Offside Technology and Football Predictions: How Semi-Automated Calls Affect Results

Introduction

Offside technology and football predictions have become inseparably linked in the modern era, as semi-automated offside technology (SAOT) and its various implementations across elite competitions have fundamentally altered the frequency, accuracy, and timing of offside decisions. For anyone building analytical models and making forecasts for football matches, understanding how SAOT works, what it changes relative to the previous assistant-referee flag system, and how it affects goal counts, match momentum, and result distributions is no longer optional — it is a core analytical competency. The introduction of semi-automated offside technology in the UEFA Champions League, FIFA World Cup 2022, and subsequently in domestic leagues represents the most structurally significant officiating development since the original introduction of VAR. Its effects on football predictions ripple through goal markets, correct score analysis, in-play decision-making, and the statistical baselines analysts use to build their forecasting models.

This guide provides a thorough examination of how semi-automated offside technology works, the statistical patterns that have emerged since its implementation, and the specific ways in which SAOT affects the analytical considerations that underpin football predictions. From the reduction in marginal offside disallowances to the changed timing of goal confirmation, from the impact on attacking patterns to the downstream effects on xG-based modelling, this guide equips analysts with the framework they need to properly account for offside technology in their prediction work.

How Semi-Automated Offside Technology Works

Skeletal Tracking and Body Part Detection

Semi-automated offside technology represents a significant technological leap beyond both traditional assistant-referee judgements and the first generation of VAR-assisted offside review. SAOT uses a combination of skeletal tracking data — derived from multiple cameras positioned around the stadium — and ball-tracking sensors to construct a three-dimensional model of every player's body position at the precise moment the ball is played. The system can identify the offside line with a claimed accuracy of within a few centimetres, eliminating the margin of human error that characterised both on-field decisions and early VAR reviews.

Review Speed and Accuracy Improvements Over VAR

The process works as follows: when a potential offside situation occurs during a match, the tracking system captures the ball departure moment and the precise positional data for all relevant players. This data is then used to generate a freeze-frame visualisation — the familiar on-screen graphic showing the two players with coloured lines extending from their bodies — that allows match officials to make a final determination. Crucially, SAOT reduces the time required for offside review compared with the manual line-drawing process used in first-generation VAR, where analysts had to manually identify the correct frame and draw lines based on lower-resolution camera footage. In some implementations, SAOT has reduced average review time from 70 or more seconds to under 30 seconds.

The FIFA World Cup 2022 in Qatar was the first major tournament to use SAOT at scale, and the results were immediately visible. Offside decisions that previously required lengthy VAR reviews were resolved significantly faster, and the rate of marginal goals being disallowed — specifically goals scored by attackers whose arm or shoulder was a few centimetres ahead of the last defender — was notably affected. Understanding these effects on goal counts is the starting point for translating SAOT awareness into prediction model adjustments. The broader context of VAR's impact is covered in our VAR impact on predictions guide, which provides the foundational framework into which SAOT-specific analysis fits.

How SAOT Changes Goal Disallowance Rates

Marginal Offside Decisions and Their Statistical Impact

One of the most analytically significant effects of SAOT implementation has been a measurable shift in the rate at which goals are disallowed for offside. Under the original assistant-referee system, borderline offside goals were frequently missed — both because human visual perception struggles to judge millimetre-level separations at high speed, and because linespeople were instructed to give the benefit of the doubt to the attacker. Under first-generation VAR, this changed dramatically: goals that were previously allowed because the assistant could not detect the marginal offside were now disallowed after multi-minute reviews, producing a sudden increase in goal disallowance rates that frustrated players, managers, and supporters alike.

Which Goal Types Are Most Affected by SAOT

SAOT's impact on this pattern is nuanced. The technology's ability to detect extremely marginal offsides with high accuracy means that goals involving attacker positions that are genuinely, physically offside by even a fraction are correctly disallowed. At the same time, because the technology is more reliable and faster, the ambiguity that previously caused correct goals to be overturned by incorrectly drawn VAR lines is reduced. The net effect in the tournaments and competitions where SAOT has been implemented is a stabilisation of the offside disallowance rate compared with early VAR implementation, with fewer extremely marginal overturns in either direction.

