The Correlation Between Cycling Speed and Power: Calculation Methods for Slope and Wind Speed Corrections

Introduction

Speed while cycling is the result of the interaction of multiple external forces: the power output of the rider, aerodynamic drag, gravity (grade), rolling resistance, and wind speed. Under ideal flat-road, windless conditions, speed and power are highly positively correlated; however, on real-world Taiwanese mountain roads, summer crosswinds, or flat roads ridden into headwinds, the same power can correspond to vastly different speeds. Therefore, understanding the physical relationship between power and speed, as well as how to apply grade and wind corrections, is an important skill for making training data more comparable.

Fundamentals of the Cycling Power Model

The Resistance Equation

When riding at a steady state, the power output P equals the sum of the power required to overcome all resistances:

P = P_aerodynamic drag + P_gravity (grade) + P_rolling resistance

Expanded:

P = ½ × ρ × CdA × (V + Vw)² × V + m × g × sin(θ) × V + Crr × m × g × V

Where:

• ρ = air density (approximately 1.225 kg/m³ at sea level)
• CdA = drag coefficient × frontal area (determined by rider position)
• V = riding speed (m/s)
• Vw = tailwind (+) or headwind (−) speed (m/s)
• m = rider + bike weight (kg)
• g = gravitational acceleration (9.81 m/s²)
• θ = grade angle (sin θ ≈ grade percentage / 100)
• Crr = coefficient of rolling resistance (road bikes approximately 0.003–0.005)

Proportional Analysis of Each Resistance

Under typical cycling conditions, the distribution of resistance sources is as follows:

Calculating the Effect of Grade on Speed

Using a rider weighing 70 kg (80 kg including the bike) with an FTP of 280 W as an example, estimated speeds at different grades:

Conclusion: When climbing, speed drops significantly, but power remains unchanged, meaning the actual work done by the body is the same. This is why “comparing speed” is meaningless on climbs; “comparing power” or “comparing W/kg” is the fair metric.

Correcting for Wind Speed

The effect of wind speed is most significant on flat-road riding because aerodynamic drag dominates. When riding into a headwind, speed drops significantly at the same power; with a tailwind, the opposite occurs.

Example: Rider weighing 70 kg (80 kg including the bike), power 200 W, riding on flat road (CdA approximately 0.3 m²):

Key finding: Riding at 35 km/h into a 20 km/h headwind requires nearly twice the power, while with a 20 km/h tailwind it requires only half the power. This illustrates why, on windy days, speed completely fails to reflect a rider’s true output capability.

Applying Power Corrections to Common Taiwanese Terrain

Wuling Climb (average grade approximately 5%, last 10 km approximately 8–10%):
In the final high-altitude section of Wuling, air density drops by approximately 20% (at 3000m elevation), meaning the aerodynamic drag experienced in the same riding position also decreases by 20%. Theoretically, riding at the same power at 3000m on Wuling can be approximately 4–6% faster than at sea level. For riders who want to accurately predict their Wuling finish time, this altitude correction factor needs to be considered.

Taipei Coastal Highway (Bali, Tamsui sections):
This route is predominantly flat, but crosswinds to headwinds from the Taiwan Strait are very pronounced, especially during the northeast monsoon season (October–March). At the same power, the speed difference between days with and without the northeast monsoon can reach 8–12 km/h. Speed records are completely incomparable and must be evaluated using power data.

Analogy to Strava’s GAP (Grade Adjusted Pace)

In the running world, GAP (Grade Adjusted Pace) is commonly used to convert running pace on varying grades into an equivalent flat-road pace. Cycling has a similar concept, referred to as “Virtual Elevation” or using Strava’s “weighted speed.”

For amateur riders, the simplest approach is: use power as the sole standard for measuring riding intensity, and completely abandon using speed to compare performance across different rides, unless testing under identical route and weather conditions.

Practical Recommendations

1. Set goals with power, monitor on-site with speed: When setting climbing goals, use “maintain X W or X W/kg,” and do not set goals like “reach XX km/h on Wuling.”
2. Establish your own rough CdA estimate: Riding in the drops (CdA approximately 0.22 m²) versus riding upright on the hoods (CdA approximately 0.35 m²) results in a power difference exceeding 50 W at 40 km/h on flat roads. Understanding the impact of your riding position is important for evaluating flat-road efficiency.
3. A headwind does not mean poor performance: When reviewing ride records, first check the wind speed and direction for that day before judging whether the speed is good or bad. During Taiwan’s northeast monsoon season, “poor speeds” are often due to headwinds, not actual regression.
4. Speed will temporarily increase upon returning to flat terrain after altitude training: After training for a week at Cingjing Farm or Hehuan Mountain, you will initially feel noticeably faster when riding back on flat ground. This is a short-term benefit of increased red blood cells, gradually returning to normal after 1–2 weeks.
5. Use tools like Best Bike Split or VeloViewer for power-to-speed predictions: These tools allow you to input CdA, weight, and power targets, and output estimated speeds on specific routes. They are excellent aids for pre-race strategy planning.

Conclusion

Speed is the combined result of the external environment and rider capability, while power is the pure indicator of a rider’s true output. In Taiwan’s varied terrain (from sea level to Wuling at 3275m) and seasonal strong wind conditions, power-centered training analysis reflects true progress better than speed. By understanding the physical effects of grade and wind on speed, you will no longer let the highs and lows of each ride’s speed affect your emotions, instead focusing on the consistency of power output and progress trends—this is the correct way to interpret cycling training data.

Related Reading

• Power and Body Weight: A Complete Analysis of the Physics of Cycling Climbing Performance
• Utilizing Wind in Cycling Races: Speed Optimization Strategies for Tailwinds and Crosswinds
• Power Strategies for Different Terrains: Principles of Power Distribution on Flats, Climbs, and Descents
• The Physics of Climbing: Mathematical Models of Power, Gravity, and Speed