This is the third in Neil Watson’s series of blogs for OptaPro. In these articles Neil approaches data analysis within rugby union from an academic standpoint while maintaining a focus on real-world applications.
Neil is a Statistical Sciences lecturer at the University of Cape Town.
This is the third post in a series of blogs that will consider various aspects of team performance. In the previous post, I analysed only the top and bottom teams in the European Champions Cup (ECC), Heineken Cup (HC) and Super Rugby (SR).
I chose the two finalists and the two teams with the lowest log points in each tournament, and contrasted them with respect to their rankings on a number of Key Performance Indicators (KPIs) across six areas of play: attack, defence, discipline/errors, kicking, possession/territory and set pieces. Read the article here.
Setting a game plan
A question of great debate in the global rugby community is whether specific game plans exist that contribute towards winning. In this post I will compare the top vs. bottom teams in greater detail, to contrast the different game plans adopted within each competition and across competitions.
Table 1 below shows the rankings of the top and bottom teams on a subset of KPIs within these three competitions. Figure 1 (further below) displays the average rankings in each gameplay area (using all 69 KPIs) for the top and bottom teams.
The average rankings displayed in Figure 1 are calculated as the weighted mean of the KPIs in each category, with the weights being the respective effect sizes of each KPI (i.e. those KPIs that displayed greater differences between winning and losing teams were given more weight in determining the average rankings).
Table 1: Top vs. bottom team rankings on KPIs within each competition. TLN = Toulon, SRC = Saracens, TRV = Benetton Treviso, ZBR = Zebre, CLR = Clermont Auvergne, CST = Castres, SS = Sale Sharks, HGH = Highlanders, HRR = Hurricanes, BLS = Blues, FRC = Force
Figure 1: Average Rankings by area of play All competitions (top left), ECC (top right), HC (bottom left) and SR (bottom right)
While there are many insights that can be drawn from the data, it is important to remember that no set of statistics can ever capture the ‘entire picture’, especially in a game as multi-faceted and complex as rugby. Rather, they enable us to make inferences as to potential factors at play in a team’s success. The statistics here are also aggregated over entire seasons, which may mask the effect of certain factors like opposition strength. With this in mind, here are some insights I gleaned from the data presented in this article:
– A team cannot excel in only one area of play if they want to be successful. Although they may be strongest in certain areas (e.g. TLN – defence in ECC, SRC and HRR – attack in HC and SR), they need to boast a good all-round game.
– The most prominent difference between top and bottom teams across the three tournaments is in the area of attack. In all three competitions, the top teams averaged considerably higher in this area than the bottom teams. The disparity between top and bottom teams across some of the attack KPIs is evident in all three competitions in Table 1. One KPI to note is linebreaks – the top two teams consistently rank higher than the bottom two teams. This points to the importance of being able to identify and exploit space in the opposition’s defensive system.
– While some teams place right near the top on certain KPIs in each area of game-play, no one team dominates across all the KPIs. This is most clearly seen in the top left graph in Figure 1. Of the teams displayed, TLN comes up top in four out of the six areas of play but doesn’t average a higher rank than approximately 13th out of 51 teams in any of the areas.
– Maintaining a high standard of play throughout the round robin/league stage of does not always translate into winning the tournament. In both the HC and SR, the losing finalists (SRC and HRR) outperformed the eventual winners (TLN and HGH) in 5 out of 6 areas of gameplay across all games in the season, yet still lost the final. This lends credence to the theory that ‘peaking’ at the right time plays an important role in a team’s chances of winning in the knockout stages of a tournament.
– When contrasting the three tournaments, the difference in ‘distance’ between the top and bottom teams is striking. It is greatest in the HC, where there is a clear distinction between the top two and bottom two teams. The gap closes somewhat in certain areas in the ECC, although the distinction between top and bottom teams is still clear. However, in SR, there is considerable overlap between top and bottom teams in certain areas (defence, discipline/errors and possession/territory). This is indicative of SR being a more competitive tournament (although the most recent expanded iteration of this competition did display greater disparity between the top and bottom teams).
– In the HC, TLN appears to have employed a more defensive game-plan than SRC. They preferred to play ‘without the ball’ and rely on their defence (ranked 5th overall for the ratio of breakdowns won and lost on defence) to force the opposition into conceding penalties, which they were largely successful in converting (ranked 1st in successful penalty goals). SRC, however, adopted a more ball-in-hand approach where they aimed to score tries through gaining territory in carrying past the gain line/breaking the defensive line. This is reflected in them ranking higher in all the attacks KPIs included in Table 1 than TLN, highest for successful conversions and second for % of total tries. Of interest, SRC ranks higher than TLN overall in defence in the HC. This is an interesting comparison, as it indicates that it is possible to achieve success in the league stages of competitions by employing game plans that focus on different aspects of gameplay.
– In the ECC, TLN embraced a more possession and carrying-focused game plan while also improving their defence. They continued to dominate at the breakdown and force penalties out of the opposition, while markedly improving their attacking game overall. One can argue that TLN were a more balanced side in 2014 (when they won both the ECC and Top14). CLR played a similar style, placing more emphasis on and excelling with their kicking game (ranking 1st for kicks where possession was regained).
– In SR, the HRR had a similar game-plan to SRC in the HC – one with a focus on successfully gaining territory through carrying the ball past the gain line. This is reflected in them ranking highest on all KPIs related to carrying the ball. The HGH adopted a similar strategy, with the key differences being that they kicked more out of hand and enjoyed more possession in general than the HRR. Overall the HRR dominated the HGH in many aspects of play and finished top of the league stage by quite some margin (1st place with 66pts to the 53pts of the HGH). In the final match of the tournament between these two sides, the HRR persisted with their strategy and carried more, beat more defenders, made more clean breaks and more metres then the HGH. The HGH kicked more and made fewer errors and ended up scoring two tries to one and winning the game.
In sum, successful teams boast good all-round games, and there has been a shift towards more attack-minded game plans that focus on maintaining possession while advancing towards the opposition try line. This is achieved through strong ball carrying where space is sought for players to break the line or at least carry over the gain line, clever kicks that present opportunities to win back possession (either via a lineout or aerial challenge), and ultimately scoring the majority of points through converted tries.
