The Guardian Line
The Guardian Line is the de-vigged median price across every sportsbook we track— the betting market’s own estimate of what is actually going to happen, with the bookmaker’s margin removed. No single book is gospel; a dozen books, stripped of their cut and reduced to their middle value, are hard to beat.
Every fair-price product in this market is proprietary. Unabated sells the Unabated Line, Betstamp sells True Line, and neither shows its working. The Guardian Line is the one you can check: the whole method is below, with a worked example you can reproduce on a calculator.
How it is built
Four steps, in order. Nothing in here is proprietary and nothing is withheld.
- Convert every book’s prices to implied probability. A price of −150 implies 150 ÷ 250 = 60%; +130 implies 100 ÷ 230 = 43.5%.
- Remove each book’s own margin.Those two add to 103.5%, and the 3.5% excess is the book’s cut. Divide each probability by that book’s total and what remains sums to exactly 100% — that book’s honest opinion, with its fee taken off.
- Take the median across books, per outcome. Not the mean. One book with a stale number should not move the answer.
- Convert back to odds. That is the Guardian Line.
On a three-way market — soccer’s home, draw and away — the draw is included at every step. Leaving it out normalises over two outcomes instead of three and inflates both remaining probabilities by roughly half, manufacturing enormous edges that do not exist. See de-vigging and market consensus.
A worked example
Three sportsbooks price the same game. The prices disagree, and none of them is the fair price, because all three include a fee.
| Book | Home | Away | Implied total | Margin |
|---|---|---|---|---|
| A | -150 | +130 | 103.48% | 3.48% |
| B | -145 | +125 | 103.63% | 3.63% |
| C | -155 | +135 | 103.34% | 3.34% |
De-vig each book — divide its probabilities by its own total — and the three honest opinions on the home side come out at 57.11%, 57.98% and 58.82%. The median is 57.98%.
The Guardian Line
Home -138 · Away +138
57.98% / 42.02%. The market thinks the home side wins about 58 times in 100.
Now look back at what was actually on offer. The friendliest home price was -145 and the friendliest away price +135, and both are worse than the fair number. That gap is the vig, and it is there on every screen in the market. Shopping does not remove it, it shrinks it — from 3.48% at a typical book to 1.74% taking the best of these three, which is the same halving we measured across 315 games.
Why the median, and not the average
This is the one methodological choice worth defending, because it is not obvious and it matters more than it looks.
While collecting European soccer prices we measured both. In the most liquid competitions the two agreed almost exactly — Premier League read 5.78% mean against 6.03% median. In the least liquid, they came apart completely: MLS read 23.3% mean against 8.5% median, entirely because a handful of books had stopped maintaining prices and were quoting numbers nobody could bet into.
An average treats a dead quote as an opinion. A median treats it as one vote among many. The divergence between the two is itself a useful signal — when they separate, the market is thin — which is why we record both and publish the median.
What it is not
The Guardian Line is not a prediction, not a pick, and not advice. It is a measurement of what the market currently believes, stated without the fee that normally obscures it.
It can be wrong. Consensus is a good estimator, not a true probability — it is the number to beat, not the answer. A price better than the Guardian Line has positive expected value against the market’s own view, which is the honest definition of an edge and also a long way from a guarantee. And a consensus built from books that share information is not independent evidence; when everyone copies one sharp book, the median is that book.
Limitations, stated plainly
- Our book set is not the whole market. Consensus means consensus across the regulated North-American sportsbooks we track, which excludes offshore books and exchanges.
- Exchanges are deliberately excluded. An exchange quotes before commission and takes its cut from winnings, so its prices look tighter than they are. Mixing them with sportsbook prices measures nothing.
- It moves. The Guardian Line at noon is not the Guardian Line at kickoff. The closing number is the sharpest one, which is why we grade against it — see closing line value.
- Thin markets are noisier. Fewer books means a less reliable median. We require at least three before calling anything a consensus.
Cite the Guardian Line
Free to cite and quote with attribution and a link. No permission needed.
Bankroll Guardian (2026). The Guardian Line: an open, checkable fair price. https://www.bankrollguardian.com/guardian-line
Questions about the method, or want a figure checked before you publish it? Email support@bankrollguardian.com. Related: full methodology · what the vig costs, by sport · no-vig calculator
Common questions
- What is the Guardian Line?
- The Guardian Line is a fair price: the median de-vigged probability across every sportsbook we track, converted back into odds. It answers one question — given everything the betting market currently believes, what are the real chances, with the bookmaker's cut taken out? It is not a prediction and not a pick.
- How is the Guardian Line calculated?
- Four steps. Convert each sportsbook's prices to implied probabilities; divide each book's prices by its own total to remove that book's margin, leaving probabilities that sum to 100%; take the median across all books for each outcome; convert back to odds. The median rather than the mean, because one stale book should not move the number.
- Why the median instead of the average?
- Because a single stale or outlier quote can drag an average a long way and barely moves a median. We measured this while collecting soccer prices: in the least liquid league the mean margin read 23.3% against a median of 8.5%, entirely because a handful of books had stopped maintaining their prices. The median is the robust choice, and robustness matters more than elegance when one bad quote can invent an edge that never existed.
- How is this different from Unabated's line or Betstamp's True Line?
- Mostly in that you can check ours. Both are proprietary fair-price estimates and neither publishes its method, so you are asked to trust the output. The Guardian Line's method is on this page in full, with a worked example you can reproduce with a calculator. We would rather be checkable than clever.
- Can I use the Guardian Line in my own work?
- Yes. It is free to cite and quote with attribution and a link. If you are writing something that depends on it and want to check a figure or ask about the method, email support@bankrollguardian.com.
- Does a price better than the Guardian Line mean I should bet it?
- It means the price is better than the market's own consensus, which is where positive expected value comes from — but it is not advice and it is not a guarantee. The consensus can be wrong, your book can be right, and a single bet at a good price still loses most of the time when the probability is low. Nothing here tells you what to bet.
See the Guardian Line on every game
Bankroll Guardian prices every market against it, tracks your bets, and shows whether you are beating the closing number — free to start, no card required.