- Verdict
- Excellent
- Coins in the cashier
- 15
- Payout ceiling
- 10,000 EUR per day clause number not recorded
- ID can be requested
- case by case
Value betting explained: when a price pays more than the chance deserves
Guide · Strategy
A value bet is a price that pays more than the outcome's real chance justifies. This guide shows how to put a number on that chance, how to tell from a record of many bets whether the edge is real, and why a sound edge can still sit underwater for hundreds of slips.
Written by: Callum Reddaway·Checked by: Joana Teixeira·Updated
Value betting is the habit of staking only when the odds on offer pay more than the outcome's true chance deserves. It is the one approach to sports betting that holds up on paper, and it is also the one that tests patience hardest, because a correct price judgement still loses a large share of individual slips.
8 sportsbooks, those with provably fair games first
- Verdict
- Excellent
- Coins in the cashier
- 9+
- Payout ceiling
- not published
- ID can be requested
- case by case
- Verdict
- Excellent
- Coins in the cashier
- 12
- Payout ceiling
- US$100,000 per week cl. 9.6
- ID can be requested
- case by case
- Verdict
- Excellent
- Coins in the cashier
- 9+
- Payout ceiling
- not published
- ID can be requested
- case by case
- Verdict
- Good
- Coins in the cashier
- in cashier
- Payout ceiling
- no limit
- ID can be requested
- case by case
- Verdict
- Good
- Coins in the cashier
- 10
- Payout ceiling
- not read
- ID can be requested
- case by case
- Verdict
- Excellent
- Coins in the cashier
- in cashier
- Payout ceiling
- 1,000,000 USDT per week cl. 6.10
- ID can be requested
- 2,500 EUR
- Verdict
- Excellent
- Coins in the cashier
- 13
- Payout ceiling
- from 5,000 USD in instalments cl. 11.5
- ID can be requested
- US$2,500
What turns a price into a value price
Every decimal price carries a promise: for each unit staked, this many units come back on a win. Whether that promise is generous depends on one number you never see printed, the real probability of the outcome. Put the two together and you get expected value:
Expected value per unit staked = probability × decimal odds − 1.
Suppose you rate a tennis player at exactly 50 per cent to win. The fair price for that chance is 1 ÷ 0.50 = 2.00. Anything above 2.00 is a value price; anything below it is a price where the bookmaker is collecting from you on average. The prices below are an example:
| Your estimate | Price offered (example) | Expected value per unit | Verdict |
|---|---|---|---|
| 50% | 1.90 | −0.05 (−5%) | Below fair, skip |
| 50% | 2.00 | 0.00 | Fair, no edge |
| 50% | 2.10 | +0.05 (+5%) | Value |
| 50% | 2.20 | +0.10 (+10%) | Value |
The arithmetic is trivial. Everything difficult sits in the first column. If your 50 per cent is really 45 per cent, the 2.20 price is worth 0.45 × 2.20 − 1 = −0.01, a small loss, and the whole plan rests on an estimate that was wrong by five points.
Estimating the true chance
There are three honest ways to arrive at a probability, and serious bettors usually combine them.
- A model of your own. Team ratings, expected goals, serve and return numbers, pace figures in racing. A model gives a number you can test, but it needs a long record before you can trust it over the market.
- The market with the margin taken out. A sharp, high-volume price is a strong estimate once the bookmaker's cut is removed. On a two-way market at 1.90 and 1.90, each side implies 52.63 per cent, the pair adds to 105.26 per cent, and scaling back to 100 gives 50 per cent each, a fair price of 2.00. If another board then offers 2.08 on one side, that is 0.50 × 2.08 − 1 = +4 per cent on paper. The guide to the vig sets out the margin-removal step in detail.
- Information the price has not absorbed yet. A confirmed line-up, a change of goalkeeper, a weather shift on a golf course. This edge is real but short-lived: prices catch up within minutes.
Comparing boards is the cheapest of the three, which is why the page on crypto betting sites with the best odds starts from the same overround sum. A book that runs a thinner margin hands you a better starting point before any judgement is involved.
Pre-matchFootball, cricket, tennis and racing, priced before the startFund the account in USDT, BTC or ETH and build a single or an accumulator on this week's fixtures.Bet nowIn playMarkets that move with the match, and cash-out where offeredPrices update as the game runs; the same balance covers the casino side of the account.Bet live
Measuring the edge over many bets
Winning or losing a month tells you very little. The better measuring stick is the closing price, the last price before the event starts, when the market has absorbed the most information. If you regularly take prices that are longer than the fair closing price, your estimates are probably beating the market.
