Space evaluation at the starting point of soccer transitions

TL;DR

Space evaluation during soccer transitions is underdeveloped, especially at transition starting points. The authors introduce OBPV, extending OBSO with a Field value model and a Transition kernel model to enable pitch-wide assessment during attacking phases and transitions. Key contributions include a field-value-based spatial weighting, a KDE-based transition kernel learned from pass distributions, and empirical demonstration on La Liga 2023/24 data showing OBPV captures counter-attacking space and team-specific transition patterns. The work provides an interpretable framework for evaluating off-ball space with practical implications for coaching and opposition analysis.

Abstract

Soccer is a sport played on a pitch where effective use of space is crucial. Decision-making during transitions, when possession switches between teams, has been increasingly important, but research on space evaluation in these moments has been limited. Recent space evaluation methods such as OBSO (Off-Ball Scoring Opportunity) use scoring probability, so it is not well-suited for assessing areas far from the goal, where transitions typically occur. In this paper, we propose OBPV (Off-Ball Positioning Value) to evaluate space across the pitch, including the starting points of transitions. OBPV extends OBSO by introducing the field value model, which evaluates the entire pitch, and by employing the transition kernel model, which reflects positional specificity through kernel density estimation of pass distributions. Experiments using La Liga 2023/24 season tracking and event data show that OBPV highlights effective space utilization during counter-attacks and reveals team-specific characteristics in how the teams utilize space after positive and negative transitions.

Figures (8)

• Figure 1: Overview OBSO and OBPV. The attacking and defensive players are represented in red and blue, respectively, and the ball is represented in black. The original point (0,0) is the center of the field. One of the attacking players holds the ball near the center and attacks from left to right (x-axis). These models consist of three components: the Score model, PPCF, the Transition model for OBSO, and the field value model, PPCF, the Transition kernel model for OBPV. OBSO shows low evaluations in this pitch. OBPV allows for a more pitch-wide comprehensive assessment.
• Figure 2: The distribution of transition model and Number of passes used for kernel density estimation. (Left) The distribution of transition kernel model is shown in each area. The direction of attack is to the right. (Right) Number of passes used for kernel density estimation is displayed. The dataset includes mirrored data, so the number of passes used is twice the actual number of passes made. In Area 17, the number is lower because players tend to opt for shooting rather than passing.
• Figure 3: The correlations of OBPV increase during positive and negative transitions and other statistics. (Left) The relationship between OBPV increase during positive transitions and the number of passes per sequence LaLigastats is shown. (Right) The relationship between the increase in OBPV during negative transitions and the number of balls won within 40 meters of the goal LaLigastats is shown.
• Figure 4: The comparison of OBPV and OBSO, and two similar situations but different OBPVs. (Left) OBPV and OBSO value histograms for the events evaluated in Section \ref{['ssec:Transition_OBPV']}. (Right) Two similar example situations but different OBPVs are shown. In both examples, the red team is attacking from left to right, with the blue team defending. The black circle represents the ball. In these two scenes, OBSO scores were very similar: 0.01816 and 0.01817 (both red #33). However, OBPV evaluations differed greatly: 0.29968 (upper: red #33) and 0.50267 (lower: red #20).
• Figure 5: Sigmoid function and field value model. (Left) The sigmoid function is applied along the X-axis of the pitch. (Right) Field value model when attacking from left to right. Darker red indicates higher importance, while areas closer to white represent lower importance. The weight decreases as the position moves away from the goal along the length of the pitch, and also as it moves toward the sidelines. The area adjacent to the penalty box, marked in green, and the vital area, marked in blue, are assigned higher values.