How It Works

TL;DR

Total Over is a football statistics and online sports analytics platform covering 120 leagues worldwide. This page explains what data we hold, how often it updates, how our football metrics work, and where the platform has limitations. Total Over is not a gambling, betting, casino, lottery, sportsbook, sweepstakes or commercial gaming service.

Put simply: Total Over is an online statistics and analytics application, and the planned fantasy football is a free feature built on top of that same data. The statistics are the product; the fantasy layer is an optional, no-cost extension of them. Fantasy points and leaderboard positions cannot be redeemed and hold no monetary or material value, so the fantasy layer does not turn Total Over into a gambling, betting or paid-contest product. It remains a statistics and analytics tool throughout.

1. Data Source

Total Over is built on a single football data feed: Sportmonks Football API v3. All scores, statistics, ratings, xG figures and predicted line-ups visible on Total Over come from that source.

We chose a single-source model deliberately. Mixing data from multiple providers introduces inconsistencies — different xG values for the same shot, conflicting goal-scorer attributions, mismatched player IDs — and we would rather show one verified number than blend several into a figure no one can verify.

Our data feed gives Total Over:

· Live match data for every league we cover: scores, events, statistics and line-ups in real time.

· Live xG updated every few minutes during matches, not only after the final whistle.

· Player Ratings updated every minute during live matches.

· Human-curated expected line-ups before kick-off, refined as new pre-match information arrives.

· Enough capacity to keep all of this fresh across every league at the same time, including on the busiest matchdays.

The rest of this document explains how we use that data, what users see on the platform, and where the limits are.

2. Coverage — Which Leagues and Seasons We Show

2.1. League, team and player scope

Total Over covers around 120 leagues worldwide. This makes the platform a broad global football reference, not only a top-five-leagues statistics site.

Database scale (approximate):

The exact totals fluctuate from season to season as squads change, leagues add or remove participants and the list of competitions we cover is updated. The figures above are rounded approximations based on typical squad sizes applied across the leagues we cover.

What is included in this portfolio:

· The five major European leagues: Premier League, La Liga, Bundesliga, Serie A and Ligue 1.

· The full UEFA club competition pyramid: Champions League, Europa League and Conference League.

· Other major domestic leagues across Europe, such as Eredivisie, Liga Portugal and Pro League.

· Major national-team competitions, including the World Cup, European Championship, Copa America, Africa Cup of Nations and Nations League.

· Major South American leagues and Copa Libertadores.

· MLS and other North American competitions.

· Domestic cup competitions tied to the leagues above.

· Selected second-tier and reserve leagues that act as feeder competitions to the top divisions above.

The exact mix of data categories available per league varies by competition. Top leagues have the full advanced dataset, while smaller competitions may have fewer categories available.

2.2. Historical depth, by data type

This is the single most important nuance of our database and the one with the largest practical impact:

This means three concrete things for users:

1. Career-long comparisons are not possible for advanced metrics. A player’s xG per 90 for season 2018/19 versus 2023/24 cannot be shown because xG was not collected before 2024.

2. Historical rating sparklines are short. No algorithmic rating exists for matches played before January 2025, so we cannot show them.

3. Old seasons of basic stats are not in our database. A player profile on Total Over currently shows the last three seasons of basic stats. Older seasons are not currently available on the platform.

3. The Full Data Catalogue

This section lists every type of data that Total Over can display.

3.1. Leagues, seasons and standings

3.2. Fixtures and match insights

3.3. Teams, players and officials

3.4. Advanced metrics

4. Player Rating — How It Works

4.1. What it measures

The Player Rating on Total Over is a numerical assessment of a player’s performance in a single match. It is generated by an algorithm and reflects only what happened on the pitch — no editorial weighting, no human override and no fan voting.

4.2. The scale

4.3. How it is calculated

Each starter begins the match at 6.5 and that score moves up or down based on 51 individual statistics weighted differently per position. The system covers goalkeepers, defenders, midfielders and forwards using position-aware weightings for the same underlying actions.

Actions that increase the rating:

· Goals scored

· Duels won

· Blocked shots

· Key passes

· Saves for goalkeepers

· Successful tackles, interceptions and clearances

· Successful dribbles

· Accurate passes above the expected level for the player’s position

Actions that decrease the rating:

· Goals conceded, weighted by position and context

· Red cards

· Yellow cards with a smaller penalty

· Errors leading to a goal

· Big chances missed

· Penalties given away

· Dispossession in dangerous zones

4.4. The substitute rule

A player who comes off the bench does not receive a rating unless one of these two conditions is met:

1. They spend at least 5 minutes on the pitch.

2. They produce a prominent event at any point while they are on the pitch, no matter how short their playing time.

Prominent events include:

· Goals

· Assists

· Big chances created

· Big chances missed

· Clearance off the line

· Errors leading to a goal

· Penalties committed

· Red cards

· Yellow cards

· Key passes

· Penalties missed

If a substitute plays 3 minutes and none of the events above occur, they appear in the line-up with no rating for that match. This is intentional, not a bug.

