/cdn.vox-cdn.com/uploads/chorus_image/image/44184194/458623152.0.jpg)
With the international break swiftly coming to an end, now seems like a good time to take a deep breath and boil Villa's season down to the numbers and what metrics we should be looking at in the upcoming weeks.
Statistics can't capture football in all its glory and each week I've been trying to mix a measure of the numbers and more descriptive analysis but I've diagnosed the current 4-3-3 system as far as is possible until Paul Lambert shows a willingness to change or some key players come back into the side. And in the world of advanced statistics there are some very useful concepts that can help show whether a side is just getting lucky/unlucky or whether structural issues are affecting their performance and what we might expect in the future.
So let's take one of the simplest and most useful of those metrics - Total Shots Ratio (TSR). This is a measure I am taking from James Grayson, who has done some wonderful work on football statistics.
It is calculated like this - Total Shots For / (Total Shots For + Total Shots Against).
It's a very simple metric and while it looks like it's primarily concerned with attacking play, it's actually much more representative of control of the ball. This is due to the fact that it doesn't measure the quality of shots or their likelihood of going in, so unusually efficient or inefficient attacking sides might not be fully represented.
However as is visually explained here by Grayson, it is essentially the best predictor so far of a side's points and final position in the league - confirming the common sense idea that a side that has control of the ball will be more successful. It is also one shorn of the large amounts of luck that often go into save percentages and shots on target percentages which tend to go through streaks before regressing to the mean.
So let's take Villa's season so far and plug in the numbers (statistics from footstats.co.uk). Villa have attempted a neat 100 shots so far this season. Their opponents have attempted 168 shots. Plug this into the formula - 100/(100+168) - and we get a rounded TSR for Villa of 0.372
Of course statistics are only useful in context. Here's the Premier League ordered by TSR -
Arsenal | 0.692 |
---|---|
Southampton | 0.627 |
Manchester City | 0.615 |
Manchester United | 0.600 |
Liverpool | 0.582 |
Chelsea | 0.570 |
Newcastle United | 0.568 |
Tottenham Hotspur | 0.519 |
Everton | 0.516 |
West Ham United | 0.515 |
Stoke City | 0.514 |
West Bromwich Albion | 0.494 |
Queens Park Rangers | 0.456 |
Sunderland | 0.451 |
Burnley | 0.428 |
Crystal Palace | 0.395 |
Swansea City | 0.382 |
Aston Villa | 0.372 |
Leicester City | 0.362 |
Hull City | 0.356 |
Credit to Michael Caley over at the SBNation Tottenham site Cartilage Free Captain for the TSR statistics which can be found here.
So according to the most accurate single statistical predictor of final league position, Villa are in line for a relegation spot.
Now of course, this statistic is not everything - after all Man United fans would be surprised to hear they're in line to do better than Chelsea. And statistics are liable to change as the sample size increases - hopefully that sample size increase will be a kind one in the next string of fixtures. But what we're seeing on the pitch and what the numbers are saying confirm one another: Villa need to pick up the pace quickly.