Same, same but different: analyzing uncertainty of outcome in Formula One races

ABSTRACT

Purpose:

Formula One (F1) is considered as one of the most popular sports worldwide. However, no previous research inspected its determinants of attendance. This paper aims to cover this gap.

Methodology:

The dataset includes races from 2015 to 2019. The impact of uncertainty of outcome (UO) is firstly addressed by coefficient of variation, Herfindahl–Hirschman and Gini. Later, a specific metric for F1 is presented. The econometric approach consists in Ordinary Least Squares regressions with season and circuit fixed effects.

Findings:

The results evidence that an increase in price slightly reduces the demand as well as a significant inverted U-shaped relationship between attendance and UO. Moreover, the findings suggest that the latest performance of a local driver has a significant positive effect in the attendance level.

Practical implications:

The empirical results provided may aid F1 races’ organizers maximizing live attendances.

Research contribution:

The findings offered here suggest the use of specific UO indexes for different sport disciplines in further research.

KEYWORDS:

• Attendance
• competitive balance
• demand
• Formula One
• uncertainty of outcome

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Notes

1 See it at http://www.watsonpost.com/how-does-formula-one-make-money/.

2 We do not discuss these papers in details because they modelled broadcast demand rather than tickets demand. However, we highly recommend them for those interested in the economics and management aspects of motorsports.

3 We do understand that a larger sample could offer valuable insights, essentially because of the potential higher variation of winners, but, unfortunately, we were not able to collect data. This limitation might be dealt in further research if more data are available. Nonetheless, from an econometric point of view, a panel data with five years’ timeframe would not constitute an issue.

4 All the information provided can be double-checked at F1 Tickets Race Calendar at http://tickets.formula1.com/en.

5 The relative points are calculated by the sum of points of a driver prior to his domestic race divided by the number of previous Grand Prix until that date in a given season. It adjusts the points by the number of races, where 25 is the maximum and 0 minimum.

6 For countries which had two or more local drivers in a domestic Grand Prix, all variables considered the values from the best of them at that moment, assuming that he would be the “local hero”.

7 We carried out alternative estimations using GPD per capita rather than both GDP and population as well as excluding the unemployed rate getting similar results. Only marginal effects slightly changed, but keeping similar significance and directions. We could also use of the Human Development Index as an alternative variable, but we have not included here once it is highly correlated with the GDP.

8 One could suggest multicollinearity among these factors, but we have not detected it by any statistical test. Additionally, we carried out alternative estimations including and excluding each of the variables related to the local driver and none of them – except “previous race performance” – shown statistical impact. Our result is robust regardless the presence (or not) of each of the others factors concerning the local drivers in all models.

9 Unfortunately, we cannot identify when a person purchased a ticket. F1 fans can buy tickets over the whole season, but there is no clear rule about when tickets become available – it is a decision of each race organizer. However, most of papers dealing with tickets demand for sports faces similar issue – season ticket holders purchase tickets before season starts, but single tickets can be purchased in months, weeks or days before a particular match/fight/race. However, researchers using secondary data are generally not able to identify these elements correctly. We assumed here that UO could imply an increase/decrease in the fans interest and tested it: our empirical findings show some preliminary evidence that race uncertainty is not relevant for F1 fans purchasing weekend tickets, although the “circuit uncertainty” exhibits some degree of statistical effect.

Additional information

Funding

This paper is an output of a research project implemented as part of the Basic Research Program at the National Research University Higher School of Economics (HSE).