Every time you place a bet at a sportsbook, you lose before the ball is even in the air.
Not the bet itself - the vig. The juice. The hold. Whatever your book calls the cut it skims off the top of every line. Standard NFL juice is -110 both sides, which means you risk 110 to win 100 on a coin flip. The book is not betting against you. The book is taking a tax on the fact that you and your buddy disagree, and it collects that tax whether you win or lose. Over a season it is the only thing on the screen that is guaranteed to go up.
So here is a question almost nobody in the golf-betting space asks out loud: what if there were no house?
The bet you can actually beat
Walk it through. A sportsbook offers you Scottie Scheffler to win the U.S. Open. Behind that number is a pricing team, a stat model, and a liability desk whose entire job is to make sure the line is sharper than you are. You are betting into a market that has been beaten into shape by people who do this for a living. You can win a weekend. You cannot win a year.
Now picture the other bet. Your buddy Dave is on the first tee. He switched putters two weeks ago and it has not gone well, his back is barking, and he opens with a triple on hole one roughly half the time, which wrecks his whole round when it happens. Does Dave break 90 today?
That is a bet you can actually win. Not because you are smart - because you have information the sportsbook will never have. You watched Dave miss four three-footers last Saturday. You know the new putter is a problem. The entire edge in golf betting is information asymmetry, and you have more information about your foursome than any book on earth ever will.
The book on Scheffler is the bet you slowly lose. The market on Dave is the one you were born to win. So that is the one we built a market around.
How a pool with no house pays out
Here is the mechanic that makes it work, and it is older than sportsbooks: it is pari-mutuel. Same structure as the betting pools at a horse track, minus the track's cut.
Everybody who wants in on a question puts their stake into one pot. When the event resolves, the pot is split among the people who called it correctly, in proportion to what they put in. There is no bookmaker setting a price and pocketing the spread. The "odds" are just a reflection of how the crowd bet - if everyone piles onto the same side, the payout for being right shrinks, because you are splitting the pot with everyone else who agreed with you. Take the unpopular side and turn out to be right, and you clean up.
The key number is the one that is missing: the rake. In a no-house pool, every unit staked is a unit paid out. On caddie.fun a Market is exactly that pool: the pot is your crew's XP, the math splits it among the people who got it right, and we do not take a cut. A four-person slate where everyone antes the same amount pays the winner the entire pot, not the pot minus a tax.
That is the difference, concretely. Ante 20 XP into a friend-only no-vig slate of four and the winner takes the full 80 - the only XP that ever stays in the pot is rounding-down on an uneven split, and even that goes back to the players, never to us. Run the equivalent through a traditional book and the structure quietly bleeds it - the hold is baked into every line so the sum of what you can win is always less than the sum of what everyone risked. One of these is a game. The other is a slow withdrawal.
The round is the referee
There is a second problem every prediction-market operator on earth fights, and it has nothing to do with money. It is settlement. Who decides what happened? Books pay teams of people to resolve disputes, buy data feeds, and adjudicate the edge cases when a stat gets corrected after the fact. That machinery exists because the operator does not own the truth - it has to go buy it.
A friend-golf market does not have that problem, because the round is the oracle. The scorecard your group was going to fill out anyway is the source of truth. Did Dave break 90? The card says 91. Settled. No feed, no dispute desk, no "pending official review." Every market resolves the instant the scorecard finalizes - usually before you can refresh the screen. We own the scorecard, the live scoring, and the round itself, so resolution costs nothing and waits on no one.
To keep it honest, every market shows a one-line "how it settles" note up front. You always know exactly where the answer is going to come from before you put anything in. And if a market does not draw at least three picks before it locks, it auto-voids and refunds everyone - no dead pools, no hollow "I won twelve XP" non-events.
Why a sportsbook literally cannot ship this
This is not a thing DraftKings is choosing not to do. It is a thing they structurally cannot do.
A book makes money on the vig. Remove the vig and you have removed the business model - there is no version of a sportsbook that takes zero cut, because the cut is the entire reason it exists. We are not a book. The product is the round, and the market is the spice that makes a Saturday game feel like it has stakes even when you played it for fun. We do not need a hold, so we do not take one.
There is a regulatory wrinkle too. State-by-state sportsbook law is why DraftKings is dark in California and Texas. Pools resolved entirely by your own performance, friend-gated, with zero house edge, are a different animal - skill-based contests among people who know each other, not a faceless book taking action on strangers. That is a meaningfully lighter footprint, and it is a direct consequence of the no-house structure, not a loophole bolted on after the fact.
Where the money part actually stands
Straight answer, because the alternative is the kind of vague hype that should make you close the tab: Markets are XP, and that is the design, not a placeholder. The same XP that powers your tier and your level. It is free to ante, the payouts are real bragging rights, and nobody is risking rent money on whether Dave three-putts the 17th. The no-house, no-vig structure in this whole post is the Markets pool - that is the surface that takes zero cut, and the reason it can take zero cut is precisely that there is no money to skim.
That is not an accident we are hoping to grow out of. A real-money version of this would need exactly the thing we spent this whole post arguing against: a house. Somebody has to seed thin pools so a four-person slate has real depth, somebody has to custody the funds and backstop the settlement, and the moment somebody does that, they are the market maker - and they will, eventually, want paying. The cleanest way to keep Markets no-house is to keep them XP. So that is where they live.
If you want to play your golf for real money, that already exists on caddie.fun - it is just a different product: round wagers, side bets, Nassau, format payouts you settle in cfUSD. Those carry a small, flat, disclosed platform fee to keep the lights on. Not a sportsbook vig hidden inside a line - a posted cut on a settled pot. But it is a cut, and we are not going to call something no-house that isn't. The Market is the no-house game. The wager is the real-money game. Two different things, on purpose.
The Cut
The whole golf-betting category is built on the same quiet assumption: there has to be a house, and the house has to get paid. Sportsbooks built an entire industry on the vig and a fight over who owns the truth.
A foursome does not need either one. You already know more about Dave's putter than any pricing desk ever will. The scorecard already knows who won. All that was missing was a market that did not skim a tax off the top for the privilege of standing between you and your buddy.
So we took the house out, took the vig out, and pointed the whole thing at the one game you actually have an edge in: the people you tee it up with every Saturday.