How Far Pinnacle Prices Move Before Kickoff, and What It Proves

What one published study of Pinnacle odds movement measured

Joseph Buchdahl, an analyst who runs football-data.co.uk, published the study in 2018 on Pinnacle's own betting resources site. The host matters: an operator hosts it, and it treats that operator's closing price as the benchmark, so it reads as the operator's own material, without an independent audit.

For every home and away team, Buchdahl divided the opening price by the closing price after removing Pinnacle's margin (the ratio sits above 1.00 when a price shortened, so 1.10 means roughly a tenth off the opening price). The ratios ranged from 0.18 to 4.36 with a standard deviation of 0.12, and three quarters of them fell between about 0.88 and 1.12. He calls most of the moves fairly limited in size. Longer prices swung more than short ones: a third of opening prices of 10.00 held an expected value above 110 percent against the price they closed at, against less than 1 percent for opening prices of 1.25. Across the whole sample, only 2.8 percent of cases held an expected value above 125 percent.

The 2.8 percent is the figure worth remembering, because it says how rare a large gap to the closing price was in that sample. A 10 percent move sits at about 0.8 standard deviations from the average (0.10 divided by 0.12), so it is unremarkable there; whether it leaves any value is a separate question, answered further down. For the wider question of where the lowest margins sit, start from the best odds guide.

What the article does not tell you

The article lists no leagues, gives no start or end date for the full sample (only a subset of 50,149 opening prices is described, as stretching back over 10 years) and publishes no method for removing the margin. It comes without code or data and was not published in a journal.

The word "opening" is the weakest point. It may not mean the moment the market opened. football-data.co.uk, where Buchdahl publishes Pinnacle prices, says on its own pages that it collects its pre-closing odds on Friday afternoons for weekend fixtures and on Tuesdays for midweek ones, and the article does not say whether its data is that dataset. If it is, the movement measured runs from a snapshot taken before kickoff to the close, and the first price Pinnacle posted falls outside it.

The figures also stop at football: they do not carry over to other sports, to handicap lines, which settle by their own rules (set out in how handicap lines settle), or to today's market. They are football numbers that stop in 2018 at the latest.

A 2016 claim and the data behind it

A 2016 Pinnacle article says that when a line moves on a game, it always happens first at Pinnacle. That is the operator's own claim with no measurement behind it, and the same article leaves it to you to decide whose opinion to trust after a big shortening, the market's or your own.

What a Pinnacle odds drop is worth after the margin

Every closing price carries the book's margin, so a move has to cover it before it is worth anything. Compare the price you took with the closing prices of all the outcomes, strip the margin out of those closing prices (the page on what a Pinnacle margin contains shows how) and divide. The break-even is plain: the gross gap to the published close must beat the overround, which Buchdahl puts at about 2.5 percent on average for a football match. The four rows below use invented prices and two ways of stripping the margin, proportional and power.

Row one is the trap: a price that looks better than the close and is worth less than zero once the margin is out. Row two is the one where the drop stays large under both methods (+7.35 and +8.19 percent). Rows three and four hang on the method, and the last one changes sign with it.

So how much margin does a 10.00 price really carry? The article sketches roughly 10 percent on such a price against the 2.5 percent average, and calls its own curve idealised. If the whole margin sat on that outcome, the same 10.00 that closed at 9.00 would be worth about +8.40 percent at a 102.5 percent book and about +1.01 percent at a 110 percent one. Pinnacle does not publish the split.

Where does a history of Pinnacle prices come from?

Pinnacle is not a documented source of past prices. A README in a GitHub repository for its API, whose ownership no official link confirms, says access has been closed to the general public since 23 July 2025, with bespoke data services for select bettors and commercial users by application. Pinnacle's own API page says access is requested, so the date is the README's claim. The page on Pinnacle data access sets out how access works by application and what the terms say about automated collection. The sources behind this page document no public archive of past Pinnacle prices.

Two sources remain. The first is free: football-data.co.uk offers CSV files for football only, with Pinnacle pre-closing and closing prices, under the reservation about "opening" above. The second is commercial providers. Some of their own pages say their prices come from the public website and may be delayed, and the others say nothing about their source. Pinnacle's terms (clause 3, versions 3.5 and 1.2) ban automated access to its odds, prices and lines, including for monitoring, unless a separate written agreement exists, and reserve the right to suspend accounts tied to it. How the clause applies to a provider's tool, or to a customer who uses one, the text does not say, and this page offers no legal view.

Six questions for any alert tool

An alert is a claim about a price at an instant, so the useful test is what the provider will put in writing. This page does not recommend any alert tool or data provider. The six questions follow from the facts above.

A drop alert is not worth a subscription until the first question has a written answer.

What is worth keeping an eye on instead?

The study's own caveat is that the ratio is a retrospective exercise: the closing price is known only at kickoff, when it is too late to bet. An alert can therefore only guess at where the close will land. The comparison that survives is your own price against the close, set out in the closing line value guide, and the gap between two routes to the same book, in the two-route price check. Both stop short of promising a profit: aggregate data, as the author puts it, cannot say bet by bet whether the value the ratio suggests was really held. And what would a 2.8 percent tail look like in a sport the study never measured? Betting is for adults only, 18+, and what you may do depends on where you live.