Oh yeah, don't get me wrong, I have an appreciation for statistics. After all, I am currently working as a researcher on national statistics.
But making informed inferences from these statistics is incredibly difficult. My two main concerns are in arbitrariness (and limitations) and assumptions made:
1. Arbitrariness. Using these statistics to judge how "good" a player is in order for national selection necessitates a somewhat arbitrary decision as to what "good" would look like in the statistics. (Usually, statistics are better at quantifiable things like income, survival rate, obesity rate etc., not quality of a player.) One person might say that a high meters made figure for a front rower is impressive. Another might say that it's indicative that that particular front rower is avoiding tight carries (usually made for minimal ground, but important in building momentum between phases). Take Dave Wilson - we've seen him make great carries for England and Bath, yet his meters made isn't anywhere near as high as Sinckler's. Perhaps Wilson is more inclined to be taking on the important carries in the tight? Which is why these statistics are also limited: to make a full assessment of the impact of the carries made by players, we would want the different types of carries quantified. Number of meters made from tight carries, from carries in the backline, for carries on the edge of the contact zone, for carries taken in the back three etc. We are also missing important statistics such as number of rucks hit. What if, for example, Wood had an absurdly high number of rucks hit comparative to the rest? He has what appears to be weaker statistics than the other backrowers shown, but perhaps his value to the Saints comes from a statistic not actually given? How do you even quantify "slowing the ball down at the contact zone", an important aspect to Kvesic's game?
2. Assumptions made. The biggest assumption being that we can compare players across different teams, without factoring in that they play for different teams. Kvesic is better at the contact zone than his turnovers won column suggests - it's that Gloucester being constantly on the back foot with a miserable defensive record that doesn't allow him to provide the statistics of others. He's (and May from the backs) also given away the most turnovers of the backrowers - probably more to do with his team not giving him the support, than an intrinsic trait of him to cough up possession. The next biggest assumption is that players play in their given position all season. A winger-fullback should be expected to post different statistics to both wingers and fullbacks. To compare a winger-fullback with a winger or fullback may lead to fallacious inferences.
I would suggest that these statistics would be useful for looking at individual trends, rather than comparing players head-to-head. For example, as a Bath coach, I'd see that Ford is making a ton of meters, but also that he's been turned over a lot. This might be used as evidence to inform that he's running away from support, getting isolated, or his team aren't doing enough to support him, and take action. But I would say that the statistics are too limited to make any serious head-to-head comparisons. Although some can be made. As I've said, Yarde's stats are impressive enough to stand out for a player so far down the division. Enough for me to say he's obviously better than the other candidates? I don't think there's enough evidence. But it's certainly eye-catching.
