Alternative methods
+EV betting
Bet only when a price pays more than the bet is worth. Here is what expected value is, where the fair price comes from, how the +EV Finder works, how much to bet, and why you judge it over hundreds of bets, never one.
What expected value is
Every bet has a chance of winning and a price. Expected value, or EV, is what the bet makes or loses on average if you could place it many times. Take a true coin flip at +110: half the time you win $110, half the time you lose $100. On average that is 50% × $110 − 50% × $100 = +$5 per $100, or +5%.
At the usual -110 the same coin flip is worth about -$4.55 per $100. That is the normal case: the book's margin makes most bets slightly negative. A +EV bet is one of the exceptions, a price better than the real chance behind it. You still lose plenty of them. The profit shows up across many bets, not on any single one.
Sign-up offers are +EV because the book hands you value. +EV betting looks for it in the prices themselves, with no promo needed. It is also riskier: there is no hedge, so every bet wins or loses in full.
Fair odds: take the margin out
To know whether a price is good, you need the fair price: the chance of each side with the book's margin removed. Removing it is called devigging. Say a sharp book prices a game at -150 and +130:
| Side | Price | Implied chance | Fair chance | Fair odds |
|---|---|---|---|---|
| Favorite | -150 | 60.00% | 57.98% | -138 |
| Underdog | +130 | 43.48% | 42.02% | +138 |
The two sides add up to 103.48%, so the simplest method divides each by 1.0348. Other methods split the margin differently, because books tend to shave more off underdogs than favorites. The power method puts the underdog at 41.60%, fair odds of +140. The +EV Finder starts on “Worst Case”, which takes whichever method is least generous to your bet, so a borderline edge gets filtered out instead of flattered.
Why a sharp book sets the fair price
Most US books limit customers who win and follow the wider market with their prices. A sharp book works the other way: it takes big bets from winning bettors and moves its line as soon as they bet. Pinnacle is the best known, and its price is widely treated as the best estimate of the real chances. Exchanges and prediction markets like Novig, Kalshi and Polymarket get close for a similar reason: customers bet against each other, and mispricing gets bet away.
Pinnacle does not take US customers. BreadSync only reads its prices as the reference; you bet at your own books. The finder uses Pinnacle first, then Novig, Kalshi and Polymarket in that order when Pinnacle has no line.
Now say one of your books offers the underdog at +150 while the fair price is +140. Per $100: 41.60% × $150 − 58.40% × $100 = $62.40 − $58.40 = +$4.00, a 4.0% edge. That gap, between a book slow to move and the sharp price, is what the finder looks for.
Use the +EV Finder
Do it in BreadSync
+EV Finder
Compares every book's price with a fair price taken from sharp books, ranks the bets priced better than fair, and sizes a stake from your bankroll.
- 1.Open "Books", click "Deselect All", then tick only the books where you have an account and money. It starts with every book on, including ones you cannot use.
- 2.In "Markets", start with Moneyline, Spread and Total under "Mainline". Props, alt lines and periods are there too, but their fair prices are rougher.
- 3.Open Settings (the sliders button at the right). Leave "Fair-value anchor" on Pinnacle and "Devig" on "Worst Case". "Min EV" starts at 1%; 2% keeps only the clearer edges.
- 4.Pick a bet in the queue and read the middle panel before anything else (below).
- 5.Under "Stake it", set "Bankroll" and pick ¼. Check the price is still up at your book, place the bet, then click "Track this bet".
Each queue row leads with its EV. Green is a normal edge. Amber means the reference market itself is wide, so the fair price is shaky and the edge may be noise. An amber “$8 available” under the bet means an exchange has almost nothing at that price to take.
Read a bet before you take it
- Offered is the price at your book, what it implies, and how long ago it last moved. Fair value is the devigged price and its “true win chance”.
- Trust is how wide the sharp market is, in cents: -150 / +130 is 20¢ wide. On main lines up to 12¢ reads “tight” and up to 25¢ “normal”. Wider means the sharp book is not confident, so neither should you be. Props are judged on a wider scale.
- Check the line appears when your book's price last moved more than 20 minutes before the sharp price did. That is the most common false edge: a book that has not caught up yet. Confirm the price is still there before you bet.
