More Competitive Balance in Gaelic Football

By John Eakins

An earlier post by John Considine looked at the topic of competitive balance in the All-Ireland Senior Football championship from the perspective of the distribution of All Ireland titles. John correctly suggested that a title for Mayo would improve the overall level of competitive balance in the competition but one could look at more than just the winners to get a sense of the degree of competitive balance. In particular, it would appear that the same teams are competing at the quarter final and semi-final stages of the championship. For example, Kerry has appeared in every quarter-final since 2001(13 in total). Dublin has appeared in 12, Tyrone in 10 and Mayo in 8. In terms of semi-finals, since 2001, Kerry has appeared in 11, Dublin in 7, Tyrone in 5 and Mayo in 5.
 
Many have argued that the introduction of the new qualifier format in 2001 have benefitted the so-called stronger countries more so than the weaker counties. In short the qualifier format allowed losing teams in the provincial championships to enter the ‘qualifiers’ for a place in the All-Ireland quarter finals. To look at this is more detail, an index can be calculated based on how similar (or dissimilar) the counties that reach a particular stage are compared to the same stage the previous year. For example, if 5 teams that reach the quarter final stage in 2012, also reach the quarter final stage in 2013, the value of the index is 5/8 = 0.625. So if the value equals 1, the same teams are in the quarter final (or semi-final stage) as compared to the previous year. If the value equals 0, there are entirely new teams.

Comparing one year to its previous may not properly capture the ‘persistence’ of a counties appearance at the latter stages of the football championship so an additional index can be calculated based on an average of the similarity between the current year and previous year, the current year and two years back and the current year and three years back. Essentially this index represents how similar are the teams that compete at a particular stage of the championship over a four year period.
 
The index values are calculated from the 1970’s onwards to observe any differences in the long term trend. Three graphs based on the results of the analysis are presented below. The first is based on data of the teams that reached the provincial final (effectively the quarter final stage before the introduction of the qualifier format in 2001) or the quarter final stage post 2001. The second graph adds in the values of the indices using the teams that reached the provincial final from 2001 onwards. Essentially this presents a comparison between what happened and what would have happened without the change in format. The final graph is based on the teams that reached the semi-final stage of the championship.

Looking first to Figure 1, in the 1970’s, 1980’s and early part of the 1990’s, on average, about 5 of the teams at the quarter final stage were the same in comparison to the previous year or over a four-year period. In the second half of the 1990’s the index fell to between 3 and 4 teams. This is probably due to the introduction of the open draw in many provincial championships which saw some of the weaker counties reaching provincial deciders. The introduction of the qualifiers however does appear to have reversed this trend with the ‘similarity’ of teams reaching the quarter final  stage increasing since the early 2000’s. 
 
Figure 2 is interesting as for most years post 2001, the ‘similarity’ of teams reaching the quarter final stage is higher under the new qualifier format than compared to the old provincial format (if that was continued). In the years 2009-2012, the gap is especially wide although it has closed again in 2013. There are only two years, 2007 and 2008, where the old provincial format generates a higher similarity index compared to the new qualifier format.

Figure 3 also appears to indicate a reverse in the trend against a more open football championship. In the 1980’s and 1990’s, on average, approximately 1 to 2 of the same teams reached the semi-finals year in year out, whereas in the second half of the 2000’s and beyond this has increased to between 2 and 3 teams. 
 
As Humphreys (2002) states “Competitive balance reflects uncertainty about the outcomes of professional sporting events” pg 133. So a Mayo win in the All Ireland Football final in four weeks’ time will be welcome from the neutrals perspective but it still doesn’t get around the potential problem that the GAA have where the top 8 and top 4 in the competition are getting easier and easier to predict.