowded market is the honest one. Depth begets tight quotes, tight quotes beget more volume, and the loop runs until the touchdown board looks nothing like the prop counter next to it. On Thursday's board, 53% of touchdown contracts are quoted a single cent wide. Volume-weighted, the whole family sits at 1.16¢.
Meanwhile, longest rush trades 31 cents wide. Buy that one and the market has to move a third of a dollar before you break even on a round trip. Same exchange, same night, same sport. It is Happy Gilmore's putting game versus his drive — one of these is a real skill and one of them is a guy flailing at a market nobody else showed up to.
The Number A Sportsbook Never Prints
Here is the mechanical reason the penny matters, and it is the thing I wish somebody had explained to me years earlier.
On a binary contract, the two sides have to add to a dollar. James Cook is 54 bid, 55 ask on Thursday. Buying Yes costs 55. Buying No costs what's left over the bid, 46. Add them and you get 101 cents — that single penny is the market's whole charge for being in and out.
Then add the exchange's fee, which is the honest way to do this: 1.74 cents on each side at that price, so 56.74 to own Yes and 46.74 to own No. The two implied probabilities add to 103.5%. At a sportsbook you'd call that 3.5% hold and move on — but on an exchange the two halves have different owners. One cent of it is the spread, which belongs to the trader on the other side of you, and you only pay it if you're the one crossing. The rest is Kalshi's fee, published in a formula, charged on each side you trade.
The number to beatboth sides of a penny-wide touchdown contract cost about 103.5% all-in. A standard -110/-110 sportsbook market holds 4.8%, and the margin there is in the price whether you like it or not.
Compare the two honestly. An anytime touchdown at a book is priced nowhere near as kindly as -110, and you can run any pair through our hold calculator to see it.
And when the number goes the other way, below zero, that is not a typo. A negative hold is an arbitrage: both sides available at prices that add to less than 100%, so you can buy the pair and lock a profit, or take a free stance on the side you actually like. Our arbitrage finder was flagging 126 of them above 2% ROI this morning, the fattest at 19.3%, and that one was a touchdown market too.
Now go look at an anytime touchdown at a sportsbook and try to buy both sides. The over is a fair-looking number. The under is where they live — and on the thin ones it is priced like the book would rather you didn't. You cannot easily see the hold, because it isn't printed anywhere; it's built into the pair.
That is the real answer to "are the odds better on Kalshi." Sometimes the top book pays more on the side you want. But you are choosing between a market that charges about a penny to be in and out, and a counter where the round trip is the product.
And you can leave. A touchdown contract has a live bid on it all night, which means a first-quarter injury or a blowout doesn't trap you — our Kalshi trading strategy guide spends its fifth move on exactly that, because it's the habit that takes longest to learn coming from books.
Thursday Night, Read Off The Board
Lions at Bills, 8:15 p.m. ET Thursday, is the tightest prop board of the week. Every quote below is live off the exchange as of Wednesday morning:
| Contract | Bid / ask | Spread | Effective after fee | Contracts traded |
|---|---|---|---|---|
| Jahmyr Gibbs 1+ TD | 68 / 69 | 1¢ | -239 | 12,317 |
| James Cook III 1+ TD | 54 / 55 | 1¢ | -131 | 35,039 |
| Josh Allen 1+ TD | 48 / 49 | 1¢ | -103 | 34,313 |
| Amon-Ra St. Brown 1+ TD | 41 / 42 | 1¢ | +129 | 19,877 |
| DJ Moore 1+ TD | 34 / 35 | 1¢ | +173 | 7,995 |
| Dalton Kincaid 1+ TD | 31 / 32 | 1¢ | +198 | 10,230 |
| Jameson Williams 1+ TD | 28 / 29 | 1¢ | +228 | 16,502 |
| Khalil Shakir 1+ TD | 27 / 28 | 1¢ | +240 | 2,528 |
| Sam LaPorta 1+ TD | 28 / 29 | 1¢ | +228 | 31,508 |
| Isaac TeSlaa 1+ TD | 14 / 15 | 1¢ | +529 | 56,577 |
| Jared Goff 1+ TD | 4 / 5 | 1¢ | +1,773 | 77,956 |
Effective odds are the ask plus the taker fee, converted to American, because a cent price with the rake still in it is not a number you can bet. All 11 of those contracts are quoted a single cent wide, and the Jared Goff rushing-touchdown market has traded 77,956 contracts — more than any other prop on this board, which tells you something about the people trading this thing at 1 a.m.
