Business Analytics Blog

Business Analytics Blog

Opinions expressed on this blog reflect the writer’s views and not the position of the Capgemini Group

Does home advantage exist in international rugby? by Angus Taylor and Guy Baxter

Categories : Figure It OutSport

New Zealand’s win brought to a close an amazing World Cup. From Japan’s unexpected win over South Africa to Bryan Habana equalling Jonah Lomu’s Rugby World Cup try scoring record, it’s no surprise RWC 2015 has been hailed as the biggest and best tournament to date! It’s a shame that for some fans...this is definitely a tournament best forgotten.

Home advantage

One question must be on the minds of many English rugby fans: ‘what happened to home advantage?’ Home advantage in sport is the supposed benefit to a team of playing on their home ground. Possible drivers of home advantage include the distance opposing teams have to travel, familiarity with the home grounds, and crowd effects. It is also possible that referees are not always as impartial as they should be when handing down decisions.

Evidence in historical rugby match data

Some may point to England’s knockout in the pool stage as evidence that home advantage does not exist in international rugby. The team that stormed to victory this month – New Zealand’s All Blacks – also had the furthest to travel. However, evidence for the existence of home advantage can be found in historical international rugby data. We analysed results for 1,890 rugby matches between 27 international teams held as part of Rugby World Cups, Six Nations Championships and Rugby Championships from 1970 to 2015. Teams in each game were labelled as either home or away depending on where the game took place. Over this time, the average score for home teams was 25 points, whilst the average score for away teams was 18.5, a lead of 35%. This suggests that home advantage has a real and strong impact on a team’s chances of winning a game. The chart below shows estimates of home advantage over this period. According to this analysis, home advantage in international rugby peaked in 1987, in which home teams scored on average 16.44 points more than away teams. By contrast, in 1974, home teams scored on average 2.5 points less than away teams. It is clear that the extent of home advantage varies from year to year but the overall consistency in which home teams outperform away teams suggests that the phenomenon exists and is significant.

Home advantage for winning teams

An alternative method of tracking home advantage over time is to look at the average points scored by home teams when they win. By comparing this to the average points scored by away teams when they win, we can more closely measure the influence home advantage has for winning teams. Analysing the data in this way shows that, on average, winning home teams score 32 points whereas winning away teams score 28 points, a lead of 14%. By this measure, home advantage is a smaller effect than if all games are included but it appears to be a strong positive influence for winning teams. The chart below shows this measure of home advantage over time. The effect is generally positive and significant but this is not always the case, particularly in recent years where the effect appears to be non-existent.

Measuring the strength of home advantage

However, the above analysis may be too simplistic because in a given year, the number of home games a team plays will not necessarily equal the number of away games due to where the international competitions are hosted. The estimate of home advantage will be over-inflated if high ability teams get to play a disproportionate number of games on their home ground against low ability teams in a given year. To correct for this problem, we need to analyse the affect of home advantage on a per game basis while controlling for the differing abilities of the opposing teams. We performed a regression analysis to predict the outcome of individual games in terms of point differential, the points scored by the home team minus the points scored by the away team. For each match, the following four metrics were calculated to account for the opposing teams’ offensive and defensive strengths: expected home points for (xHPF), expected home points against (xHPA), expected away points for (xAPF) and expected away points against (xAPA). These were computed as the average number of points scored (or conceded) by the home (or away) team per match during the year. The model predicts the outcome of a match as follows

The term of interest is home advantage as this constant value holds the systematic advantage for home teams over away teams, after the teams’ abilities have been accounted for. The final term is the error of the model and can partly be interpreted as a ‘luck’ factor.

The above model fits the historical match data reasonably well: it accounts for roughly 75% of the variability in point differential, with the remainder accounted for by other factors including luck. As expected, the terms accounting for team ability are most important for determining point differential. However, analysis of the home advantage term suggests that a systematic home advantage does exist but the effect is quite small. Home teams can on average expect to achieve only 0.33 points more than an opposing away team.

Conclusion

Our analysis has shown that the phenomenon of home advantage does exist in international rugby and that it positively contributes to a team’s chances of winning. However, regression analysis shows that the effect is quite small when the relative ability of opposing teams is taken into account. The strength of the effect also varies significantly over time, and appears to be non-existent in recent years for winning teams. Japan will need to rely on more than just home advantage when it hosts the next Rugby World Cup in 2019.

About the author

Guy Baxter
Guy Baxter
Guy is a cross channel analytics specialist with 8 years experience having worked in both consulting and corporate environments. He is a managing consultant at Capgemini leading the businesses customer insight team within the digital transformation and business analytics department. His last role was managing the digital analytics team for a leading organisation in the media and telecomms industry. Prior to this, he was a senior consultant at Europe's largest customer experience consultancy where he managed an analytics practice for the business. His areas of expertise include analytics and measurement strategy, web analytics, social analytics, customer analytics, media & mobile analytics, digital analytics, survey analytics, multi-variant testing (MVT), behavioural targeting, multi channel analytics, market Intelligence.

Leave a comment

Your email address will not be published. Required fields are marked *.