Odds movement: what it means, with data from 16,674 matches

• Across 16,674 matches, the favorite's odds moved by an average of 5.7% between open and close (Pinnacle).
• The closing probability predicted the result better than the opening one, by a statistically clear margin.
• A move does not tell you who caused it: it tells you the market changed its mind.

What does odds movement mean?

A match's odds are not fixed.

They change from the moment the bookmaker posts them until the game starts.

Those changes are called odds movement.

Odds that drop indicate the outcome became more likely according to the market.

Odds that rise indicate the opposite.

It sounds simple, but it has traps.

In this article we do not repeat theory.

We download real data and measure what happens.

We only publish what we calculated.

Where the data comes from and how we measured it

We used the public files from football-data.co.uk.

For each match they include opening and closing odds from several bookmakers.

We chose Pinnacle's.

Leagues: Premier League, Championship, Bundesliga, Bundesliga 2, LaLiga, Spain's Segunda División, Serie A, Serie B, Ligue 1, Ligue 2, Eredivisie, Portugal's league, Belgium's, Turkey's, Greece's and the Scottish Premiership.

Seasons: 2022/23, 2023/24 and 2024/25.

Total: 16,674 matches with all six odds complete.

In 2025/26 the file includes Pinnacle only for part of the matches.

That is why that season was left out of the Pinnacle calculations.

We do not mix incomplete data.

How much odds move between open and close

In each match we took the favorite at the open.

That is the outcome with the lowest odds.

We measured how that price changed up to the close.

Two things stand out.

First, there are almost no still odds: only 3.6% do not move.

Second, the favorite drifts more often than it shortens.

The favorite drifting more often does not, by itself, mean the market is losing confidence.

We did not measure the reason for that difference.

So we do not invent an explanation for it.

How many moves are big?

Not every move matters equally.

A 1% change is noise.

A 10% one is news.

In one match out of six, the favorite's odds change by more than 10%.

That is a big difference, for example from 1.80 to 1.62.

In almost three out of four, they change by more than 2%.

Practical conclusion: odds that move a few points are normal.

You do not need to look for a hidden explanation for every change.

The bookmaker's margin: it falls from the open to the close

At Pinnacle, the average margin was 3.75% at the open.

At the close it fell to 3.21%.

The bookmaker trusts its price more as the match approaches.

That is why, to compare odds, you have to remove the margin.

That is what we do in the next section.

If you want to review the concept, read what the bookmaker's margin is and implied probability.

Does the close predict better than the open?

This is the central question.

If the market learns between the open and the close, the closing probability should be closer to the real result.

To check it, we converted the three 1X2 odds into probabilities and removed the margin proportionally.

Then we measured prediction error with two metrics: Brier and log-loss.

For both, a lower value is better.

The close wins on both metrics.

The Brier difference is 0.00256 with a standard error of 0.00041.

That is more than six times the standard error: it is not chance.

For reference: a forecast of one third for each outcome would have a Brier of 0.667.

Pinnacle's odds are far below that.

The improvement from the close is small, but consistent.

Does it happen in every league?

We checked league by league.

The close had a lower Brier in all 16.

The improvement varies a lot.

In the big leagues the improvement is modest.

In Ligue 2 it is larger, perhaps because its opening is less precise.

In the Scottish league it is almost zero.

If the favorite shortens, does it win more?

Another way to look at the data.

We split the matches by the favorite's movement.

And compared with what the probabilities said.

When the favorite shortened, it won more: 54.9% against 48.8%.

The opening probability fell short in the first group and ran high in the second.

The movement carried information.

But look at the closing probability.

In both groups it sits close to the result, within one or one and a half points.

The close had already absorbed almost everything the movement knew.

What about another bookmaker? A check with Bet365

We repeated the calculation with Bet365.

We used 22,113 matches from the same 16 leagues, from 2022/23 to 2025/26 (the latter season incomplete).

The pattern repeats.

The mean change in the favorite's odds was 5.6%.

The opening Brier was 0.58916 and the closing one 0.58708.

The difference is also clear, more than five times its standard error.

A notable difference: Bet365's margin was about 6.3% both at the open and at the close.

That is almost double Pinnacle's.

That is why Pinnacle is used as the reference for the fair price.

What you can do with this

1. Do not panic over a small move: it is normal in three out of four matches.
2. See the move as information, not as an order: it tells you the market changed its mind, not why.
3. Compare against the close: if you got better odds than the close of an efficient market, you got a good price, whether you won or not.
4. Always remove the margin: without it, you compare apples with oranges.
5. Look for the best odds: the advantage of comparing bookmakers is explained in line shopping and in the odds comparator.

If you want to convert odds into probabilities, use the odds converter.

To understand how price connects with edge, read value betting and odds and probabilities.

The limits of this analysis

It is worth being clear about what we did not test.

We did not measure tennis, basketball or other sports.

Nor leagues outside Europe.

We do not know who moved each price.

The public data does not say.

We develop that topic in the article on sharp money.

And the data are in the past.

Something holding over three seasons does not guarantee it holds in the next.

At Gambeta we analyze the matches every day in the Betting Journal.

What to read next

You now know how much odds move and why the close is a good reference.

The next question is who moves them, and whether that can be detected.

Continue with sharp money in betting.

For the full picture, the AI for betting guide brings it all together.

Bet only what you can afford to lose, and if you stop enjoying it, see the responsible gambling guide.

Frequently asked questions

What does it mean when odds drop?

It means the bookmaker assigns more probability to that outcome.

It can be due to news, money wagered or a risk adjustment.

Odds that drop do not guarantee the outcome will happen.

What are opening and closing odds?

The opening odds are the first price at which the bookmaker offers the match.

The closing odds are the last ones before kickoff.

In our data, the source records them at fixed times before the match, not always the first one published.

Which bookmaker is used as the reference?

Pinnacle is the usual reference because it accepts large bets and works with a low margin.

In our data its average margin was 3.75% at the open and 3.21% at the close.

Other bookmakers have higher margins.

Do closing odds predict the result better?

Yes, in our data.

Across 16,674 matches, the Brier score of the close was 0.5838 against 0.5864 for the open.

Lower is better, and the difference is statistically clear.

Is it better to bet when odds fall or when they rise?

There is no rule that works on its own.

When the favorite shortened it won 54.9% of the time, but the close had already absorbed that information.

What matters is comparing the price with the fair probability.

Do these data apply to other leagues or next season?

They describe 16 European leagues over three seasons.

They are a good starting point, not a law.

Each league and each bookmaker behaves differently.