What Actually Makes a Car Fast?

A Data Analysis of 176 Vehicles from 2013-2019

Introduction

I originally started this project after losing an impromptu race to a Ford Mustang EcoBoost while driving my 2018 Honda Accord Sport 1.5T 6MT.

Like many automotive enthusiasts, I assumed the answer was simple: the Mustang had more horsepower.

But as someone actively developing data analysis skills, I wondered whether horsepower alone truly explained vehicle acceleration. Rather than rely on opinions and internet debates, I decided to build a dataset of 176 vehicles produced between 2013 and 2019 and analyze the factors most closely associated with 0-60 performance.

The project ultimately evolved into a broader question:

What actually makes a vehicle fast?

The Dataset

Vehicles Analyzed

176 vehicles

Years

2013-2019

Variables Collected

• Horsepower

• Torque

• Curb Weight

• Drivetrain

• Transmission

• Engine Displacement

• Forced Induction

• Wheelbase

• Front Tire Width

• Rear Tire Width

• Quarter Mile Time

• Horsepower Per Ton

• Torque Per Ton

• 0-60 Time

Vehicle data were manually aggregated from manufacturer specifications and publicly available performance testing sources. Vehicle specifications were compiled with assistance from Google Gemini and cross-referenced against multiple automotive reference sources, including Edmunds, MotorTrend, U.S. News & World Report Cars, MotorWeek, Cars.com, Kelley Blue Book (KBB), and official manufacturer websites.

Performance metrics for 0–60 mph acceleration and quarter-mile elapsed times were obtained from ZeroTo60Times.com. The vehicle model year range was intentionally constrained to facilitate comparisons among vehicles developed within a similar technological era, thereby reducing variability attributable to major advancements in automotive engineering across different generations.

To maintain consistency within the dataset, only vehicles equipped with conventional geared transmissions were included. Vehicles utilizing continuously variable transmissions (CVTs) or other non-geared transmission technologies were excluded. Additionally, the analysis was limited to vehicles powered exclusively by internal combustion engines (ICEs), excluding hybrid-electric and fully electric vehicles.

Hypothesis

Many enthusiasts believe horsepower is the primary factor influencing acceleration.

This project tested whether horsepower was truly the strongest predictor of vehicle performance or whether other factors better explained acceleration.

Methodology

To identify which characteristics most strongly influence acceleration, a correlation analysis and regression analysis were performed using vehicle specifications and 0-60 times.

The strength of each relationship was measured using:

• Correlation Coefficients

• Scatter Plots

• Trendlines

• R² Values

Key Findings

Finding #1: Torque Per Ton Was the Strongest Predictor

Variable: R²

Torque Per Ton: 0.699

Approximately 70% of the variation in acceleration performance could be explained by torque per ton alone.

This was the strongest relationship identified in the dataset.

Finding #2: Power Per Ton Outperformed Raw Horsepower

Variable: R²

Torque Per Ton: 0.699

Horsepower Per Ton: 0.661

Horsepower: 0.645

Rear Tire Width: 0.501

The analysis showed that weight-adjusted power metrics consistently outperformed horsepower alone.

This suggests that vehicle performance is influenced by the relationship between power and mass rather than raw engine output.

Finding #3: Rear Tire Width Matters

Rear tire width produced an R² value of 0.501.

While tire width alone does not make a vehicle faster, high-performance vehicles consistently featured wider rear tires, suggesting traction plays a meaningful role in acceleration performance.

Finding #4: Engine Size Was Less Important Than Expected

Displacement showed a weaker relationship with acceleration than horsepower-per-ton and torque-per-ton.

This finding supports the modern trend toward smaller turbocharged engines producing performance levels once reserved for larger naturally aspirated engines.

The Accord vs Mustang Question

After identifying the strongest performance predictors, the analysis returned to the original question:

What would it take for a 2018 Honda Accord Sport 1.5T 6MT to outperform a Ford Mustang EcoBoost?

Current Comparison

Vehicle: Horsepower Per Ton

Accord Sport 1.5T: 121.7

Mustang EcoBoost: 172.0

The Mustang possesses approximately 41% more horsepower per ton than the Accord.

Scenario 1: Match Mustang Power-to-Weight

To match the Mustang's horsepower-per-ton ratio, the Accord would require approximately:

271 Horsepower

This would place both vehicles on equal footing from a power-to-weight perspective.

Scenario 2: Match the Average Five-Second Car

A regression model built from all 176 vehicles predicts that achieving a 5.0-second 0-60 time would require:

222 Horsepower Per Ton

At the Accord's current weight, this translates to approximately:

350 Horsepower

This estimate reflects the broader performance vehicle market and includes factors beyond horsepower such as traction, drivetrain configuration, tire width, transmission tuning, and torque delivery.

Conclusion

The analysis revealed that horsepower alone is not the best predictor of acceleration.

The strongest predictors were:

1. Torque Per Ton

2. Horsepower Per Ton

3. Horsepower

4. Rear Tire Width

The findings suggest that acceleration performance is driven primarily by the relationship between power and vehicle weight rather than raw engine output alone.

While a simple comparison indicates that approximately 271 horsepower would allow a 2018 Honda Accord Sport 1.5T 6MT to match the Mustang EcoBoost's power-to-weight ratio, a broader regression model suggests that roughly 350 horsepower would be required to achieve the acceleration performance commonly associated with five-second 0-60 vehicles.

Ultimately, the answer to "What makes a vehicle fast?" is not horsepower alone—it is the combination of power, torque, weight, and traction working together as a system.