Entem is a sports analytics solution, aiming to not only provide coaches with relevant insights to improve game outcomes, but to also facilitate understanding on what leads to peak performance.

The Problem

Oftentimes on the playing field, coaches are in high pressure situations where they need to intuitively make a decision quickly that fosters positive outcomes. Seasoned coaches are able to efficiently know what to focus on and what went wrong during game performance. However, a common barrier is that coaches do not have concrete evidence or data to back up their intuition or gut feelings.

Not having data to drive decisions that affect individual athletes can prompt conversations that make them feel awkward or under-valued. Having certain statistics to support coaches’ decision-making can back up conversations with athletes, empowering both parties and making them feel confident during game time, thus further enhancing performance and outcomes. This is why sports analytics solutions are popular, as they can provide the insights that coaches could not have had. While most solutions out there measure physical factors, many coaches will agree that emotions are indeed the driving force behind game performance. 

Research in sports psychology strongly supports this notion of emotions having a critical role in sport and exercise (Hanin, 2000) and being the primary motivational system (Izard, 1993). Since emotions can affect athlete performance, using these insights can support and accelerate coaches’ strategic decisions, ultimately strengthening outcomes.

The Solution 

To bridge the gap between making efficient, robust decisions in a fast-paced context and ensuring positive game impact, we researched and developed a solution that utilizes the primary driver behind game performance as the key variable: emotions. This approach is different from existing solutions in that we measure the predictor and antecedent variable of game engagement and performance, rather than the outcomes. For instance, common sports analytics solutions will measure physical factors such as speed or distance. Emotions play a vital role in these factors. Components of emotion include: physiological changes, action tendencies, and subjective experience (Deci, 1980; Young, 1973).

The Role

I conducted user research, conducted literature review, collaborated with sports coaches, ideated and evaluated the product features, and designed the high-fidelity mockups before sending off to the development team.

• I led the research, development, and ideation of key product features in addition to the emotional model framework.

• I created the sketches,  wireframes, low-fidelity mockups, and defined the technical requirements and product rules.

• I collaborated with the UI Designer to design the mockup and high-fidelity prototype.

• I worked with Backend Engineers and Data Scientists to brainstorm what were feasible features for presenting the data.

  High Level Goals

• Make sure the insights are easy to interpret, no complicated data

• Make it clear how managers where managers should focus on improving and take action

• User research - I chose to do user interviews to have a better understand of how coaches make decisions during games and practice sessions, and what kind of insights they would find useful

• Literature review - I wanted to do more research about the role of athletes' emotions and how it contributes to game performance and other factors

Based on primary research, there is a strong consensus from sports coaches and performance directors that emotions are the primary driver behind game performance, can make or break the game. Physical performance is contingent on the mental and emotional.

★ Insights 

Understanding emotions during game performance can help coaches:

• Facilitate tactical decision-making during high pressure game context and development of game strategy

  • Example: If you know your team is emotionally high, you can adjust defensively vs offensively

  • Example: Understand how to motivate the team, especially for offense

• Understand team dynamics and how to improve team culture for optimal performance

  • Example: Player energy and emotions from sidelines can affect game performance, affect people who sees it

• Know whether the team is where they need to be emotionally during game and adjust coaching accordingly

• Identify the drivers and impact of fan engagement on game performance

  • Example: "I would make the argument that if we have 3,000 people watching our game, our girls would play better 100%, but I can't prove that." - Richard Moodie

• Have insight to individual players and facilitate player development and growth for long-term, especially since every player deals with emotions differently

Understanding emotions during game performance can help players:

• Be aware of their emotions for individual player development and its impact on team dynamics and culture

  • Example: Overall team emotions were low because individual level was negative and it spreads

• Understand the role of their emotions on game performance

  • Example: better manage emotions on the field, such as not losing temper

• Our proposed framework is based on the IZOF model (Hanin, 1978, 1983a, 1986, 1989,1995,1997a, 1997b), which attempts to predict individual performances based on emotion states 

• The IZOF Model 

  • Has been extensively tested and validated across cultures, sports, and ages, and between genders  (Gould, Tuffey, Hardy, & Lochbaum, 1993; Hyvönen, 1992; Krane, 1993; Morgan, O'Connor, Ellickson, & Bradley, 1988; Pons, 1994; Prapavessis & Grove, 1991; Raglin, Morgan, & Wise, 1990; Salminen, Liukkonen, Hanin, & Hyvönen, 1995; Turner & Raglin, 1996)

  • Examines two major functional aspects in emotion-performance relationships: energizing (de-energizing) and organizing (disorganizing) effects of emotion upon performance

  • Is based on results of systematic observation of top-performers in real-life contexts

• Emotion influences performance and in turn performance affects emotion content and intensity

• Research has emphasized how and why athletes experiencing different levels of emotion are consistently successful or unsuccessful

• IZOF represents:

  • Individual - unit of analysis within context of high-achievement competitive sport

  • Zones - implies specific relationship between perceived intensity of optimal and dysfunctional emotional states and quality of a performance

  • Optimal - optimal emotions are those most relevant and appropriate for particular athlete under specific conditions

  • Functioning - refers to specific optimal (dysfunctional) effect of emotion upon the quality of performance process

1. Playback: Video playback of games or practices with AI-based analysis of emotions and body language allows coaches and players to see non-verbal cues and performance in real time. Research in sports psychology shows that visual feedback helps athletes gain self-awareness and understand their actions more clearly, facilitating targeted improvement (Carroll & Bandura, 1982). By capturing emotional expressions, coaches can address subtle cues like body language and positioning to optimize performance.

2. Playback insights: Highlighting how individual players contribute to team dynamics through positive or negative actions encourages accountability and strengthens team cohesion. Sports psychology suggests that acknowledging positive behaviors can increase motivation, while constructive feedback on negative behaviors promotes resilience (Weinberg & Gould, 2014). These insights align with the UX principle of actionable feedback, as coaches and players receive direct, observable data.

3. Line graph: A line graph displaying dimensions such as energy, confidence, engagement, and control over time aligns with the sports psychology concept of tracking mental and physical states to identify trends in performance (Hanin, 2000). Real-time visualization provides insights into how each factor fluctuates during gameplay, allowing coaches to adjust strategies based on athletes' peak and low moments.

4. Radar chart: The radar chart gives a holistic view of an athlete’s emotional and physical dimensions at a glance, valuable for assessing multi-faceted performance (Eccles & Tenenbaum, 2007). Displaying these dimensions in real-time facilitates quick adjustments based on dynamic changes in athlete states, supporting an adaptive coaching approach.

5. Individual athlete profile: Providing historical performance data allows coaches to tailor training programs and track progress over time. According to sports psychology, monitoring patterns in emotional states and performance aids in creating personalized interventions, increasing the athlete’s sense of control and motivation (Locke & Latham, 1990).

6. Simulator: The simulator empowers coaches to test various team compositions and predict team dynamics, enhancing strategic planning. Sports psychology suggests that simulations can help athletes and coaches anticipate outcomes, manage anxiety, and improve decision-making under pressure (Feltz & Landers, 1983). This feature aligns with predictive UX by allowing users to simulate outcomes based on historical data.

7. Athlete impact ranking: Ranking players based on their impact (positive or negative) fosters a competitive atmosphere and helps identify areas for individual improvement. Sports psychology research highlights that ranking systems can enhance intrinsic motivation by providing athletes with specific, measurable goals (Ryan & Deci, 2000).

Our solution draws on the relevant attributes for peak performance so coaches can see at a glance how their team is doing, both at the individual and collective level. 

With supporting research, a robust framework, and cutting-edge technology, our team is confident in generating the necessary meaningful insights coaches seek to guide their athletes to optimal game performance. 

Our solution was eventually reviewed and positively received by senior executives at the Boston Red Sox, which we are currently working on a partnership with.