Two rows are bolded because on Wednesday the exchange is the best price in the entire market on both, books included.
The Plays
Amon-Ra St. Brown, 1+ touchdown, 42¢, effective +129. The most targeted receiver in Detroit, in a game the total expects points from. He is in the Week 1 table above too: the exchange had him at +101 while FanDuel wanted -110 for the same touchdown, and he went out and caught both of Goff's scores. Second straight week the board prices him better than most of the market. The liquidity is the part I like as much as the price: nearly 20,000 contracts traded, more than 18,000 still open. That is a market you can size into and, more to the point, get back out of.
Dalton Kincaid, 1+ touchdown, 32¢, effective +198. Buffalo's tight end has an end-zone role that shows up most in exactly this kind of short-week, keep-it-simple game plan.
The fine print, and there is real fine print here. Detroit opened the week with three of five starting offensive linemen on the injury report and plans walkthrough-only practices before a short-week Thursday game, which is not the setup you'd draw up for a clean red-zone script. St. Brown is also splitting a Detroit end-zone pie with Gibbs, who is 69 cents for a reason. If the Lions' scoring drives all end in Gibbs plunges, the St. Brown ticket watches — which is exactly why these are two separate tickets and never a parlay.
Post Your Own Number Instead Of Paying Theirs
One more thing the exchange gives you that a book structurally cannot, and almost nobody uses in their first month.
You do not have to take 42. You can post a bid at 41 and wait.
On most sports markets a resting order that fills pays no trading fee at all. So the same touchdown, bought by being patient instead of impatient, is 41 cents flat — +144 instead of +129. Fifteen points of American odds, for the crime of not mashing the buy button.
Now run it on Kincaid, the other side of the same board:
| Take the ask | Join the bid | |
|---|---|---|
| Price | 32¢ + 1.53¢ fee | 31¢, no maker fee |
| Effective | +198 | +223 |
Twenty-five points of American odds sitting in the gap between impatient and patient, on a contract nobody is racing you to. And on a board where a bid can walk a cent or three in a single morning, joining it is not a heroic act of patience — it is standing where the buyers already are.
That is the trade a sportsbook customer structurally cannot make. There is no window at a book where you slide your own number across the counter and wait for someone to take it.
The trade-off is real: a resting bid can sit there all night and never fill, and a fill you never got pays nothing. My rule hasn't changed. Close to kickoff in a liquid market, taking a fair ask is fine. Any time the spread is wider than a penny or two, or you're hours from kick, be the one posting the price. On a board where 53% of the contracts are already a penny wide, you are rarely giving up much by asking.
That is the part of this that turns a bettor into a trader, and there is no version of it at a sportsbook. You cannot walk up to a counter and offer to sell them a touchdown at your number. Here you can just... do it.
Two separate tickets. Never a parlay. One quiet Detroit night should not take the Buffalo ticket with it.
More on this: One Score Is Priced. The Second One's Hiding In Plain Sight. · 49ers Vs Rams Touchdown Odds: The Crowd Bought A Penny And Missed Corum · The Longest Shot On This Board Might Be The Best-Kept Secret · Patriots Vs Seahawks First TD Odds: Kalshi's Mystery Buyer · Five Anytime Tickets and Two First TDs for Monday Night
Final Thoughts
The Week 1 lesson wasn't that a touchdown is always cheaper on an exchange. The lesson is that St. Brown's touchdown paid $90.91 at one app and $115 at another, that the exchange paid $101 on a board charging a penny for the round trip, and that all three of those numbers were sitting there in the same minute while everybody argued about who was going to score.
Two things before you go, same as always.
Every longshot ticket is more likely to miss than hit. A 14-cent contract is the market telling you it misses 86% of the time, and it does not care that you like the guy. Say the number out loud before you buy, not after.
And size it like a prop, because it is one. Quarter-unit stakes, no chasing a price that already ran from you, no hero sizing because a board looks soft on a Wednesday. The penny you save on the spread means nothing if the stake behind it was three times too big.
Kalshi prices were read from the exchange on Wednesday, September 16, 2026, and will move before Thursday's 8:15 p.m. ET kickoff; confirm the bid and ask before trading. Sportsbook prices for Week 1 are from OddsShopper's board, captured September 9, 2026. Volume and spread figures cover the full NFL prop board on the exchange, pulled September 15 and 16, 2026. Fees are Kalshi's published taker rate of 0.07 x price x (1 - price), rounded up. The edges here are model estimates, not predictions of fact; nothing here is financial advice. If you or someone you know has a gambling problem, call 1-800-GAMBLER.
