Archive : Dublin Kerry Semi Final 2016
The following article was published in the Kerryman in August 2016
The biggest irony of Sunday’s All-Ireland semi-final between Dublin and Kerry is the fact that most people will likely consider it a day that Stephen Cluxton will rather forget in terms of his own personal performance. Most memorably, it was his intercepted kick-out which led to Kerry’s equalising goal. Though he could hardly be blamed as such, his failure to beat Paul Geaney in the air under a 50/50 ball resulted in Kerry’s second.
However, despite a day which was undoubtedly below par for his standards, when looked at statistically from the point of view of kick-outs we’ll see that even on such a day Cuxton was still king. Or more significantly in terms of kick-out decision-making and delivery, Kerry’s keeper, Brian Kelly, was pauper. There’s no soft way to put it. Not so much that Dublin won based on the decisions and delivery of their kick-outs, but Kerry undoubtedly lost it on theirs.
The first point to note in all of this is that one key element went completely against what I, and I can only assume, Maurice Fitzgerald would have expected. Down in the Delahunty’s pub on Dorset Street before the game, amongst a cohort of Kerry acquaintances, I was bemused that nobody considered it particularly strange that Anthony Maher wasn’t scheduled to start.
If one thing had seemed obvious to me and was borne out by statistics taken by me at the time, it was that key to Donegal putting Dublin to the sword in 2014 was their complete and utter aerial dominance at midfield. Equally significantly in Kerry’s victory over Donegal in the final that year was the fact that they won more than their fair share at midfield under the long kick-out. Key to this was the massive aerial presence of the twin towers of David Moran and Anthony Maher.
It seemed obvious to me that any side who would expect to beat Dublin would need to force them to kick their own kick-outs long and dominate them there. Alas, as the ball was thrown in, it all made sense. There, lined up for the throw in were the towering figures of the “6”3″ David Moran and the “6” 5″ Anthony Maher. This would give them an inch and a half on the Dublin pairing of Mick Macauley and Brian Fenton. Dominate your own long kick-outs and force Dublin to kick it down these boy’s throats on some of theirs at least and you’ll be in with a chance. The relative beating that the height advantaged Kerry would receive at midfield under long kick-outs could hardly have been expected by Éamonn Fitzmaurice and his charges, though their efforts/lack of system under the break left a lot to be desired. In total, under 24 long kick-outs tallied from both ends Kerry would win ten and lose fourteen. The massively significant factor, however, is what would happen with these kick.
From start to finish Kerry kicked eighteen kick-outs long. In total they won eight of these and lost ten (this accounts for the first team with clean, uncontested possession). This figure is not so significant in itself. What is startling is that from the eight which they won they didn’t garner a single score. From the ten which they lost, they conceded a startling six points! That means that from eighteen long Kerry kick-outs their net score was a loss of six points. In statistical terms that’s a value of 0.33 points conceded every time they kicked a long kick-out.
What will, or at least should, bring Kerry to tears is their statistics on short kick-outs. As already mentioned, typically, short kick-outs played more than 9.5 seconds after the ball goes dead tend to result in more scores conceded than scored. Kerry, however, are a different beast. Almost certainly owing to their immense capacity to carry the ball intelligently and “pick and poke” their way through opposition blanket defences, they actually score significantly more than they concede upon the first turnover when playing short kick-outs, even after the magic 9.5 seconds. Their figures from Sunday’s game are startling. They played ten of these. From these ten they scored five points and didn’t concede any upon the first turnover. That is to say, from a statistical point of view every time Kerry played a short kick-out more than 9.5 seconds after the ball went dead, they produced a net figure of 0.5 points per kick-out!
In cold statistical terms this all leads to one very simple conclusion. Based on the average expectancy per kick-out, had Kerry chosen/been able to play all of their kick-outs short to the full back line instead of long, even after the magic 9.5 seconds, they’d have saved themselves six conceded points and would have scored a staggering nine points! That is to say that instead of the final score reading 0-22 to 2-14, you could have expected it to have ended 0-16 to 2-23 to Kerry!
There are two key points to note here. In the absence of serious competitive matches for either of these sides already this year, there wouldn’t have been the evidence of previous patterns to suggest that we’d see such a pattern from either side. To that end you could hardly blame Fitzmaurice for attempting to pummel Dublin with long kick-outs.
It’s also worth pointing out that the 0-16 to 2-23 figure is somewhat one dimensional. Apart from the fact that Kerry wouldn’t have had the amount of kick-outs they did had they scored nine points more and conceded six less, this figure potentially falls foul of falling into the category “A” statistical pitfall. It doesn’t give weight to the fact that maybe on some of these long kick-outs that the short kick-out wasn’t available/a safe option. Perhaps Kerry only kicked long when they didn’t have any other choice.?
The issue, however, is the fact that this pattern was already established by half time. Kerry had already conceded four points and scored none from their own eleven long kick-outs. They had scored one from two on their own delayed short kick-outs and one from one with quick, short kick-outs without conceding any. One can only assume that the correct half time analysis would have been central to Kerry’s second half kick-out policy. Evidence points towards the fact that it was, though maybe not as strongly as it should have been. Kerry certainly tried to play more short kick-outs early on in the second half until Dublin upped the ante and started splitting the extra man in Kerry’s defence with more vigour. All in all Kerry played three short and eleven long kick-outs in the first half. They played seven long and ten short in the second half!
The key element in all of this is Brian Kelly’s ability, or lack thereof, to pick out those spare players around Kerry’s full back line and half back line. For example in the 65th minute, with Kerry still a point ahead, the RTE cameras showed Kerry’s kick-out from behind the goal with a full view of the field. It wasn’t an outrageously easy option but it wasn’t terribly difficult either, between the full back line and the half back line was an unmarked Kerry player. Quite simply, armed with the statistical patterns which were evident both at half time and at that point, the percentage option would undoubtedly have been to dink the kick-out into the clear path in front of that free man. And key to all of this is the fact that nineteen times out of twenty, you’d expect that Cluxton would have! That one time, Dublin might have conceded a score, maybe even a goal as they did on Sunday. But over the long term, based on percentages, he’d have tried it, would have been correct to have and probably 95 percent of the time would have nailed it. Kelly didn’t try it, he kicked it long, and as patterns had suggested would be the case,Dublin won it. Diarmuid Connolly broke it down for Paul Mannion, Dublin earned a free and scored the equalising point.
Conclusion? Either Kerry’s management team weren’t suitably aware of just how significant the statistical pattern at hand was, they didn’t have a defence intelligent enough to consistently create the space to make their spare man in defence (which they had all day) an easy target or they didn’t have a keeper brave enough or suitably skilled enough to nail it and hit that man when he was available. This 65th minute incident suggests the first and third of these factors.
Though less spectacular, Dublin had on the day, and generally have equally significant stats related to long and short kick-outs/kicks to the half back line. In the 56th minute, a point down, Cluxton had a kick-out with whereupon his free man, 40 odd yards away, was significantly more difficult to hit than the aforementioned Kerry kick. He nailed it down the throat of the free man and Dublin went straight down and earned a free just outside the “45”.
Even on a bad day, broken down statistically, Cluxton was still king.
The stats in this article were compiled using the newly developed GaaProstats statistical and video analysis program. This program has been specifically designed for Gaelic football and hurling. It is now available for all clubs to download free for a month’s trial.
By Stephen O’Meara