HoneyBadger35
Small, bitter, fascinatingly unimpressive.
- Aug 11, 2011
- 31,902
- 93,210
- AFL Club
- West Coast
- Moderator
- #1
We've all heard for three years now that our beloved back-to-back Coleman medallist/steak-knives in the Judd Kennedy trade is the quintessential Simpson Eagle; that is to say, a soft, down-hill skiing flat tracker.
Not content with JK leading the comp in goal kicking (twice), or the fact that a staggering majority of clubs would trade the farm for prime Josh Kennedy, an increasing number of posters have been hitting the all-too familiar "doesn't deliver when it counts" line. I thought I could dive into Kennedy's numbers to assess the validity of this train of thought, at least from a statistical perspective. Does Kennedy's output drop in losses, is it all anecdotal, and do I have a third point to finish this sentence? For the answer to the first two questions, read on (the answer to the third is no).
I started by charting the Eagles' game-by-game goals since 2015, along with Kennedy's goals, marks, tackles and disposals in those games. What follows is stat heavy AF, so I've used spoiler tags to hide a bunch of tables. If you want the quick results, lower your eyes slightly. If you can still read this, you'll need to lower them slightly more.
In 2015, Kennedy tallied 2.3 more marks, 0.9 more tackles and 3.8 more disposals in wins compared to losses. However, he was responsible for 22.6% of our scores when we won, and 22.2% when we lost - a difference of just 0.4%. Essentially, JK's scoreboard impact was the same regardless of the team's result.
In 2016, Kennedy actually tackled more frequently and gathered more possessions in losses when compared with wins. He did, however, take more marks in victories, and contributed 4.5% more to our winning scores than he did in losses.
2017 stats are perhaps the most in line with the flat track theory, but the majority of this I'm putting down to sample size. This will likely fall back in line with previous season's figures by year's end, although as we're going to finish the year undefeated from this point in, perhaps not.
JK is taking nearly three more marks in wins, finds the footy an additional 3.5 times per game and contributes 10.3% more to our score in wins, certainly not an insignificant amount.
The first thing that jumps out from this is the fact that, since 2015, we've averaged seven fewer goals in losses than wins, which is pretty damning. Back to Kennedy, and the bottom line is the focus here. JK's output increases to the tune of 2.0 goals, 1.7 marks and 2.0 disposals in victories, while tackles held steady. The 3.5% difference in goals responsible for is a fairly insignificant margin in my mind; more on that below.
In short: Since the beginning of the 2015 season, Josh Kennedy averages 2.0 more goals, 1.7 more marks and 2.0 more disposals in wins. He averages the same number of tackles in both wins and losses.
However, as you intuitively kick more goals in wins than losses, the raw number can be misleading. As such, the numbers were then broken down into % of goals responsible for e.g. if Josh Kennedy kicks 10 goals and we kick 20, he was responsible for 50% of our goals. If he kicks three but we're getting thumped and only finish with six goals, the figure remains at 50%
On average, Kennedy is responsible for 23.7% of our goals since 2015. In wins, he contributes 24.4%, while that number dips to 20.9% in losses. That is to say, Kennedy has, on average, just 3.5% less scoreboard impact in losses than wins.
The Eagles have averaged 14.5 goals per game since 2015, with 3.5% of 14.5 being 0.5. This entire narrative, at least statistically, is born out of Josh Kennedy playing half a goal worse by his standards in losses, while also tallying 1.7 fewer marks and 2.0 fewer disposals. I'm not suggesting this was a quick and lazy attempt at analysis in 2015 that then stuck and became the Fox Footy party line, but that is actually exactly what I'm suggesting.
Stats are not the be all and end all of this debate. A pack mark on a crucial kick-out is clearly worth more than an uncontested mark out the back when you're switching play. Not all goals are created equal. A fantastic lead or drawing two men to the contest doesn't show up in the stats, just as a costly fumble or critical dropped mark would not. If you think Kennedy is a front runner, then you'll likely argue that stats aren't the way to prove that, and vice versa. However, hopefully this is food for thought for the next time you're entering into the debate, regardless of what side you're on.