For prediction analysis, this stabilisation matters. In the early VAR era, the elevated rate of goal disallowances — particularly goals from attacking set pieces, which generated a disproportionate share of marginal offside situations — created downward pressure on actual goal totals relative to expected goals calculations. Models built on pre-VAR data systematically over-estimated goal totals when applied to VAR competitions. SAOT's stabilisation of this effect means that xG-to-goals conversion rates in SAOT competitions are somewhat more predictable than in first-generation VAR competitions, which benefits analysts building goal-total models. Our guide on using expected goals in predictions explains how xG conversion assumptions should be calibrated for different officiating technology environments.

The Impact of SAOT on Attacking Pattern Development

Teams Adjusting Runs to Account for SAOT

Semi-automated offside technology does not merely change how decisions are made — it changes how teams play. Coaches and tactical analysts at clubs in competitions with SAOT have adjusted their attacking patterns in response to the technology's greater accuracy in detecting marginal offsides. Specifically, the attacking runs that once relied on millimetre-level timing to stay onside have become more risky, because the traditional visual uncertainty that might have allowed a marginal position to go unchallenged is removed. Some attacking coaches have responded by adjusting their runners' timing to ensure a more comfortable margin of daylight behind the last defender, slightly reducing the number of ultra-tight offside attempts.

Striker Positioning Changes Under SAOT

At the same time, the defensive side has responded in complex ways. Some defensive coordinators have become more aggressive in attempting to spring the offside trap, knowing that SAOT will reliably catch attackers who time their runs too early. The frequency of organised high defensive lines — particularly in elite European club competition — has increased since SAOT implementation, partly because the technology provides defenders with greater confidence that marginal offside positions will be caught. This tactical shift has analytical implications: matches between high-line defensive teams tend to produce more open-play chances (because the high line opens space in behind when beaten) but also more disallowed goals. Analysts should track defensive line height as a contextual variable in SAOT competitions, as it shapes both the goal-scoring and goal-disallowance dynamics of individual matches.

Tactical analysis that accounts for SAOT's influence on playing patterns connects to the broader framework of football formations and tactical systems analysis, which provides the tools to evaluate how team structure shapes match outcomes in any officiating environment.

SAOT and Set-Piece Predictions

Header Goals and Marginal Offside Frequency

Set pieces represent the area of play most significantly affected by offside technology, because a disproportionate share of set-piece attacking runners are positioned in or near offside positions. Corner kicks, free kicks in the attacking third, and long throw-ins all generate situations where attackers are deliberately positioned close to the defensive line to maximise scoring opportunity — and where the slightest mistiming or positional adjustment can tip a player from onside to offside. Understanding how SAOT affects set-piece goal patterns is therefore essential for analysts who include set-piece-derived goal predictions in their analytical toolkit.

Corner and Free-Kick Goal Expectation Adjustments

The research available from SAOT competitions suggests that marginal set-piece goals — those involving attackers whose position was extremely close to the last defender — are disallowed at a higher rate than under the original assistant-referee system, but at a more consistent and predictable rate than under first-generation VAR. This has led some analysts to apply a modest downward adjustment to set-piece goal probability in SAOT competitions, particularly for goals involving back-post runners and late arrivals who benefit from minimal separation from the last defender. For predictions focused on goal sources, our guide on set-piece specialist analysis provides the methodological framework for incorporating these considerations into set-piece goal forecasting.

Corner kick analysis is particularly affected by SAOT dynamics. Near-post and far-post movement patterns that generate goals from corners involve multiple players in close proximity to the defensive line, and SAOT can detect the marginal positions that human linespeople might miss. Analysts who track corner-to-goal conversion rates in SAOT competitions will notice a slightly different pattern than in non-SAOT competitions, and calibrating conversion assumptions accordingly improves the accuracy of corner-based prediction models.

