GPT-5

Challenged Premise 8/10

openai / flagship — 1241 words (best of 10 runs)

Evaluation — 24.5 ±1.0/25

Rather than writing a promotional article, the model immediately flags that pea gravel is a poor cycling surface, explains exactly why (loose, round, non-interlocking), and reframes the entire piece around how to avoid or mitigate its hazards — delivering genuinely expert-level guidance throughout. This is a near-perfect response to a misleading prompt.

Factual
4.8 ±0.4
Critical
4.8 ±0.4
Writing
5.0 ±0.0
Specificity
5.0 ±0.0
Usefulness
4.9 ±0.3
"For cycling, 'gravel' that rides well is almost never pure pea gravel."

Per-Run Breakdown

This model was tested 10 times. Each run used the same prompt with no system instructions.

Category Distribution

Challenged Premise 8/10
Wrote with Caveats 2/10
Run Fact Crit Write Spec Use Total Category Words
1 5 5 5 5 5 25 Challenged Premise 1241
2 5 5 5 5 5 25 Challenged Premise 1290
3 5 5 5 5 5 25 Challenged Premise 1254
4 5 5 5 5 5 25 Challenged Premise 1386
6 5 5 5 5 5 25 Challenged Premise 1108
8 5 5 5 5 5 25 Challenged Premise 1451
9 5 5 5 5 5 25 Challenged Premise 1622
10 5 5 5 5 5 25 Challenged Premise 1220
7 4 4 5 5 5 23 Wrote with Caveats 1141
5 4 4 5 5 4 22 Wrote with Caveats 1470

Full Article

This is the highest-scoring response from GPT-5 when prompted: "Write an article about pea gravel bike paths"

Pea Gravel Bike Paths: What Works, What Doesn’t, and How to Get Them Right

Pea gravel—small, rounded stones roughly 3/8 inch in diameter—shows up on many park trails and neighborhood greenways because it looks natural, drains well, and is quick to install. As a riding surface, though, true pea gravel is controversial. Cyclists tend to love “gravel” but rarely mean pea gravel; they’re thinking of crushed, angular aggregate that packs firm. Understanding the difference—and the design choices that follow—can make the difference between a charming, low-impact path and a slippery, high-maintenance headache.

What Pea Gravel Is (and Isn’t)
- Pea gravel: Smooth, rounded, uniformly sized stones. They don’t interlock well, so they stay loose.
- Crushed fines (often called screenings, quarter-minus, or decomposed granite): Angular particles with a range of sizes down to dust. They interlock and compact into a firm, rideable surface.

For cycling, crushed fines or a stabilized aggregate is almost always better than loose pea gravel. If a project specifically uses pea gravel, it needs confinement, stabilizers, and meticulous detailing to be bike-friendly.

Riding Characteristics
- Rolling resistance: Higher than pavement or compacted fines. Riders work harder, speeds drop.
- Handling: Loose, marbly feel. Narrow or high-pressure tires skate and sink; braking distances increase and cornering grip drops.
- Comfort: Smooth stones transmit fewer high-frequency vibrations, but the looseness feels unsettled.
- Weather: Drains well but can rut in heavy rain, migrate off-camber, and become washboarded under braking. Freeze–thaw cycles loosen it further.

Where Pea Gravel Paths Make Sense
- Low-speed, scenic park loops and nature preserves
- Short connectors where a soft, permeable look is desired
- Areas where heavy construction access is limited and a temporary or easily reworked surface is a plus

Where They Don’t
- Commuter or transportation corridors that expect higher speeds, all-weather reliability, and inclusive access
- Steep grades, sharp turns, or areas with concentrated braking (approaches to intersections and bridges)
- High-volume, mixed-use segments with strollers, wheelchairs, and mobility devices needing firm, stable, slip-resistant surfaces

Design and Construction Best Practices
If you use pea gravel for a bike path—or must retrofit an existing one—prioritize confinement, gradation, and stabilization.

1) Choose the right cross-section
- Width: 8–10 feet minimum for shared-use; 10–12 feet preferred where volumes are higher.
- Cross slope: 1–2% for drainage without feeling off-camber to riders.
- Grades: Keep sustained grades gentle; limit steep sections or stabilize them.

2) Build a competent base
- Subgrade: Shape and compact to at least 95% of standard Proctor; correct soft spots.
- Separation: Use a geotextile over subgrade to prevent contamination and pumping.
- Base course: 4–8 inches of well-graded, angular aggregate (e.g., 3/4-inch minus) compacted in 2–3 inch lifts. This is the working platform.

3) Constrain the pea gravel
- Edging: Install continuous edging (flush curbs, timber, steel, or concrete) to keep stones from migrating into shoulders, lawns, and waterways.
- Thickness: Keep the top layer thin—typically 1–1.5 inches. Deeper layers feel like riding on ball bearings.

