This commit is contained in:
2025-12-01 22:01:58 +01:00
parent 086fdc1ea1
commit f717abf0a0
14 changed files with 3305 additions and 1780 deletions

224
docs/concept/RATING.md Normal file
View File

@@ -0,0 +1,224 @@
# GridPilot Rating
*Our own competition rating system designed for league racing — because iRating was never made for this.*
---
# 1. Why GridPilot Needs Its Own Rating
iRating works great for official races, but it's not built for leagues.
**League racing is fundamentally different:**
- you race against the same people each week
- racecraft matters more than raw pace
- team dynamics influence outcomes
- attendance and consistency matter
- incidents have real consequences
- formats vary massively between leagues
- results depend heavily on season structure
- small grids distort iRating logic
iRating cant measure:
- season performance
- team contribution
- consistency
- clean driving
- decision-making
- rivalry strength
- long-term commitment
- reliability
So we build a rating *specifically for league racing*
where context, progression, and human rivalry actually matter.
---
# 2. What GridPilot Rating Measures
GridPilot Rating = **league racing performance + racing behavior + reliability**.
It captures the full picture of a driver, not only their pace.
### **Core components**
- **Results Strength**
How well you finish relative to the field strength.
- **Consistency**
Stability of finishing positions over a season.
- **Clean Driving**
Incidents per race, weighted by severity.
- **Racecraft**
Positions gained/lost vs. incident involvement.
- **Reliability**
Attendance, DNS/DNF record.
- **Team Contribution**
Points earned for your team; lineup efficiency.
### **Why these matter**
A great league driver is not only fast.
A great league driver is:
- clean
- consistent
- reliable
- competitive
- predictable
- respected
- valuable to teams
Our rating reflects **real league value**, not just raw pace.
---
# 3. Principles of GridPilot Rating
## **1. Competition over matchmaking**
iRating is designed for matchmaking.
GridPilot Rating is designed for **championships**.
## **2. Context matters**
Beating a strong field → worth more.
Clean races → rewarded.
Incidents → penalized.
Attendance → essential.
## **3. No punishing drivers for leagues with small grids**
In iRating, small splits cause unstable jumps.
GridPilot Rating normalizes for field size.
## **4. No obsession with “perfection”**
This is not a chess ELO.
This is racing.
Racing has chaos.
Rating should reflect that.
## **5. Transparent & understandable**
Drivers should always know:
> “Why did my rating change?”
Our UI makes this clear.
---
# 4. Rating Inputs (Conceptual, not equations)
GridPilot Rating is built from weighted contributions:
### **1. Race Placement Score**
Normalized by:
- field size
- field strength
- track difficulty (optional future)
- gaps between positions
### **2. Incident Factor**
Scaled to:
- incident count
- impact on others
- severity
- penalty involvement
### **3. Reliability Score**
- attendance
- quitting early
- DNS
- DNF
### **4. Team Factor**
- points contributed relative to team total
- team finishing efficiency
- team strength
### **5. Season Factor**
Bonus for:
- championship position
- improvement
- performance under pressure
- completing the full season
Everything is transparent and cumulative.
---
# 5. Why Teams Benefit from Rating
Teams need a way to scout drivers that isnt:
- “Do you have 3k iRating?”
- “Are you fast in officials?”
- “Do you join our Discord?”
GridPilot Rating shows:
- reliability
- cleanliness
- consistency
- contribution
- teamwork
- multi-season performance
Thats what teams care about in real league competition.
---
# 6. Why Admins Benefit
Admins get:
- balanced rosters
- fairer splits if needed
- data for assigning classes
- better recruitment
- fewer problems with unreliable drivers
- objective metrics behind decisions
Rating turns leagues into **professional, predictable competitions**.
---
# 7. Why Drivers Benefit
Drivers finally get:
- recognition for more than pace
- a way to show reliability
- stats that show improvement
- incentives to race clean
- a reason to finish the season
- meaningful identity across multiple leagues
GridPilot Rating makes league racing **feel like a true career**, not a collection of spreadsheets.
---
# 8. How It Differs From iRating
| iRating | GridPilot Rating |
|--------|------------------|
| Matchmaking tool | Competition identity tool |
| Random opponents | Same rivals weekly |
| No context | Context-heavy |
| Pace only | Pace + behavior + consistency + reliability |
| Does not reward team play | Team contribution included |
| Penalizes small grids | Normalized for field size |
| No season meaning | Season progression core to system |
| Binary win/loss logic | Full racing context |
GridPilot Rating doesnt replace iRating.
It fills the gap iRating never tried to solve.
---
# 9. Why We're Building This
Because league racing deserves:
- fair recognition
- long-term meaning
- personal progression
- team value
- traits beyond speed
- a competitive environment that feels like real motorsport
No other platform gives league racers their own rating.
And no existing metric truly represents what makes someone great in a league.
GridPilot Rating changes that —
**by measuring the things that actually matter in league racing.**