@Batty an interesting article ...
There?s a common saying that goes ?statistics are like a bikini ? they show everything but the important bits?. Other people say that ?statistics mean nothing?, and further still, there are some who believe that you can analyse a game of rugby league by looking at a sheet of paper. This essay will investigate the use of player statistics in rugby league, how much can be inferred about a player from his statistics.
Statistics are raw information, so one benefit they have in terms of analysis is that they are unbiased, unlike casual observation. There are countless ways of interpreting statistics, which means it is impossible to know everything about a player through statistics. It is reasonable to imagine a brilliant halfback failing to get any try assists because of a weak forward pack, and outside backs who are incapable of holding onto a ball. It is in this process of interpretation of statistics that the art lies, yet an ordinary halfback may get better statistics simply due to better players surrounding him. For example, debate is heated regarding
Brett Finch, who led the NRL in 2004 in try assists and line break assists. Do these statistics mean that he was the better halfback than more glorified halves such as Matt Orford and
Craig Gower?
Imagine two lots of statistics, taken from one game. One lot belongs to player C, and consists of plenty of metres gained, offloads and try assists, and the other, belonging to player D, really poor, with very low quantities in these areas. It still remains possible that player C is a mediocre player who was part of a top quality side, whereas player D was in a poor side, and was targeted by the opposition defence. It is very plausible that if player D was given the same opportunities as the player C, would have equalled or even bettered his performance. We see this often in the NRL, when players from good teams dominate selection for representative teams. Given the same opportunities, it is reasonable to think that players of lower teams would be able to equal or better the performance of some of these representative players.
Statistics are made in such a way that they can hold very little ambiguity, for example, one might argue that a try was scored due to poor defence, and another might argue that the try was scored due to intelligent attack, however neither could argue the fact that a try was scored. There may be disagreement as to whether a try was created by the fullback who charged onto a short ball, or the halfback who gave the short ball and drew two defenders, but it can not be disagreed that the halfback receives a try assist, as this is the definition of a try assist: to give the final pass before a try is scored, whether or not is was a bad pass, a fluke pass or a brilliant pass. The try assist statistic simply does not give any information as to which of the two, the try assister or the try scorer, was more responsible for the try being scored. It does give out a clue, though, and this is the case with many statistics. For example, passes or kicks which are put through to lead to a try have to be of a minimum standard in order for a try to be scored, for example, a kick which goes out on the full is no use to any winger, no matter how good at catching they are. This tells us that the try assister possesses that minimum standard of ability ? but is this all it tells us?
It is still unknown exactly how much skill the halfback possesses, as the try assist statistic represents quantity only, not quality. The next step in this situation is not to disregard statistics altogether as meaningless, but rather to look at other related statistics, in order to get more clues. The process of looking at the whole picture is very important in interpretation of statistics, and it is when statistics are studied in isolation that they become misleading.
Given all this room for interpretation, it is not surprising that opinions regarding individual players vary so highly. Statistics don?t say nothing, but they don?t say everything, either. More often than not, the art lies in the interpretation of statistics, not the statistics themselves