An example: you take 2.20 on a selection. By kick-off the market has moved, and with the margin removed the fair closing price is 2.05, an implied chance of 48.8 per cent. Your bet was worth 2.20 ÷ 2.05 − 1 = +7.3 per cent against the close, even though the slip itself may lose. Repeat that check across every bet and you get an average edge that settles down far faster than profit does.
A useful record holds, for each slip: the sport and market, the price taken, the fair closing price, the stake, the coin, the result, and the profit in coin. A spreadsheet is enough. The betting strategy guide shows a fuller layout and how often to review it.
Variance: why a real edge still loses for long stretches
Value betting is a long game because results scatter widely around the average. Take a bettor with a genuine 5 per cent edge, staking one unit each time at even money (a true chance of 52.5 per cent at a price of 2.00). Using the normal approximation, this is how the picture looks:
| Number of bets | Expected profit | Typical swing (one standard deviation) | Chance of being behind |
|---|---|---|---|
| 100 | +5 units | ±10 units | about 31% |
| 500 | +25 units | ±22 units | about 13% |
| 2,000 | +100 units | ±45 units | about 1% |
Longer prices widen the swing. The same 5 per cent edge at odds of 5.00 (a true chance of 21 per cent) has a per-bet standard deviation of about 2 units instead of 1, and after 500 bets the chance of being behind is close to 29 per cent. Outright markets in golf are the extreme case; the crypto golf betting page explains why a long-odds winner can also run into a payout ceiling.
The practical consequence is that a losing run of 30 or 40 slips is not evidence that the method has stopped working. Equally, a winning month is not evidence that it works. Only the closing-price record and a large sample separate the two.
Sizing the stake on a value bet
The textbook answer is the Kelly fraction: stake = (b × p − q) ÷ b, where b is the decimal price minus 1, p your probability and q = 1 − p. For the 2.20 price on a 50 per cent chance, b = 1.2, so (1.2 × 0.5 − 0.5) ÷ 1.2 = 0.083, or 8.3 per cent of the bankroll.
Full Kelly assumes your probability is exact. It never is, and an overestimate of only a few points turns the Kelly stake into an overbet. Most disciplined bettors use a quarter or less of the Kelly figure, here about 2 per cent, or simply a flat 1 per cent of the bankroll on every value bet. Flat stakes give up a little growth in exchange for far smaller drawdowns.
SportsbookDeposit in USDT, BTC or ETH and bet the same dayPlace a bet18+ · the operator's terms apply to every betThe limits a value bettor runs into
A bookmaker earns its margin from customers whose estimates are worse than its own. An account that keeps beating the closing price does the opposite, and operators respond by trimming the maximum stake that account may place, sometimes to a token amount. Read the operator's terms for what it reserves the right to do with stakes and accounts. The page on bookmakers that take crypto reads those settlement clauses operator by operator.
This is the uncomfortable truth of value betting: the better you are at it, the less a single sportsbook wants your money. Spreading activity across several of the crypto betting sites on the main board is the usual response.
The coin layer on top of the edge
An edge is earned in the unit of the balance. A 5 per cent advantage in bitcoin is 5 per cent more bitcoin than you started with on average, but if bitcoin falls 8 per cent before you withdraw, the result in pounds is a loss. A dollar stablecoin keeps the edge close to a dollar figure, which is why the betting sites that take USDC page is worth reading if value is the plan.
Network fees act like an extra margin. An example: a 1 USDT fee on a 50 USDT deposit costs 2 per cent before the first bet, which wipes out much of a 5 per cent edge on the first few slips. Fewer, larger transfers keep that cost small.
What is value betting in one sentence?
Value betting is backing an outcome only when the decimal odds, multiplied by your estimate of its probability, come to more than 1. A price of 2.20 on an outcome you rate at 50 per cent gives 1.10, an expected return of 10 per cent of the stake.
How many bets do I need before I know my edge is real?
More than most people expect. With a 5 per cent edge at even-money prices, roughly one bettor in eight is still behind after 500 bets of equal size. A few hundred results say little; the price you took against the closing price says more, and sooner.
Can value bets lose?
Most individual value bets at prices around 2.00 lose almost half the time, and long-odds value bets lose most of the time. Value describes the average over many slips, not the outcome of the next one.
Does it matter which coin I bet with?
The edge is earned in the coin of the balance. In a stablecoin it stays roughly a dollar figure; in bitcoin or ether the coin's own price movement can add or remove more than the edge itself between the bet and the withdrawal.
Why do bookmakers limit accounts that find value?
A bookmaker sets prices to earn its margin. An account that keeps beating the closing price costs it money on average, so the operator can reduce the maximum stake it accepts. The terms of each sportsbook describe what it reserves the right to do.