4.5. Coverage

The vast majority of our 120 leagues carry Player Ratings, including the major European and international competitions. A small number of secondary leagues do not yet have Player Ratings; in those cases, we display the match data without a rating value rather than showing an estimate.

5. xG Metrics — Full Reference

Total Over shows a full set of 13 xG metrics, available on every fixture from season 2024 onwards.

5.1. The 13 metrics

Note that two metrics share the abbreviation xGP. One is offensive (expected goals from penalties) and the other is defensive (expected goals prevented). On Total Over, we always use the longer name to avoid confusion.

5.2. Calibration reference

5.3. How xG is calculated

xG values for individual shots are computed by a statistical model trained on a large archive of historical shots. The inputs include shot location, angle, distance, shot type and the positioning of the goalkeeper and defenders at the moment of the shot.

Player identity is not a factor. For the model, a shot taken by a world-class forward has the same xG as the same shot taken by a centre-back if the shot context is identical. A player’s or team’s total xG for a match is the sum of the xG of all shots taken.

5.4. Important caveats about xG

Two common misconceptions affect how xG on Total Over should be read:

1. A higher xG does not mean the team should have won. xG measures chance quality, not match outcome.

2. xG is not a perfect comparison of finishing skill. It measures the context of a shot, not the identity or quality of the player taking it.

5A. Match Probabilities

Total Over shows model-based outcome probabilities for upcoming and live fixtures. These come directly from the data feed’s prediction output, computed from historical and current match data. Probabilities become available up to 21 days before kick-off, and the underlying model refreshes its predictions daily as new data arrives. They are presented as football and fantasy context — a way to understand how a match is likely to unfold — and never as betting odds.

5A.1. What we show

The probability set available to Total Over includes:

· Full-time result probabilities (home win, draw, away win), shown as percentages.

· The probability that both teams score at least once.

· The probability distribution of total goals in a match — for example, the likelihood of a match producing three or more goals.

· Other modelled outcome distributions, such as the likelihood of specific final scores, where available.

We display these figures as percentages or index values, sit them alongside form, xG and head-to-head context, and label them clearly as statistical probabilities for football analysis and free fantasy decision-making.

5A.2. What we deliberately exclude

The prediction part of the data feed also offers a “value bet” output, which compares modelled probabilities against bookmaker prices to flag potential betting value. Total Over does not use this output, and it is excluded from the product entirely. We do not ingest, display, derive from or link to value bets, bookmaker prices or betting markets of any kind. Match probabilities on Total Over are a statistics and analytics feature only.

Probability figures are provided for football analysis and free fantasy decision-making only. They are not betting odds, betting advice, financial advice or a recommendation to place a bet.

6. Expected Lineups

6.1. Three phases of line-up display

6.2. How the prediction evolves

Expected line-ups in our data feed are human-curated: analysts monitor team news, injury reports and pre-match coverage in real time, rather than relying on an algorithm alone. The expected line-up is updated whenever new pre-match information arrives. Triggers include:

· A new injury or suspension confirmation.

· A pre-match press conference quote about the XI.

· A late training-ground report.

· A player being added to or removed from the squad.

The closer to kick-off, the more accurate the prediction. Twenty-four hours before kick-off, expected line-ups are useful but volatile. An hour before kick-off, they are typically very close to the eventual starting XI — and once the club confirms its team, the prediction is replaced by the official line-up. Across the top European leagues, the data provider reports prediction accuracy of roughly 75–88%, varying by competition.

6.3. What’s inside a line-up

For each predicted or actual line-up, we have:

· Player identifier, name and shirt number.

· Position and formation slot.

· Team identifier.

· Captain flag.

· All available xG metrics for the player in that specific match.

7. Update Frequency

This section shows how fresh each type of data on Total Over is.

7.1. How often each data type updates

7.2. How we poll the data feed

During live matches, Total Over checks the data feed every 5 seconds, picking up only what has changed since the last check. When no matches are live, the cadence drops to once every 5 minutes. Reference data such as venues and league metadata is refreshed weekly.

Our infrastructure is sized so that even on a heavy matchday with 10 or more simultaneous live matches across the 120 leagues, this polling rate uses only a small percentage of available capacity. In normal conditions there is no need to slow updates down at peak times.

7.3. End-to-end latency

In live mode, Total Over shows updates within roughly 5 seconds of any change in the underlying data feed. The data feed itself publishes goals within roughly 10 seconds of the real-world event and other match events within 10–30 seconds. The end-to-end delay from a goal being scored to it appearing on a Total Over page is therefore typically under 20 seconds.

