Have F1 cars become faster during the past 30 years?

Advances in technology have impacted performance sports, including F1 over the past 30 years. As we look to explore how these changes have impacted speeds achieved by F1 drivers, we must consider a number of critical variables, including a definition of “Faster”, differences in performances achieved during stages of the race (first lap, last lap, fastest lap, mean speeds), account for variances caused by undesired or necessary events (accidents, pit stops, technical delays) and isolate for other material variances which might cause noise in our calculations (isolating performances by Individual drivers and achieved on specific tracks)

Key Variables and Considerations

Definition of Fast

For our analysis, we will utilize lap times as our measure of "Fast".

Year over Year

Year over Year aggregate comparisons reduce noise and facilitate a macro level comparison.

Track Name

Unique site of race, We must consider impact of nuanced design and layout of each track.

Racing Team

Budgets, resources, technologies vary across teams, how does this impact our definition?

Driver

Enabling isolation to single individual, perhaps the purest form of overtime comparison.

Driver Age

Does the age of the Driver impact performance? Can increases be attributed to non car related factors?

First Lap

Initial lap after formation lap, notably slower as drivers develop a comfort for racing conditions.

Fastest Lap

A representative metric for the best achievable performance given a set of conditions.

Mean Lap

A representative metric of what is most likely to happen and be achieved given a set of conditions.

The Analysis...

Please explore the below visualization, which enables filtering of Lap Times against key variables, as you search to understand and we answer the question, have F1 cars become faster during the past 30 years.

Observations

As above, we can see that lap times have not decreased over time. While there have been individual year over year gains evidenced, they have not been absolute over our study, neither have they been consistent or persistent . Our analysis enables us to explore and evidence interesting and critical observations related to individual driver, track, team and driver age giving us the confidence to validate our hypothesis, there remains countless other interactions and nuances which further support. As below, are a summary of interesting observations, approaches to review the data, and perhaps some ideas to spark your curiosity to further explore factors underlying causation.

Location, Location, Location.

While average lap times remain the best approximation we have of our definition of fast, it is critically important to remember that every track is unique, and custom built for unique interactions. Courses can be built for speed, challenging navigations, while others (Like Las Vegas GP which held its inaugural race in November 2023) are being explored, and developed to pose new and unique driver and fan experiences.

As evidenced by the heat map, there are no races which have evidenced mean lap time reductions which is consistent with our hypothesis.

Further, it is interesting to note a number of tracks which have been introduced since 2020, which will continue to provide data points for our continued analysis and might pose some additional questions, such as whether familiarity with a course impacts lap time.

Team Performance

Over the past 30 years only 12 teams have participated for 10 or more seasons, with only 4 participating for 20 or more seasons. This fact proves interesting for several reasons, firstly, when trying to compare performance across teams it is difficult to find appropriate relative comparisons, as each team is on their own unique journey, and comparing Ferrari ( a company with deep traditions in racing, and having competed annually for 30+ years) with a team entering today (which is learning all aspects of the sport). Secondly, given the prevalence of the sport, it might be surprising that so few teams sustain over the long-term, however given the cost to develop, operate and continued need to invest in research and develop, perhaps it is not surprising that only a few highly motivated entities pursue F1. That being said, the number of teams competing today is higher than 30 years ago, which bodes well for the long term prospects.

Experience Does Matter.

While we do not see prolonged increases in performance in our data set, if we isolate the initial 10 years of team participation, we can evidence a very real and meaningful reduction in Lap Time, evidencing that there unquestionably is a steep learning curve, which teams initially overcome, but there is a very real revision to the mean time, which as discussed has not decreased.