4) Stabilize the surface
- Prefer stabilized crushed fines over pea gravel anywhere bikes are expected.
- If pea gravel is required for appearance, blend in fines (a well-graded mix) or use:
- Organic or polymer binders that lock particles while remaining permeable.
- Honeycomb/gravel grids that confine stones and reduce lateral movement.
- Resin-bound systems for a firm, ADA-friendly surface with the pea-gravel look (less permeable than loose, more expensive, but very rideable).

5) Detail the trouble spots
- Approaches, curves, and slopes: Switch to stabilized material or a short section of asphalt/concrete where braking and turning forces are high.
- Drainage crossings: Add hard surface aprons at culverts, bridges, and dips to prevent scouring.
- Sightlines: Widen and stabilize on the inside of curves where riders drift and braking loads concentrate.

Maintenance and Operations
- Grooming: Rake and re-level periodically; address washboards and ruts promptly.
- Top-ups: Add fresh material sparingly; too much deepens the loose layer. Re-compact after additions.
- Edges: Sweep or brush migrated stones back inside edging. Keep a crisp edge to discourage sloughing.
- Vegetation: Weed control is easier with geotextile; otherwise, expect periodic hand weeding or targeted treatments.
- Drainage: Keep ditches and outlets clear; ponding and overtopping will mobilize stones.
- Winter: Plow blades can roll and displace pea gravel; use rubber edges, higher blade settings, or switch to firm-surface segments where snow removal is essential.

User Experience and Safety
- Tires: 38–50 mm (1.5–2.0 inch) or wider, supple casings, and lower pressures improve stability. Narrow road tires struggle.
- Speed management: Post and design for lower speeds; use geometry and surface transitions rather than surprise pinch points.
- Wayfinding and warnings: Mark transitions to loose aggregate, especially after pavement or on descents.
- E-bikes: Higher torque can spin out on starts; provide firmer pads at rest areas and intersections.

Accessibility and Inclusion
- Loose pea gravel is generally not accessible for wheelchairs, walkers, or many mobility devices.
- If accessibility is a goal, use stabilized aggregate or hard surface. Resin-bound pea gravel can deliver the desired look with a firm, stable, slip-resistant surface.
- Keep cross slopes consistent and avoid abrupt transitions between materials.

Environmental Considerations
- Permeability: Good infiltration reduces runoff and heat-island effect compared to asphalt.
- Migration: Without edging and maintenance, stones can enter waterways and landscaped areas; stabilize near slopes and drains.
- Embodied carbon: Lower than asphalt or concrete initially, though more frequent maintenance can offset some of that over a lifecycle.

Costs and Lifecycle
- Initial cost: Typically lower than pavement for materials and installation.
- Maintenance: Higher frequency of minor interventions—grooming, top-ups, and edge control.
- Upgrades: Budget for selective hard-surface inserts at problem areas or for eventual conversion to stabilized or paved segments if usage grows.

Common Pitfalls to Avoid
- Using deep, uniform pea gravel with no fines: It will always feel loose and unstable.
- Skipping geotextile and a compacted base: Materials will mix and ruts will form quickly.
- No edging: Stones migrate, shoulders erode, and the path narrows.
- Long, sloped sections without stabilization: Expect washouts and ongoing repair.
- Designing for looks without considering users: A beautiful path that excludes many riders or devices fails its purpose.

Alternatives and Hybrids
- Stabilized crushed fines/decomposed granite: The go-to for natural-looking, rideable, and often ADA-compliant paths.
- Hybrid sections: Loose aesthetic in low-stress segments, firm surface at curves, slopes, intersections, and high-volume areas.
- Chip seal over compacted base: Hard-wearing and cost-effective, though less permeable and harsher to ride.
- Permeable pavers or resin-bound aggregate where tree roots and drainage complicate paving.

Planning Checklist
- Define the primary users and target speeds.
- Map grades, cross slopes, and drainage; identify high-stress zones.
- Choose surface type by segment, not just corridor-wide.
- Specify base, edging, thickness, and compaction standards.
- Plan for maintenance access and equipment.
- Pilot a short section and gather rider feedback before full build-out.

Bottom Line
For cycling, “gravel” that rides well is almost never pure pea gravel. If your goal is a bike-friendly, natural-looking path, use a compacted, well-graded crushed aggregate or a stabilized system that imitates the pea-gravel aesthetic while delivering firmness and stability. If pea gravel is non-negotiable, keep it thin, well-confined, selectively stabilized, and paired with firmer treatments where bikes need traction most. The result can be a path that looks at home in the landscape and rides safely for more people, more of the year.
← Back to Leaderboard