Start With the Two-Layer Model
The Saiki report places two numbers beside each other, but they do not have the same evidence status.
Layer 1: self-report
You answer 100 items from the HEXACO-100 item set. Saiki reverse-scores the keyed items, averages items into facets, then averages facets into six domain scores:
- Honesty-Humility;
- Emotionality;
- Extraversion;
- Agreeableness;
- Conscientiousness;
- Openness to Experience.
These values describe how you answered the questionnaire.
Layer 2: gameplay
Saiki applies hand-set anchors to selected statistics from recent League matches. It normalizes those statistics to a 0 to 100 display scale and combines them into experimental indicators.
These values are product heuristics. They are not validated personality scores, clinical measures, or population norms.
Read the Self-Report Scores
The questionnaire calculates raw domain means on a 1 to 5 response scale. The report converts each raw mean to 0 to 100 using a linear display formula:
(raw score - 1) / 4 × 100
A displayed 75 therefore represents a raw mean of 4. It does not mean you scored higher than 75 percent of people.
Saiki does not currently calculate true percentiles. A real percentile requires an appropriate reference sample, score distribution, and documented norming method. Do not label the display values as percentiles or use fixed bands such as “top 25 percent.”
When reflecting on a domain:
- read the scale description from the official HEXACO site;
- think about the situations you had in mind while answering;
- consider response style, mood, language, and context;
- avoid turning a continuous score into a permanent type;
- remember that every domain can be expressed differently across settings.
Read the Gameplay Indicators
The current rules are visible in the product code and summarized here.
| Indicator | Match statistics used | Important alternative explanations |
|---|---|---|
| Openness | Champion variety and role variety | Auto-fill, patch, team needs, deliberate specialisation, account age |
| Conscientiousness | Vision score per minute, wards per game, control wards per game | Role, support item, game length, rank, team strategy |
| Extraversion | Kill participation, kills, assists | Champion, role, team strength, game pace, play objective |
| Agreeableness | Assists, ally saves, healing | Champion kit, support role, itemisation, team fights |
| Emotionality | Deaths, KDA variance, early-surrender rate | Match difficulty, role, sample size, disconnects, team state |
Honesty-Humility has no gameplay mapping. It remains self-report only.
The app labels Emotionality as the weakest conceptual match because HEXACO Emotionality is not the same construct as in-game volatility. The same general caution applies to every row: a match statistic can have many causes.
Understand Concordance
Saiki calculates the absolute gap between the self-report and gameplay display values:
- less than 15 points:
close; - 15 to 30 points:
partial; - more than 30 points:
divergent.
These thresholds are product choices, not validated psychometric cutoffs.
A close result does not prove both values are accurate. Two methods can agree because they share a confound, because the rules were tuned that way, or by chance.
A divergent result does not prove hidden personality. It may indicate a context difference, a narrow match sample, a role effect, a noisy questionnaire response, or a poor mapping.
Use concordance to generate questions:
- Does the match period represent my usual play?
- Which statistic drove the gameplay value?
- Would the value change if I played another role?
- Does the questionnaire domain actually describe the same behaviour?
- What evidence would make me reject my first explanation?
Free and Deep Report
The current free report shows all self-report values and one fully unlocked gameplay mapping when match data exists. The remaining mapped indicators and Strategic Play panel are locked.
The €6.99 one-time Deep Report unlocks:
- all five gameplay mappings;
- the concordance view for each mapped domain;
- Strategic Play statistics for economy and tempo;
- an AI-generated narrative constrained to the report inputs;
- an Evolution chart when enough snapshots exist.
Payment does not turn an experimental indicator into a validated score. The AI narrative can summarize the inputs, but it cannot independently verify them. Evolution shows changes in heuristic indicators across stored snapshots, not personality development.
Decisions the Report Cannot Support
Do not use Saiki results to:
- diagnose a mental-health or neurodevelopmental condition;
- estimate IQ or general cognitive ability;
- screen a candidate or employee;
- make education, insurance, credit, housing, or legal decisions;
- assess another player without their informed participation;
- choose treatment;
- label a child;
- infer honesty, risk, aggression, or clinical stability from a champion or role.
If a decision has material consequences, use an instrument validated for that exact purpose and population, administered and interpreted by a suitably qualified professional where required.
The Licensing and Review Boundary
The HEXACO authors state that non-academic use of the inventory requires permission and a licence. Saiki’s repository does not contain evidence that this commercial authorization has been obtained. That status must be confirmed before this content or the paid report is deployed.
No psychologist, psychometrician, clinician, or independent research team has approved Saiki’s gameplay mappings. Until preregistered validation and qualified review exist, the honest product description is a self-report questionnaire with transparent experimental gameplay indicators.