The practical limit on freshness is not our polling rate. It is how quickly the underlying feed publishes changes and, in the case of xG, the gap of up to five minutes between consecutive xG updates during live matches.

8. Technical Architecture

8.1. Operations we allow in our compute layer

8.2. Operations we explicitly do not perform

9. Strengths of Our Database

9.1. Single source of truth

Every page on Total Over shows the same numbers because they all come from one data feed. There is no second provider being blended in, no manually edited override and no rounding that differs between pages.

9.2. Live xG, not post-match only

Many free stat sites show xG only after the final whistle. On Total Over, xG updates live during the match, refreshed every few minutes.

9.3. Complete xG family

Many sites publish only a handful of xG variants. Total Over carries all 13, including xGoT, goalkeeper xGP, xPTS, xGSP, xGOP, xGC and xGFK.

9.4. Algorithmic Player Ratings

The Player Rating system is fully algorithmic — no editorial intervention, no fan-vote bias and no media opinion influencing the number.

9.5. Sub-20-second live data

Live scores and events arrive in Total Over in under 20 seconds end-to-end in typical conditions.

9.6. Broad league coverage

Total Over covers 120 leagues in its database, including major leagues, international competitions, domestic cups and selected lower-tier competitions.

9.7. International tournaments use the same pipeline

Champions League, World Cup, European Championship, Africa Cup of Nations and Copa America use the same infrastructure as domestic leagues, where the relevant data is available.

9.8. No throttling on peak matchdays

Total Over is designed to keep updating at full cadence on busy matchdays, rather than slowing down or limiting updates because of traffic volume.

10. Limitations of Our Database

10.1. Historical xG is not available

xG data starts at season 2024. There is no xG in the system for seasons before that date, so career-long xG comparisons and five-season xG trend charts are not possible.

10.2. Historical Player Ratings are short

The current Player Rating system went live on 6 January 2025. Matches played before that date do not have a rating value.

10.3. Older historical data is not in our database

Our database currently includes the last 3 seasons of basic historical data. Older seasons are not available on Total Over today.

10.4. No xA as a first-class metric

Expected Assists is not available in our data feed as a standalone metric. The closest proxy we have is Key Passes, which we label clearly as a proxy.

10.5. No set-piece taker designation

The data feed does not flag which player is the designated free-kick taker, corner taker or first-choice penalty taker. Where we infer this from event history, we label it as an inference.

10.6. xG model differences versus other providers

Our xG model is different from those used by other providers. A user may see a slightly different number for the same match on another site. Total Over uses one model for consistency.

10.7. Sub-5-minute live xG latency

Live xG updates every few minutes, with a maximum gap of five minutes between consecutive updates. It is a rolling figure, not a continuous live stream.

10.8. Coach data has a flat ceiling

Coach statistics are limited to a smaller set of base metrics. More advanced coach metrics must be calculated from fixture-level line-ups and events and are labelled accordingly.

10.9. No control over data corrections

If the data feed updates a historical value, Total Over picks up the correction on its next polling cycle. Users may occasionally see a number change after a match has finished.

10.10. Single point of failure on availability

If the underlying feed goes offline, Total Over enters degraded mode and shows the last known data with a visible timestamp until the feed is restored. We do not currently use a fallback provider.

11. Scope: What’s Out, What’s Coming

11.1. Total Over is not a gambling platform

Total Over is a football statistics reference and online sports analytics platform. We do not operate gambling, betting, casino, lottery, sportsbook, sweepstakes or commercial gaming products. We do not accept, process, transmit, facilitate, settle or promote bets.

Any fantasy, prediction or leaderboard feature on Total Over is free, points-based and for football analysis or entertainment only. Fantasy points and leaderboard positions have no monetary or material value.

The way to read Total Over is as a single statistics and analytics application with an optional free layer on top. The statistics and analytics — live scores, xG, Player Ratings, expected line-ups and the data pages built from them — are the product. The planned fantasy football is a free feature that re-uses the same data; it adds no stake, paid entry, prize or redeemable value, and so it does not change the product’s nature. There is no point in the product, in either layer, at which a user pays to enter, places a wager, or can win or withdraw anything of value.

11.2. Permanently outside our scope

11.3. Planned / in development

The features below are not in the current product, but are on the roadmap. They will be built on the same football data foundation and will remain free of stakes, paid entry, prizes and redeemable value.

12. Glossary

13. External References

Technical methodology for the underlying data feed:

· Sportmonks Football API v3 documentation: https://docs.sportmonks.com/v3

Document version 3.1 · Last reviewed: June 2026 · Maintained by the Total Over data team.