Note
This note was transcribed by Claude.
Overview
Lecture 2 (23.02.2026) focused on why science matters for performance analysis, reviewed content from Lecture 1, and dedicated significant time to practical exercises on academic citation, literature searching, and structuring a research introduction. The session bridged Module 1 (to be completed by this date) and Module 2 (introduction to research methods, due by 9 March).
1. Why Science Matters for Performance Analysis
Four key reasons were presented:
1.1 Developing New Methods and Approaches
- The future of match analysis will not be limited to watching video. Over the next decade, video and data will be combined, leveraging tools like machine learning to identify behavioral patterns in opposition and own-team play.
- Analysts must understand data-driven methods to remain relevant.
1.2 Validating Existing Systems and Creating New Ones
- Expected Goals (xG) was highlighted as a prime example: it is a metric now widely used to assess team behavior, and it was developed and validated through scientific research.
1.3 Contributing New Concepts and Variables
- PPDA (Passes Per Defensive Action) was defined as a metric of defensive intensity — specifically, the number of passes a team allows the opposition per defensive action. This metric was created and validated by science.
- Science has established many of the technical, tactical, and physical performance indicators used in professional football today.
- Journal of Sports Sciences studies show that data can reveal intricate aspects of player performance such as decision-making under pressure, consistency in passing accuracy, and resilience in high-stakes moments.
1.4 Reducing the Risk of Invalid Decisions
- Scientific methods decrease the risk of making decisions that are not valid or evidence-based.
2. Real-World Examples of Science in Football
Benfica Case Study
- Benfica was cited as an example where scouts characterize metrics aligned with both technical excellence and adaptability, crucial for players intended for international transfers. This approach has been covered in studies from the Portuguese Journal of Sports Science.
Brentford, FC Midtjylland, and Other Data-Driven Clubs
- Clubs like Brentford and FC Midtjylland were mentioned as examples of data-driven approaches, referencing studies published in the European Sports Management Quarterly.
Liverpool FC and AI
- An article titled “The Impact of Artificial Intelligence on Football: Liverpool Football Club Case Study” was shown as another example of research contributing to team performance improvement.
AI and Data in Scouting
- The lecturer showed a web article on “How Artificial Intelligence and Data Are Shaping the Future of Scouting”, demonstrating that AI and data are increasingly central to scouting workflows.
3. Master Report Expectations
The lecturer showed an example master report from a first-edition graduate:
- The scientific study component counts for 15% of the final grade in the second year (work placement period).
- The example study compared 82 players selected for the Portuguese U19 national team vs. 82 players with similar profiles who were not selected, analyzing which key performance indicators (KPIs) differentiated the two groups.
- Findings: successful actions, total actions, successful passes, and defensive duels were among the variables that may justify a player being called up.
- Students will receive a full template document specifying expectations for each section (to be provided around September).
4. Practical Exercises (Booklet 2)
The bulk of the session was hands-on work using Booklet 2 (a Word document submitted to the lecturer at the end of class via email).
Task 1: Literature Search
- Students accessed the designated journal and searched for three articles using specific keywords:
- Offensive/attacking performance
- Playing/game style
- Technology
- The lecturer demonstrated how to use the journal’s search interface with Boolean operators (e.g., “offensive performance AND soccer”).
- Students downloaded PDFs using institutional access via UMAIA and pasted article URLs into the booklet.
Task 2: Citation Practice
Three types of citations were taught and practiced:
Indirect Citation (Paraphrase)
- Restate the idea in your own words, then cite the author and year in parentheses at the end of the sentence, before the period.
- Example: “To create goal-scoring chances, it is crucial to explore offensive actions such as penetrative passes between the defensive line of the opposing team (Tenga et al., 2017).”
- Common error: placing the period before the citation instead of after.
Direct Citation (Quotation)
- Use the author’s exact words, placed in quotation marks, with the author named in the sentence.
- Example: “Tenga et al. (2017) stated that ‘exploring the last defensive line is crucial in sport.‘”
- Must come from a different article than the one used for indirect citation.
Citation of Citation (Secondary Source)
- When you find a reference cited within another article but have not read the original source yourself.
- Example: “(Pollard & Ripp, as cited in Tenga et al., 2017).”
Task 3: Finding Different Source Types
- Students found and properly cited/referenced a book, a thesis, a web page, and a book chapter.
Task 4: Quantitative vs. Qualitative Research
- Find one article that is clearly quantitative and one that is clearly qualitative, and provide a citation for each.
Task 5: Structuring a Research Introduction
- Students read a scientific article’s introduction and identified its structural components.
- Assessment A3 requires students to:
- Identify a possible research question
- Design a list of topics and subtopics for an introduction, organized by paragraph
- Key concept: research must address a gap in the literature — something that has not been done before.
5. Key Definitions
| Term | Definition |
|---|---|
| Expected Goals (xG) | Metric assessing the probability of scoring from a given chance |
| PPDA | Passes Per Defensive Action — measures defensive intensity |
| KPIs | Key Performance Indicators — measurable variables that differentiate performance levels |
| Indirect Citation | Paraphrasing an author’s idea with (Author, Year) before the period |
| Direct Citation | Quoting exact words with attribution |
| Citation of Citation | Referencing a source found cited within another work |
| Literature Gap | An area not yet addressed by existing research, justifying a new study |
| Boolean Operators | Logical operators (e.g., AND) used to combine search terms in databases |
6. Tools and Platforms
| Tool/Platform | Purpose |
|---|---|
| International Journal of Performance Analysis in Sport | Primary journal for article searches |
| Journal of Sports Sciences | Referenced for studies on player performance data |
| European Sports Management Quarterly | Referenced for data-driven club studies |
| Portuguese Journal of Sports Science | Referenced for Benfica scouting studies |
| Institutional access via UMAIA | Accessing paywalled journal articles |
7. Assessment and Deadlines
- Module 1: Should have been completed by 23 February. Includes quizzes on journal impact factor.
- Module 2: Starts 24 February, due by 9 March. Largest module — covers introduction to research methods.
- Booklet 2: Submitted at the end of each lecture session via email.
- Assessment A3: Design an introduction structure (topics/subtopics by paragraph) for a proposed study.
- Tasks marked “Quiz” or “Complementary work” on Moodle do not count toward the final evaluation. Only tasks marked with “A” (Assessment) count.