- Sharp money · exchange depth shows how much each exchange, and Pinnacle, will take on each side. The side they will sell you least of is the side they rate higher, and the headline says which way they lean and how many agree. Use it as a tiebreak on a close call, not as part of the edge.
- Compare odds lists every book's price for the market. The columns with a dot or a percentage under the logo are the books that set the fair price.
How much to bet: Kelly, and why a fraction
The Kelly criterion sizes a bet from its edge: the edge divided by the net odds. For the +150 bet above, the edge is 4.0% and +150 pays 1.5 times the stake, so full Kelly is 4.0 ÷ 1.5 = 2.67% of your bankroll. The panel shows it as “Full Kelly”, and the buttons under it scale it down.
| Setting | Share of bankroll | On a $5,000 bankroll |
|---|---|---|
| Full | 2.67% | $133 |
| ½ | 1.33% | $67 |
| ¼ (the default) | 0.67% | $33 |
| ⅛ | 0.33% | $17 |
At ¼ Kelly the expected profit on that bet is about $1.34. That is what +EV betting looks like bet by bet: small numbers that only add up over volume.
Why not bet full Kelly? Two reasons:
- The edge is an estimate. If the real edge is half what the screen shows, full Kelly is betting twice what it should, and overbetting a small edge can shrink your bankroll even when the edge is real.
- Full Kelly swings hard. Deep drops are normal on the way to long-run growth. Half Kelly keeps most of the growth with far smaller swings, and a quarter is gentler still.
Your bankroll here is money set aside for these bets that you can afford to lose, not everything you hold at the books.
Watch out: never raise your stakes to win back a losing week. The Kelly amount already assumes you will have losing weeks.
Variance: you can lose for weeks with a real edge
A real edge does not mean steady profit. Here is a 3% edge, betting $100 a time near even money, and the chance you are still down after each number of bets:
| Bets placed | Expected profit | Chance you are still behind |
|---|---|---|
| 100 | +$300 | about 38% |
| 250 | +$750 | about 32% |
| 500 | +$1,500 | about 25% |
| 1,000 | +$3,000 | about 17% |
| 3,000 | +$9,000 | about 5% |
At twenty bets a week, 250 bets is three months, and about one person in three with exactly this edge is still behind. Bets at longer odds swing even more. So a bad month tells you very little, and neither does a good one. That is why the next section matters.
Closing line value: the honest scorecard
The closing line is the fair price just before the game starts, when the market has the most information. If the prices you take are routinely better than the close, your edge is probably real, long before your profit can tell you. That comparison is closing line value, or CLV.
| You took | Fair price at the close | Your CLV |
|---|---|---|
| +150 | +130 (43.48%) | 43.48% × 2.50 = 1.087, about +8.7% |
| +150 | +160 (38.46%) | 38.46% × 2.50 = 0.962, about -3.8% |
In the first row the market moved your way: by kickoff the sharp price agreed your side was worth more. In the second it moved against you, and that bet was a mistake whether it won or lost.
- 1“Track this bet” saves the edge, the fair price and the method in the bet's notes in My Bets.
- 2Just before the game starts, look up the fair price for the same bet again in the finder's “Compare odds” or on the Odds Screen, and add it to the note. BreadSync does not record the closing price for you yet.
- 3After a hundred bets or so, count how many beat the close. Mostly yes: keep going. Mostly no: the edges were noise or stale lines.
Limits come sooner
Books watch for customers who beat the closing line, and +EV bettors are exactly that. Expect limits sooner than with promo play, sometimes within weeks, and sooner still on small markets like player props. A limit also tends to end the promos at that book.
- A book that still sends you good promotions is usually worth more to you than a few weeks of +EV bets there.
- Skip prices that look far too good. A book can cancel a bet on an obvious error.
- When a limit comes, it is not a penalty and your balance is still yours. If a book limits you covers what to do.
Without Pro
Do it in BreadSync
Expected Value Calculator
Found a price at your book and both sides of the same market at a sharp book? Devig them with the No-Vig Calculator, then this gives the edge.
The free No-Vig Calculator devigs a market, and the Kelly Criterion Calculator sizes the bet. Doing it by hand is slow, which is the part the finder does for you across every game.