Are Early-Season Hot Starts Real? What PDO and xG Reveal After the Opening Matchweeks

By the end of September, every league table has a surprise near the top and a giant stuck in mid-table. The question is whether those early positions reflect real quality or a run of favourable bounces. RubiScore, the live football score and data platform at https://rubiscore.com, records the shots, goals and saves that let fans test a hot start instead of simply trusting it.

This data-study asks one question: after roughly six to eight league matches, how much of a team's position is explained by sustainable performance, and how much by finishing and goalkeeping luck that tends to fade? The answer relies on two tools, expected goals and a blunt but revealing number called PDO.

Why Early Tables Are So Hard to Read

Early in a season, the sample is tiny. A team that has played seven matches has faced seven opponents, some strong and some weak, often with an uneven split of home and away fixtures. One deflected goal or one goalkeeping error carries far more weight in a seven-game table than it will across a full campaign.

That is why the opening weeks are the busiest time for data-led reality checks. Match data on RubiScore lets a reader move beyond points and look at what produced them: how many chances a team created, how good those chances were, and how often they were converted or denied. The gap between the chances and the scoreline is where hot starts are either confirmed or exposed.

What PDO Measures

PDO is the sum of two percentages: a team's shooting percentage (the share of its shots that become goals) and its save percentage (the share of opponents' shots on target that its goalkeeper stops). It is usually written on a scale where 1000 represents the long-run average, or sometimes as 100.

The metric originated in ice hockey analytics and takes its name from the online handle of the analyst who popularised it. Its logic transfers neatly to football:

• A team scoring on an unusually high share of its shots is finishing above normal levels.
• A team whose goalkeeper stops an unusually high share of shots is defending above normal levels.
• Added together, a very high PDO suggests both ends of the pitch are running hot at the same time.

Across a full league, PDO must average out close to the baseline, because every goal scored by one team is a goal conceded by another. Individual teams drift above or below that line, and the further they drift in a small sample, the more likely it is that luck is playing a part.

PDO is not a headline figure on most scoreboards, but its ingredients are. Shots, shots on target, goals and goals conceded for each fixture are logged in the match statistics that Rubi Score publishes, so the calculation can be rebuilt by hand for any team in a few minutes.

A Worked Example With Two Hypothetical Teams

Consider two invented sides after seven matches, used only to illustrate the method. Team A sits second in the table. It has scored from an unusually large share of its shots, and its goalkeeper has saved almost everything on target. Its expected goals difference, however, is only slightly positive. Team B sits eleventh. It has created more and better chances than its opponents in most games, but its forwards have missed good opportunities and its goalkeeper has conceded from several low-quality efforts.

On points, Team A looks like a contender and Team B looks ordinary. On process, the picture reverses: Team B is the stronger performer, while Team A's position depends on conversion and shot-stopping that sit well above typical levels. Neither conclusion is guaranteed to hold, but the second reading is the one historical patterns tend to support over the following months.

What the Data Tends to Show

Research across several sports points to the same broad pattern. Shooting and save percentages are among the least stable team statistics from one block of matches to the next. Shot volume, chance quality and territorial control are far more consistent.

In practice, that produces three recognisable profiles in the early weeks:

• Genuine strong starts. The team leads on expected goals difference as well as actual goal difference, and PDO sits near the baseline. The results are backed by the process.
• Hot starts with a warning light. The team sits high in the table, but its expected goals difference is modest or even negative, and PDO is well above the baseline. Finishing and goalkeeping are covering for average chance creation.
• Unlucky slow starts. The team is below where its underlying numbers suggest, with a strong expected goals difference and a PDO below the baseline. Chances are being created but not converted, or opponents are scoring from few opportunities.

The second and third groups are where regression to the mean becomes the story. When PDO sits far from the average, the most likely next move is back towards it, which usually means results cool down for the hot team and pick up for the unlucky one.

How Expected Goals Adds the Missing Context

PDO on its own treats every shot as equal, which is its biggest weakness. A team taking many low-quality long-range efforts will naturally have a low shooting percentage, and that does not mean it is unlucky. A team that creates tap-ins will naturally convert more.

Expected goals fixes that problem by weighting each shot by its historical chance of becoming a goal. Comparing goals scored with expected goals, and goals conceded with expected goals against, tells a reader whether a team is outperforming the quality of its chances rather than just the quantity.

The strongest early-season signal comes from reading both numbers together:

1. Check expected goals difference per match to judge the underlying performance level. 2. Compare actual goals with expected goals at both ends to see where any gap sits. 3. Look at PDO to see whether finishing, goalkeeping or both are driving the gap. 4. Weigh the schedule, because a run of weak opponents can inflate every number.

Confounders and Limits

Any honest early-season study has to name what the numbers cannot settle.

• Elite finishers are real. Some forwards convert chances above expected levels over many seasons. A team built around one of them can sustain a slightly elevated shooting percentage.
• Elite goalkeepers are real too. A goalkeeper with a strong shot-stopping record can hold save percentage above average for longer than a typical keeper.
• Game state distorts shot numbers. A team that scores early often sits deeper and concedes more low-quality shots, which can make its defensive numbers look worse than its actual control of the match.
• Schedule strength varies. Six matches against promoted sides and six against title contenders are not comparable samples.
• Tactical change is possible. A new coach or a new system can genuinely shift performance, and early numbers may be capturing a real change rather than noise.

For these reasons, PDO is best treated as a question rather than an answer. A high figure says "check whether this is sustainable", not "this team will collapse".

How Long Until the Numbers Settle?

There is no single threshold, but the pattern in performance analysis is consistent: volume-based metrics such as shots and expected goals become informative sooner than conversion-based metrics such as shooting percentage. By the time a league passes the ten-match mark, expected goals difference usually carries meaningful information about quality. Shooting and save percentages need considerably longer, often most of a season, before they separate real skill from variation.

That is also why October is a natural moment for a regression watch. The early table has taken shape, but the samples are still small enough that luck can dominate. Teams whose points total rests on an extreme PDO are the ones most likely to move, in either direction, by the winter break.

Verdict: Trust the Process Numbers First

Early-season hot starts are sometimes real and sometimes borrowed from future matches. The data tends to favour teams whose results are matched by their chance creation and chance prevention, and to warn against teams whose position depends on finishing or goalkeeping far above normal levels.

Match-by-match figures of the kind RubiScore tracks make these checks repeatable each week rather than a one-off exercise. A sensible early reading combines three steps: judge performance through expected goals difference, identify where any gap between goals and expected goals sits, and use PDO as a flag for likely regression. None of these tools predicts a single result, and none replaces watching the matches. Used together across the opening weeks, they turn a noisy table into a clearer picture of which starts deserve belief and which deserve patience.