Non-league attendances what I’ve learnt so far….

Long-time followers of this blog – if they exist –  will know that I have a little bit of an interest in non-league attendances. For a while now I’ve been scouring the internet and poring over old programmes trying to decode just what factors influence attendances at non-league games. Some things are, of course, fairly obvious like getting promoted or relegated which almost always has the expected effect on attendance figures, but there are also other factors, like the weather, or fish…………..

big fish

1.) The big-fish effect.

You could also call this the AFC Wimbledon bump, the FC United of Manchester  mount or the FC Halifax hump; a mathematical quirk which arises because in many cases non-league attendances are quite low and therefore just one outsize club in a division, the big-fish, can distort average figures for both individual clubs and the whole division. Just take Eastleigh’s average attendance of 723 for the 2008/09  season – down, in large part, to the 2283 who showed up to watch AFC Wimbledon game,  a large chunk of the crowd being travelling supporters. If we disregard that one game the average would in fact be a more modest 645. The effect on a league can be seen in the graph showing the change in average attendance. AFC Wimbledon were present in the division for the years 2005/06 and 2007/08 and during this time average attendances rose significantly, only to return to plummet once AFC Wimbledon were promoted.

The Isthman Premier; Home to AFC Wimbledon from 2005/06 to 2007/08

2.) Competition from nearby league clubs.

This is an interesting one. In the past  I have noted that in the cases I’ve looked at changes – even quite big ones – in the overall average attendances at league clubs do not seem to have any tangible effect on the corresponding crowd figures of their  non-league neighbors.  Initially I thought this was because league and non-league fans were essentially two different tribes, but I’ve since come to change my view – albeit only slightly

Average attendances at Bradford City and Bradford Park Avenue; High attendances at City do not result in Lower attendances at Bradford Park Avenue
Average attendances at Bradford City and Bradford Park Avenue; High attendances at City do not result in Lower attendances at Bradford Park Avenue

Whilst the season-by-season figures don’t show up anything looking at data for Portsmouth and Havant and Waterlooville over the course of three seasons (2009/10 to 2011/12) it seems that when there is a direct fixture clash Havant’s crowds are, on average, quite a bit lower – in fact the average attendance at Havant when Portsmouth are away is 787, but when the bigger club is at home the average drops to 577.


So how come there seems to be little evidence of competition when looking at the season averages? The first reason is that reason is that fixture clashes don’t happen that often; Over the course of three seasons there was only a clash nine times – so in areas like the Portsmouth region competition from league clubs fixtures directly clashing are therefore not a big issue and the non-league club may even benefit from the larger fan-base of a league club twiddling their thumbs when the league club are out of town. As for when the league club is losing spectators, the reason why so few turn up at non-league grounds  is that the supporters who abandon a league club when it gets relegated, or performs badly, are unlikely to then trade-down further to watch non-league.

So my original view needs a slight moderation, there are in fact several different types of fans; There are some who float between league and non-league and who tend to choose league above non-league when given the chance, These are relatively small in number, but still enough to be a noticable absence from a non-league ground. The other groups are league fans who will only watch football above a certain level, and non-league fans who either prefer watching their local club, whatever the level and finally there are those who prefer non-league over league football.

3.) Blame the weather

When I was watching Locks Heath play on a beautiful late-summer afternoon. I wondered how many of the, even then fairly sparse, crowd would be there when it’s freezing, or howling with rain. The answer is I suspected very few, if any. I also wonder whether this dynamic could affect non-league attendances, particularly as more than league-clubs non-league grounds are slightly more exposed to the elements. I organised the Havant and Waterlooville data from 09/10 to 11/12 to show average monthly attendances.

Over the three seasons attendances were, on average, higher for the milder  months.
Over the three seasons attendances were, on average, higher for the milder months.

Interestingly the graph shows that over the three seasons in question attendances are, on average, higher for the milder months. December seems to be a bit of an exception, but this can be explained by the fact that it is customary to play a local derby on Boxing day which may have the effect of inflating the monthly average. Though we can’t see what happens for individual games this does seem to lend some support to the view that attendances are to a degree influenced by the weather.

The weather has another affect. Waterlogged or frozen pitches mean postponements and postponements mean, typically, mid-week games. Using a regression model and data Havant and Waterlooville attendances over the 2009/10 and 2010/11 seasons we could predict that, other factors held constant,  a mid-week game would attract 150 fewer spectators.  It seems quite possible that a really damp, or freezing spell may reduce average attendance figures through both putting spectators off attending and through increasing the number of mid-week games. Perhaps a point to make is to question whether the non-league game would benefit from a winter break?

4.) Rivalry is good.

Fans love a derby and derby games at non-league typically see the biggest crowds. As well as the added interest of a derby for away fans it usually means only a short trip to watch their team playing. Again looking at the Havant and Waterlooville data for 2009/10 to 2010/11 the biggest single gates, both in four figures [1451 and 1020] were the games against local rivals Eastleigh. Using the same regression model as with mid-week games we can also predict that a derby game would, all other factors being constant, attract around an additional 190 spectators.

The number of derbies may also vary between season than another which can impact on average attendances; Certainly in the case of both Eastleigh and Havant and Waterlooville should either team change division it would be a hard fixture to replace in terms of pulling power and it seems likely both clubs would experience a drop in average attendances.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s