Walk into any branch of a UK high-street bookmaker and ask for the price on the next general election. You will get a single number. Behind it sits a trader who has built in a margin, hedged the book, and quoted you odds that, if you converted them to implied probabilities and added them across every outcome, would total something like 108 or 110 per cent. That extra eight or ten points is the house's cut. It is the price of doing business with them, and it is baked into every line they show you.
Now open Betfair Exchange or Smarkets on the same market. The numbers look different. The total implied probability across all outcomes sits much closer to 100 per cent, sometimes within a single point of it. You are not betting against the house. You are betting against another user who thinks you are wrong, and the platform is taking a small commission on your winnings rather than baking a margin into the odds themselves. The shift in structure produces a measurable, persistent difference in the prices you see.
This is the core difference between prediction markets and traditional bookmakers, and it matters more than most casual bettors realise. The question is not just which gives better odds on any given race or referendum. It is which model of pricing you trust to be honest, and what you are willing to give up to get it.
The house edge, made visible
Ladbrokes does not lose money on elections. Neither does William Hill, nor any of the legacy UK bookmakers. The reason is structural: they set the odds, and they set them to ensure that whatever happens, the book pays out less than it takes in. The technical term is the overround, and on a typical UK political market it runs between five and ten per cent. On a competitive football match it can be three to five. On a novelty market with little volume, it can climb to twenty or more.
The overround is not a fee you see itemised. It is invisible, woven into every price. You think you are getting 2/1 on Labour to win most seats; the true market probability might be closer to 38 per cent, which would be a fair price of around 13/8. The 2/1 you actually got contains the bookmaker's margin already taken. You have lost money the moment you placed the bet, in expected-value terms, before the result is even known.
Prediction markets and betting exchanges remove this layer. The price is set by what other users are willing to pay or accept, with the platform taking commission only on net winnings. The platform does not care which side wins. It cares only that trades happen.
Peer-to-peer pricing, in practice
A concrete example helps. Suppose a market on Smarkets prices the probability of the Conservatives winning the next general election at 18 per cent. The yes side trades at 0.18 and the no side at 0.82. Both numbers add to exactly 1.00, which means there is no overround at all. The exchange charges a commission, typically around two per cent on net winnings, but the price itself is clean.
Compare that to a traditional bookmaker offering 4/1 on the same outcome. Four-to-one converts to an implied probability of 20 per cent. The bookmaker has shaved two points off the true market view, and that is before you factor in the rest of the overround spread across other candidates. On a single bet of fifty pounds, the difference compounds. Over a year of betting, it is the difference between a hobby that breaks even and one that drains your account at a steady rate.
For a deeper walk through how these prices are generated and what they actually mean, our guide to how prediction market odds work covers the mechanics in detail. The short version: a market price is a probability, expressed as a number between zero and one, and the cleaner that number is, the more useful it is.
Where bookmakers still win
None of this means high-street bookmakers are obsolete. They are not, and pretending otherwise misses the trade-offs involved.
First, liquidity. A peer-to-peer market only works if there is someone on the other side of your trade. On a Premier League fixture, Betfair Exchange has more money on it than any bookmaker in the country. On an obscure non-league cup tie, you might find the order book is empty and the spread between the best yes and no prices is wide enough to drive a bus through. A traditional bookmaker will quote you a price on anything, because they are not waiting for a counterparty. They are the counterparty. That convenience has a cost, but for some markets it is the only option.
Second, the experience. Bookmakers offer a clean interface, fixed prices, no commission to calculate, and the ability to place a five-pound accumulator on a Saturday morning without thinking about market depth or matched volume. For casual bettors who want to back a hunch and not study order books, this matters. The exchange model rewards engagement; the bookmaker model rewards none.
Third, promotions. The legacy industry survives on free bets, enhanced odds, and acquisition offers. Exchanges and prediction markets rarely offer the same, because their margins are thinner. If you are a price-sensitive shopper who hunts the best promotional offer each week, the bookmakers are still where you go.
Regulation and what it covers
The UK regulatory picture splits along an interesting line. Betting exchanges like Betfair and Smarkets are licensed by the UK Gambling Commission and treated, broadly, as betting operators. Winnings are not subject to personal income tax for UK residents, which is one of the genuine structural advantages of the UK gambling regime.
Cryptocurrency-based prediction markets like Polymarket sit in a different category. UK regulation restricts how these contracts can be offered to UK residents, and the FCA has taken a generally cautious view of crypto-denominated speculation. The full picture is more nuanced than a single sentence can capture, and the rules continue to shift. Our explainer on whether prediction markets are legal in the UK walks through the current position in more detail.
For most UK users interested in peer-to-peer pricing, the practical answer is the regulated exchanges. They give you the structural advantage of the model with the protection of the licensing regime, and they let you focus on the markets themselves rather than the legal grey areas.
What changes when you trade against other users
The psychological shift is bigger than the price difference. When you place a bet with a bookmaker, you are betting against an institution. The institution is faceless, well-capitalised, and almost certainly more informed than you. The frame is adversarial in a way that feels rigged, because it slightly is.