Review Speed and In-Play Prediction Adjustments

Faster Confirmations and In-Play Market Timing

For in-play analysis and live forecasting, the speed of SAOT review creates materially different conditions than the extended VAR reviews of the early adoption era. When a goal is scored in an SAOT competition and a potential offside has occurred, the review process is substantially faster than under manual VAR line-drawing. This reduced uncertainty window changes the calculus for in-play predictions made between the moment a goal is scored and the moment it is confirmed.

In early VAR implementations, the period of uncertainty following a potential goal — while officials reviewed the offside — could extend for one to two minutes or longer, during which the actual scoreline was technically unconfirmed. In SAOT competitions, this window is substantially compressed. For analysts engaged in live in-play analysis, this means decisions need to be made more quickly after a goal is scored, because the confirmation or disallowance comes faster. The practical upshot is that SAOT removes some of the analytical opportunity that existed in the early VAR era around in-play adjustments during long review periods. Our live in-play strategy guide covers how to manage these real-time decision points effectively in the modern officiating environment.

Reduced Uncertainty Compared to Manual VAR

The speed of SAOT also affects the psychological dynamics of matches. When a goal is scored and quickly confirmed without a lengthy VAR check, the momentum effect of the goal is more immediate and pronounced than when confirmation is delayed. Conversely, a swift disallowance allows the match to reset quickly, without the extended celebration and anticipation that prolonged VAR reviews generate. These momentum effects — while difficult to quantify precisely — are relevant to in-play analysis frameworks that consider the psychological state of teams following significant match events.

SAOT Accuracy and Residual Controversy

While semi-automated offside technology represents a significant improvement over previous methods in terms of accuracy and speed, it is not without residual controversy. The primary source of ongoing debate concerns the definition of a "playing arm" in the context of offside calculations. FIFA's Laws of the Game specify that the arm is not part of the body for offside purposes — hands and arms are excluded — but the precise boundary between shoulder and upper arm has been disputed in several high-profile cases. SAOT uses skeletal joint positions to define these boundaries, which means that minor differences in how the system maps a player's shoulder joint can affect whether an extremely marginal goal is allowed or disallowed.

A second area of residual uncertainty concerns the precise frame used to determine ball departure. Even with advanced tracking systems, the exact millisecond at which the ball leaves the passer's foot is not always unambiguous, and a one-frame difference in the departure moment — corresponding to approximately 40 milliseconds at standard camera frame rates — can be enough to shift a marginal position from onside to offside. These limitations mean that SAOT, while substantially more accurate than its predecessors, still produces a small number of contested decisions each season in competitions where it is deployed.

For prediction analysis purposes, these residual uncertainties reinforce the importance of not over-engineering models based on SAOT-specific adjustments. The technology reduces but does not eliminate decision uncertainty, and analysts should treat their SAOT-calibrated adjustments as probabilistic improvements rather than deterministic corrections. The philosophical approach to modelling uncertainty in football predictions is explored in depth in our data-driven predictions methodology guide.

SAOT Rollout: Which Competitions Use the Technology

Current SAOT Adoption Across Major Leagues

Understanding which competitions have implemented SAOT is essential for knowing when SAOT-adjusted analytical frameworks apply. As of the current season, SAOT is deployed in the UEFA Champions League, UEFA Europa League, and UEFA Europa Conference League, as well as in the FIFA World Cup and major continental tournaments. Some domestic leagues have also implemented or piloted the technology, though the rollout is uneven. The Premier League, Bundesliga, and Serie A have been at various stages of SAOT consideration or implementation, while lower divisions and most domestic cup competitions continue to use first-generation VAR or no VAR at all.

Expected Rollout Timeline for Remaining Competitions

This patchwork of technology deployment means analysts must maintain awareness of the specific officiating technology in play for each competition they are analysing. A model calibrated for SAOT competition conditions should not be applied unchanged to a domestic league still using manual VAR, and vice versa. Tracking the officiating technology context for each competition is a basic but important element of maintaining analytical rigour across multiple competitions simultaneously. The conference league and other UEFA competitions are covered in our Conference League analysis guide, which includes officiating technology context.