Not content with JK leading the comp in goal kicking (twice), or the fact that a staggering majority of clubs would trade the farm for prime Josh Kennedy, an increasing number of posters have been hitting the all-too familiar "doesn't deliver when it counts" line. I thought I could dive into Kennedy's numbers to assess the validity of this train of thought, at least from a statistical perspective. Does Kennedy's output drop in losses, is it all anecdotal, and do I have a third point to finish this sentence? For the answer to the first two questions, read on (the answer to the third is no).
I started by charting the Eagles' game-by-game goals since 2015, along with Kennedy's goals, marks, tackles and disposals in those games. What follows is stat heavy AF, so I've used spoiler tags to hide a bunch of tables. If you want the quick results, lower your eyes slightly. If you can still read this, you'll need to lower them slightly more.
In 2015, Kennedy tallied 2.3 more marks, 0.9 more tackles and 3.8 more disposals in wins compared to losses. However, he was responsible for 22.6% of our scores when we won, and 22.2% when we lost - a difference of just 0.4%. Essentially, JK's scoreboard impact was the same regardless of the team's result.
In 2016, Kennedy actually tackled more frequently and gathered more possessions in losses when compared with wins. He did, however, take more marks in victories, and contributed 4.5% more to our winning scores than he did in losses.
2017 stats are perhaps the most in line with the flat track theory, but the majority of this I'm putting down to sample size. This will likely fall back in line with previous season's figures by year's end, although as we're going to finish the year undefeated from this point in, perhaps not.
JK is taking nearly three more marks in wins, finds the footy an additional 3.5 times per game and contributes 10.3% more to our score in wins, certainly not an insignificant amount.
The first thing that jumps out from this is the fact that, since 2015, we've averaged seven fewer goals in losses than wins, which is pretty damning. Back to Kennedy, and the bottom line is the focus here. JK's output increases to the tune of 2.0 goals, 1.7 marks and 2.0 disposals in victories, while tackles held steady. The 3.5% difference in goals responsible for is a fairly insignificant margin in my mind; more on that below.
In short: Since the beginning of the 2015 season, Josh Kennedy averages 2.0 more goals, 1.7 more marks and 2.0 more disposals in wins. He averages the same number of tackles in both wins and losses.
However, as you intuitively kick more goals in wins than losses, the raw number can be misleading. As such, the numbers were then broken down into % of goals responsible for e.g. if Josh Kennedy kicks 10 goals and we kick 20, he was responsible for 50% of our goals. If he kicks three but we're getting thumped and only finish with six goals, the figure remains at 50%
On average, Kennedy is responsible for 23.7% of our goals since 2015. In wins, he contributes 24.4%, while that number dips to 20.9% in losses. That is to say, Kennedy has, on average, just 3.5% less scoreboard impact in losses than wins.
The Eagles have averaged 14.5 goals per game since 2015, with 3.5% of 14.5 being 0.5. This entire narrative, at least statistically, is born out of Josh Kennedy playing half a goal worse by his standards in losses, while also tallying 1.7 fewer marks and 2.0 fewer disposals. I'm not suggesting this was a quick and lazy attempt at analysis in 2015 that then stuck and became the Fox Footy party line, but that is actually exactly what I'm suggesting.
Stats are not the be all and end all of this debate. A pack mark on a crucial kick-out is clearly worth more than an uncontested mark out the back when you're switching play. Not all goals are created equal. A fantastic lead or drawing two men to the contest doesn't show up in the stats, just as a costly fumble or critical dropped mark would not. If you think Kennedy is a front runner, then you'll likely argue that stats aren't the way to prove that, and vice versa. However, hopefully this is food for thought for the next time you're entering into the debate, regardless of what side you're on.
Last edited:
