Rating place-kicking success needs context
Andy McGeady: The two factors with a significant impact on the probability of kick success are distance and angle
Ireland outhalf Johnny Sexton: through the first three rounds of this season’s he had had a 91 per cent kicking success rate. Photograph: Inpho
Risk versus reward. That’s at the heart of the silent process a rugby union captain goes through in those moments after a penalty has been awarded before looking towards the referee and deciding whether to go for the posts.
The kicking tee will arrive and the kicker will start their routine. Clear the mental mechanism. Scrum cap might come off; gumshield perhaps tucked into a sock. Slow down the breathing. The kick could be a gimme in front of the posts or 50 metres back, hugging the touchline, but the kicker’s routine should be the same. Bob is an 80 per cent kicker. That means he’s probably pretty good. But the kicking stat usually used to quantify kicking success does not discern whether a kick is from 15 metres or 50.
The unfortunate truth is that kicking percentage lies.
Jurie Nel, 37, is an actuary in Pretoria. He’s a rugby fan. In 2011 he wondered why place-kicking success was measured via the blunt instrument of success rate. A penalty attempt from 55 metres viewed as the equal of a conversion from under the posts?
Through the first three rounds of the 2015 Six Nations Johnny Sexton had a 91 per cent kicking success rate, the result of nailing 10 goals from 11 attempts. Scotland’s Greig Laidlaw had the exact same statistics. But those 11 attempts were not the same.
While the points scoring system of rugby union is what it is, in rating the work of a kicker a long kick should be rated differently than a short kick. Similarly a tight kick from the touchline versus one in front of the sticks.
In golf, Columbia Business School professor Mark Broadie was the brains behind the Stokes Gained statistics now in use on the PGA Tour. The idea was simple: by making or missing a particular shot, how many strokes did that golfer gain (or lose) compared to how the average tour golfer would have fared on a similar shot.
Points gainedNel’s website – www.goalkickers.co.za – uses the same basic premise. It’s called points gained, and he hopes it will become a widely-used stat for rating rugby goal kickers.
Back to Sexton and Laidlaw. From those 11 attempts Sexton led the competition with five points gained compared to an average professional kicker, as calculated based on over 5,000 kick attempts Nel has entered into the software designed by his colleague Francois Theron.
Leigh Halfpenny’s 14 from 17 attempts was second, good for four points gained. Laidlaw, despite having the exact same 10 from 11 as Sexton, was awarded just a single point gained.
Why? Based on distance and angle the Scot had enjoyed the easiest attempts of the main goalkickers in the tournament. Sexton’s and Halfpenny’s had been significantly more difficult. Points gained takes that into account. It provides context. When Sexton missed a relatively straightforward penalty attempt in Cardiff, points gained was unimpressed.
In trying to add context to rugby place-kicking others have gone down a similar path to that of Nel. Mark Taylor used Opta data in 2013 to perform a similar exercise looking at a selection of matches across multiple competitions and age grades. Ken Quarrie and Wil Hopkins performed a study using 10 years of Verusco data (now Prozone) encompassing almost every Test match featuring a Tier One international side from 2002 until the end of the Rugby World Cup in 2011. That international study found that the expected success rate for a place-kick in Test rugby decreased very, very quickly when the distance went beyond 35 to 40 metres.
Bottom lineThe bottom line? Distance is vitally important. So is the angle of the kick. So are other things – the gap between the two teams’ scores; the time remaining in the game, altitude, weather. But the two factors with a consistent, significant impact on the probability of kick success are distance and angle. So that’s what Nel uses.
Unfortunately Nel is doing this without a data provider. It takes him about 20 minutes to code a match containing 10 to 12 kick attempts. That’s entirely dependent both on local television coverage being available for every game and Nel having the time available to dedicate to it. He missed out on the 2014 autumn Tests, for example, but has committed to coding every major Test in 2015 including the Rugby World Cup.
The captain deciding whether to take that kick attempt won’t be thinking about kicking stats alone. Is it in their kicker’s effective range? If so, the captain’s then moving on to other considerations. Does the opposition have a good maul defence? If it’s a weak spot, that kick to the corner looks a little more appetising. While points gained might help those on the outside; the tough decisions still must be taken by those on the field.
It’s not the perfect kicking stat, but it’s a big improvement. Nel says he needs a data provider and a sponsorship partner to take care of the overheads. With kicking being such an important component of the elite level it’s remarkable an amateur enthusiast is setting the public pace with regard to this aspect of the game. Statistics are an ever-increasing part of sports coverage in 2015. Some like them, some couldn’t care less. But giving context to an important rugby stat is something this parish can get behind.