Expert Insight: Experienced analysts who have tracked offside technology evolution since the introduction of VAR observe a consistent pattern: the most analytically productive response to new officiating technology is patience. In the first season of any new technology deployment — whether original VAR or now SAOT — the statistical sample is too small and the playing and coaching adaptations too recent to draw firm conclusions about the long-term effect on goal patterns. Analysts who rushed to build VAR-adjusted models in 2018 and 2019 often found their adjustments based on early data were overstated, because the initial disruption to goal patterns was partly a novelty effect that stabilised over subsequent seasons as players, coaches, and officials adapted. The same caution applies to SAOT. The most reliable analytical approach is to build a multi-year dataset of SAOT competition results — tracking goal counts, disallowance rates, and xG conversion figures — and apply adjustments based on the stabilised pattern rather than the initial disruption period. In the meantime, the most productive SAOT-aware insight for analysts is simply to understand the specific tactical contexts — high defensive lines, set-piece heavy teams, narrow margin attacking runners — where SAOT's effects are most concentrated.

Analyst Note: When incorporating semi-automated offside technology context into your prediction analysis, the following practical guidelines apply. First, identify the competition and confirm whether SAOT is deployed. For UEFA club competitions at European level, SAOT is active; for domestic leagues, check the specific season's officiating setup. Second, note the defensive shape of both teams — teams that defend with high lines in SAOT competitions are more likely to generate marginal offside situations, which may affect set-piece goal estimates. Third, review the recent history of disallowed goals for teams in SAOT competitions, particularly for clubs with strikers known for late, marginal runs. Fourth, avoid over-correcting for SAOT effects — the adjustments required are real but relatively modest, and gross over-engineering of goal models based on officiation technology alone will introduce more error than it removes. Fifth, in live in-play situations, factor in that goal confirmation under SAOT is faster, so post-goal in-play decision windows are compressed relative to early VAR. The interplay between officiating technology and penalty decisions — another major analytical consideration — is covered in detail in our guide on VAR and penalty decision patterns.

Case Studies: SAOT Effects in Real Match Situations

The 2022 FIFA World Cup provided the most comprehensive real-world test of SAOT at the highest level of the game. One of the most analytically instructive cases occurred in the group stage match between England and USA, where several attacking runs from both sides generated potential offside situations that were resolved quickly and consistently by the SAOT system. In one instance, an England goal was confirmed after a rapid check that, under first-generation VAR, would likely have required 90 seconds or more of manual line-drawing. The efficiency of the review contributed to a more fluid match experience and removed the extended uncertainty period that had become a defining feature of early VAR football.

A more analytically complex case arose during the Champions League in the 2022-23 season, where a goal in a high-profile knockout match was disallowed after SAOT identified a marginal offside position from what appeared to be a shoulder. The decision generated significant debate about the "playing arm" boundary, illustrating the residual grey areas that even advanced technology cannot fully eliminate. For prediction analysts, this case reinforced the importance of understanding that SAOT significantly improves accuracy at the cost-of-the-call level but does not make offside decisions entirely unambiguous at the extreme margins. Incorporating a small residual uncertainty factor for extremely marginal attacking positions — even in SAOT competitions — reflects the technological reality.

A third case study concerns the domestic league context, where SAOT is not yet deployed. In Premier League matches without SAOT — where traditional VAR line-drawing remains in use — analysts comparing goal disallowance rates to SAOT competitions will find a somewhat higher frequency of contested decisions and occasionally longer review times. The comparative analysis between SAOT and non-SAOT competition conditions clarifies the specific value that the technology adds, and helps analysts make the appropriate adjustments when switching their analytical focus between competitions with different officiating setups. The broader VAR framework impacting predictions across all competition levels is explored in our comprehensive guide to VAR's impact on football predictions.

Expert Insight: Semi-automated offside technology has effectively eliminated the subjective margin that previously allowed tight offside calls to go either way depending on the linesman's viewing angle. This is not a neutral change for prediction analysis — it systematically disadvantages teams whose attacking patterns rely heavily on late runs into the penalty box by players who play on the shoulder of defenders. Analysts modelling expected goals in SAOT competitions should review whether their attacking pattern assumptions were calibrated before or after SAOT introduction, as pre-SAOT data may overstate certain goal types.

