Epistemic Certainty in Sports
This essay poses a hypothetical and offers solution-frames to discuss epistemic certainty in scouting and recruitment.
Introduction: A Case Study
A striker is through on goal. They've beaten the off‑side trap, have no defenders on their tail, and only the goalkeeper stands between them and the back of the net. Four common finish selections present themselves:
- Dribble past the keeper and roll the ball into the net.
- Shoot early to catch the keeper off‑guard.
- Feign and nut‑meg the keeper.
- Chip ('dink') the ball over the keeper.
Our striker chooses one of these options. And misses. It is not the first time. Viral highlight reels document a season of squandered one‑on‑ones.
As the recruitment executive of a Champions‑level club YOU receive a dossier from your analysts. Inside is a file labelled '1v1s' – a super‑cut of every clear one‑on‑one the striker has faced during the last three seasons.
The dossier forces a series of epistemic questions:
- What can we infer about this player's finishing ability?
- Is it reasonable to conclude they are generally poor in 1‑on‑1 situations?
- How strongly can we project their performance into a different league or tactical context?
- How certain can we be about any of the above?
- What additional information would increase or decrease that certainty?
This essay explores those questions and, more broadly, epistemic certainty in scouting and recruitment.
Epistemic vs Psychological Certainty
Epistemic Certainty
In analytic epistemology, epistemic certainty describes the strength of justification for believing a proposition is true. It is a function of evidence, logical coherence, methodological soundness, and the absence of defeaters.
Psychological Certainty
Psychological certainty is a feeling of conviction, which may or may not track epistemic warrant. A coach convinced a striker will 'come good' might possess high psychological certainty yet low epistemic warrant if the data say otherwise.
Two Analytical Frames for 1‑on‑1 Evaluation
There are two frames we can employ to attempt the aforementioned questions. (See footnotes for my earlier scraps about this.)
| Frame | What it looks at | Core Question | Typical Tools |
|---|---|---|---|
| Frame 1: Outcome‑based Mechanics | Observable execution of the chosen technique (post‑hoc). | “How did the attempt play out?” | Video tagging, biomechanics breakdown, xG models |
| Frame 2: Intention‑based Decision Analysis | Underlying choice architecture that produced the attempt. | “Why did the player choose that option?” | Cognitive task analysis, VR replay & interview, eye‑tracking data, coach-player debriefs |
Strengths & Weaknesses
Frame 1 maximizes measurability and thus epistemic certainty but abstracts away internal decision processes.
Frame 2 accesses intent, useful for coaching and player self‑reflection, but relies on inaccessible mental states, lowering epistemic warrant.
Choosing a Frame
For forecasting (transfer recruitment, salary negotiation) we prioritize verifiable, mechanically‑grounded evidence: Frame 1 becomes primary; Frame 2 remains complementary for player development programs.
Technique, Mechanics, and Decision‑Making
- Technique (How): The specific motor pattern selected (e.g., inside‑foot placed finish).
- Decision‑Making (Why): The rapid perceptual‑cognitive process that selects a technique given constraints.
- Mechanics (What Happened): The physical outcome: ball trajectory, keeper reaction, result. This is a pivot from the usual frame I had previously published on this blog before.
While stats and technique tell us what a player does, it’s their mechanics that explain why and how they can consistently perform.
— Joel A. A. (@JoelAdejola) October 7, 2024
I break down how mechanics help scouts and coaches forecast future success in ways that numbers alone can’t.https://t.co/ud9o0ejOIW
Factors Modulating Epistemic Certainty
Back to the last two questions:
How certain can we be about any of the above?
What additional information would increase or decrease that certainty?
Here are some simple attempts at these:
| Increases Certainty | Decreases Certainty |
|---|---|
| Larger sample size of 1‑on‑1s | Small sample (highlight bias) |
| Contextual data (xG, shot location, goalkeeper positioning) | Context change (new league tempo, defensive spacing) |
| Multi‑season trend stability | Tactical system mismatch |
| Biomechanical metrics (approach speed, body shape) | Psychological volatility (confidence swings) |
| Training data & coach testimony | Incomplete or low‑quality video |
So, Should YOU Sign the Striker?
- Aggregate Evidence: Three‑season 1v1 conversion rate vs (outgoing) league average.
- Context Adjustment: Compare quality of chances (post‑shot xG).
- Model Forecast: Bayesian update incorporating club tactical fit.
- Residual Uncertainty: Highlight unknowns (injury history, adaptation to higher tempo).
Finally, provide a probability estimate rather than a binary verdict.
Epistemic Humility in Talent ID
Absolute certainty is unattainable: the goal is to bound uncertainty tightly enough that decisions are rational under risk.
Combining a mechanics‑first frame with selective insights from intention analysis provides the highest defensible warrant when millions - and trophies and jobs - are at stake.
Related essay:
Joel A. Adejola
Related thread (where this essay spawned from):
You’re amazing.
— Joel A. A. (@JoelAdejola) October 7, 2024
The entire reason why this article idled for so long — and even this morning while I wrote — was because I couldn’t completely justify one frame (how-why) over the other.
I ultimately chose Frame 1 because, while technique is all-encompassing, for the sake of…
Who is the Writer?
Joel A. Adejola is an undergraduate at the University of Kansas (KU), studying Engineering and Philosophy.