On an exchange or prediction market, you are trading against other users. Some of them are sharper than you. Some are not. Some are hedging existing positions, some are speculating on news they have read, some are simply wrong. The price you see is the equilibrium of all those views, and your job is to decide whether it is too high or too low. That is a different kind of thinking. It rewards patience, research, and the willingness to walk away when the price already looks fair.
It also means the roles can reverse. On an exchange, laying a selection means offering odds to another user who wants to back it: you take the other side and carry the risk if it comes in. This is a core part of how the peer-to-peer model functions. It is also a fast way to lose money if the liability is misjudged, which is why it is a mechanic to understand fully before using, not a shortcut to returns.
The editorial take
For anyone in the UK who places more than a handful of bets a year, the exchange model is structurally better. The prices are closer to fair, the commission is transparent, and the model treats you as a participant rather than a customer. The cost is engagement: you have to understand what you are doing, which markets have enough liquidity to trade, and how to read an order book. The upside is that you stop paying the invisible tax that the traditional bookmaker model is built around.
For casual bettors who place one Grand National bet a year and want a clean app and an easy interface, the high-street bookmakers remain a reasonable choice. They are not robbing you; they are charging you a margin for convenience. Whether that margin is worth it depends on how much you are betting and how often.
iPredicta tracks prediction markets and betting exchanges across the UK and beyond, publishing analysis on where the prices are sharpest, where the liquidity actually sits, and how the regulated venues compare. For readers weighing the peer-to-peer model against the legacy high-street one, the comparison is rarely as simple as one number against another, and the Smarkets and Polymarket comparison is a useful next step for anyone deciding which platform to actually open an account with.
Betting carries risk and is for over-18s only. If gambling stops being fun, or you are worried about how much you are betting, free and confidential support is available at BeGambleAware.org.
Frequently asked questions
Are the odds really better on prediction markets than at a traditional bookmaker?
Yes, on most liquid markets the prices are measurably better because there is no overround built into the line. A traditional bookmaker bakes a margin of roughly five to ten per cent into UK political and sports markets, while exchanges like Betfair and Smarkets quote prices that add up to close to 100 per cent of implied probability across all outcomes. The platform takes a commission on net winnings instead, usually two to five per cent. The headline difference depends on the market and the volume, but over a year of regular betting the structural pricing difference of the peer-to-peer model is significant. On thinly traded markets the picture flips: bookmakers will quote a price where exchanges cannot match you at all.
What is a betting exchange and how is it different from a bookmaker?
A betting exchange is a platform where users place opposing bets against each other rather than against a house. The exchange matches a back bet (a bet that something will happen) with a lay bet (a bet that it will not), and takes a small commission on the winner's profit. The price is set by supply and demand among users, not by an in-house trader trying to balance a book. The classic UK example is Betfair Exchange, with Smarkets as the main competitor. The model produces tighter spreads and more honest pricing, but it requires enough trading volume on each market for orders to match, which is why exchanges shine on big events and struggle on niche ones.
Do I pay tax on winnings from prediction markets or exchanges in the UK?
Betting and exchange winnings are generally not subject to personal income tax for UK residents, which is one of the structural advantages of the UK regime. This applies to wins from licensed UK operators, including the major betting exchanges. Crypto-denominated prediction markets sit in a more uncertain category, because the assets involved may attract capital gains treatment separate from the bet itself. The practical answer for most UK users is that wins on Betfair or Smarkets do not need to be declared as income, but anyone using offshore or crypto-based platforms should think carefully about both the legal status and the tax position. This is general information, not tax advice, and individual circumstances vary, so check your own position with HMRC or a qualified adviser. Our guide on tax treatment of prediction markets in the UK and US covers this in more depth.
Why do traditional bookmakers still exist if their odds are worse?
Bookmakers survive because they offer guaranteed liquidity, clean interfaces, generous sign-up promotions, and the ability to bet on absolutely anything regardless of whether other users are interested. An exchange cannot match you on a Tuesday-night League Two match if no one else has placed an order; a bookmaker can and will. Most casual UK bettors place small stakes on a handful of events per year, and the convenience of a familiar app outweighs the structural cost of the overround. The exchange model rewards engagement and research; the bookmaker model rewards none. Both serve different customers, and the legacy operators have spent decades building the brands and promotions that keep casual users coming back.
Can I lay bets on a prediction market the way I can on a betting exchange?
On UK betting exchanges like Betfair and Smarkets, yes, laying is a core feature and is what makes the peer-to-peer model work. You offer odds to someone who wants to back a selection, and if they accept, you become the counterparty. On dedicated prediction markets like Kalshi or Polymarket, the mechanic is slightly different but achieves the same outcome: you can buy the no side of a contract, which is functionally equivalent to laying the yes side. Both approaches let you take the bookmaker's role and profit when the outcome does not happen. The risk is asymmetric; a small lay liability can become a much larger loss, which is why understanding stake-to-liability ratios matters before you start.