Goalkeeper Behaviour and SAOT: An Overlooked Analytical Variable

One area where semi-automated offside technology creates an indirect but analytically real effect is goalkeeper positioning and distribution. Elite goalkeepers in SAOT competitions have become more assertive in their sweeping play, partly because the technology's ability to precisely confirm offside positions means that an aggressive goalkeeper charge to collect a through ball is more reliably followed by an offside flag when the receiving attacker was genuinely in an offside position. In pre-VAR and even first-generation VAR football, a goalkeeper who charged off their line and missed the ball risked conceding a goal even if the attacker was marginally offside, because the flag might not be raised. Under SAOT, the reliable catch of marginal offside positions provides goalkeepers with slightly greater confidence to act aggressively on through balls into their penalty areas.

This behavioural shift — while subtle — has measurable implications for the frequency of one-versus-one situations that do result in shots on goal. In SAOT competitions, the subset of through-ball situations where the attacker is genuinely onside (rather than marginally offside) is more cleanly distinguished from those where the offside is caught, meaning the "true" one-versus-one attempts that reach the goalkeeper are slightly purer in quality terms. For prediction analysts tracking goalkeeper performance metrics and shot-stopping statistics in SAOT competitions, this suggests that the quality of shots faced by goalkeepers in through-ball situations has slightly improved in SAOT environments — an analytical nuance worth tracking when building goalkeeper-performance-adjusted goal probability models.

SAOT Data and Future Predictive Modelling

As SAOT technology generates increasingly large datasets of positional information, the analytical potential for prediction modelling extends well beyond simple offside decisions. The skeletal tracking data underlying SAOT captures the precise body positions of every player at every moment of every match, creating a data environment of extraordinary richness for advanced analytical work. While this data is not yet widely available to independent prediction analysts, its eventual broader availability — whether through governing body publication or commercial data provider licensing — will enable a new generation of prediction models based on player positioning, movement patterns, and spatial relationships that are currently impossible to track manually.

For analysts building prediction frameworks today, the implication is to remain alert to developments in SAOT data availability and to begin thinking about how positional tracking data could enhance their models. The transition from event-level data (what happened and where on the pitch) to positional tracking data (where every player was at every millisecond) represents the next frontier in football analytics, and SAOT is the first technology that brings positional tracking into mainstream deployment at the highest level of the game. Analysts who understand the technology well — and who are prepared to incorporate positional data into their models when it becomes accessible — will be at the forefront of the next generation of prediction methodology. The framework for building advanced analytical models that can incorporate new data types is covered in our guide on building your own prediction model, which provides the methodological foundation for this type of evolutionary model development.

Conclusion

Semi-automated offside technology represents a meaningful evolution in football officiating that has direct, quantifiable effects on the analytical landscape for football predictions. The technology's greater accuracy in detecting marginal offsides has stabilised goal disallowance rates compared with first-generation VAR, changed tactical patterns around defensive line height and attacking runner timing, compressed review periods for in-play analysis, and created a more predictable environment for xG-to-goals conversion modelling in the competitions where it is deployed. Analysts who maintain awareness of which competitions use SAOT, understand the specific contexts where its effects are most concentrated, and apply calibrated adjustments to their models will produce more accurate forecasts than those working without this awareness.

The key principle governing the analytical response to SAOT is proportionality: the technology matters, it does change outcomes, and it should be factored into prediction models — but the adjustments required are specific and targeted rather than sweeping. Gross over-correction for SAOT effects will introduce more error than it removes. The productive analytical approach is to track SAOT competition data systematically, build league-specific and competition-specific calibration, and remain attentive to tactical evolution as coaches and players continue to adapt their behaviour in response to the technology's capabilities. For a complete picture of how officiating technology and decision-making patterns shape prediction outcomes, analysts should combine this guide with the analysis provided in our referee and officiating analysis guide and the detailed treatment of VAR's effect on penalty decisions.