How Many Points Might We Need To Go Up?

“It’s a point. It’s fine. It’s closer to the total that we want.” Michael Appleton’s words in the recent Accrington post-match press conference after Dion Charles’ last-minute equaliser writes Adam Foster.

We’ve heard similar remarks for a number of months about this total, but the figure itself has remained a closely guarded secret.

Although most of us would have been more than happy with a top-ten finish at the beginning of the season, it’s likely that back in September the management and board were setting a play-off position as the number of points to aim for. Many organisations use ambitious targets to drive standards and football is no different. After winning the first seven out of ten games by the time November rolled by, it’s safe to say that the points target the club were aiming for rose significantly, with the top two promotion places in sight.

If we’re to hazard a guess at how many points are required for an automatic promotion place, we obviously need to look back at previous seasons in the third tier. Simply adding one point to the third-placed team’s total each season gives us a ballpark figure (ignoring last season’s curtailed finish):

League One Figures

As you can see, it seems like a points haul in the mid to high eighties is about the right amount but there’s considerable variability, especially in those consecutive years of 2011-12 and 2012-13 which had a massive eleven point difference.

Every league table has its own unique characteristics each season. The Premier League, which has the biggest disparity in resources between top and bottom, generally gets quite stretched, with the top one, two or three gaining a huge number of points and the bottom few struggling to get to the so-called magic 40-point mark. One of the hallmarks of the Championship that makes it such an exciting league is that the opposite happens – teams are more bunched together – leading to tumultuous final weeks and days as more than half the division seem to have something to play for.

Teams in a stretched table, where the top teams win more matches, may need more points to do well. In a bunched table a lower total can get you where you want to be.

To get an idea of this season’s points target, what we need is a way to see if our current table is likely to be stretched, bunched or somewhere in between. Luckily there is a way of measuring this, called standard deviation.

Standard deviation is a measure of how spread out a set of numbers are. It works by calculating how far away each data point is from the mean (average) number. In a stretched table, we might expect the points totals of teams near the top and bottom to be further away from the mean number of points and therefore the standard deviation number (the spread) to be greater.

Unfortunately, it’s not as simple as that. Some divisions can have hugely competitive parts of a table.

How many points does this man want? – Courtesy Graham Burrell

We’ve all seen seasons where the bottom teams need a high points total to stay up. This doesn’t always result in a bunched table. We’ve also seen leagues where several top teams earn more points than we would expect. This doesn’t always mean this table will be stretched.

There are other factors, such as the number of draws in a season or if there is a whipping boy or two, which affect things.

However, once the season is already half-completed, as ours now is, these patterns have already begun to show themselves. Luckily, the standard deviation (the spread of points) mid-way through a season does relate to the standard deviation at the end.

There are always exceptions, but teams that draw a lot of games in the first half of a season tend to do the same in the second. Real whipping boys (sides that are a long, long way from safety) most often don’t recover.

To make a fair comparison, we need everyone to have played the same number of matches. At the time of writing, a couple of teams are still on the 25-game mark, so we will use that as our benchmark.

The graph below shows how the standard deviation changes each season from the 25th game to the end of the season. Naturally, the spread is much bigger at the end of a season because the points gap from the top to the bottom is greater, but it is interesting how in many seasons the lines go up at a really similar rate.

This means that by using the standard deviation of our current table, taking everyone’s points totals after 25 matches, we can make a prediction of the points total we’ll need at the end of the season.

Graphic by Adam Foster

Using teams’ points totals after 25 matches, the current League One table had a standard deviation of 9.5. This puts it pretty much bang in the middle of the starting points on the left hand side of the chart.

Looking at how the things changed in previous campaigns (and especially closely at seasons with similar data), my estimate is that the team finishing in second place this season will get about 86-87 points, and the team in third about 84-85.

Given that we currently have 53 points from 27 games, this would leave us needing about 34 points from the remaining 19 matches. Winning ten, drawing four and losing five would get us there – and seems to be quite achievable with the current squad – but only time will tell.

All of the information above is based on general trends and nothing can be assured. What’s for certain, however, is that whatever the outcome this season we can be hugely proud of the whole club for getting us to a position that many of us have not seen in our lifetimes.