Premier League teams on Twitter (or why Liverpool wins the league and the Queen might support West Ham)

Have you ever wondered where Premier League football teams draw most of their support from? Or what the geography of fandom is? We have too, and set about to better understand how Premiership teams are reflected in Twitter usage across the UK.

The Floatingsheep team, with the help of two researchers from the Oxford Internet Institute - Joshua Melville and Scott A. Hale (both of whom did most of the work) - have created a neat interactive map for you to both explore the geography of Twitter mentions of specific teams, and let you explore the patterns of five key rivalries. Click on the screenshot below to be brought to the full interactive map. 

The data used include all geotagged tweets mentioning any of the Premiership football teams and their associated hashtags (e.g., #MUFC or #YNWA) that were sent between August 18 and December 19, 2012. We have only included one tweet per user to prevent 'loud' fans from skewing the results. The users were then aggregated to postcode districts in order to see a fairly fine-grained geography of results. The number of tweeters per district is normalized by the total 'population' of Twitter users based on a 0.25% random sample of all tweets within the UK. 

What do the data show us, you ask? In Manchester, for instance, there is the oft-repeated stereotype that Manchester City are the 'real' local team, while Manchester United attract support from further afield. Our map doesn't really support that idea though. There are only a few parts of Greater Manchester in which we see significant more tweets mentioning Manchester City than their local rivals. We also, strangely, see more mentions of Manchester City in Scotland and Merseyside, and more support for Manchester United in Northern Ireland.

The Merseyside rivalry (Liverpool vs. Everton) is another interesting one to map. There we see that Liverpool have the slight edge in the postcode that is home to both team's stadiums. However, there is no clear winner in the rest of the region: with most postcodes having a fairly close split between the two teams. Interestingly, many postcodes in Scotland seem to have more mentions of Everton; while many in Northern Ireland have more mentions of Liverpool.

We can also zoom into particular postcodes and see which teams are most mentioned there. The  academics in Oxford (for some strange reason) mention Manchester City more than any other team. Central Edinburgh (when not focusing on Hearts or Hibs) has more mentions of Everton than any other Club. And the Queen's home of SW1A goes for West Ham.

What about maybe the most important question of all. Who wins the league based on total number of Tweets sent from anywhere in the UK? The answer is Liverpool (a team that hasn't won the actual league since 1990).  Manchester United are a somewhat distant second, joined by Everton and Tottenham in the Champions League spots. We also find out that Fulham, Swansea, and Wigan are the three teams that get relegated due to their quite abysmal scores. Apparently just not that many people want to tweet about Wigan.

There is no doubt that using Tweets as a proxy for fandom is messy and not always reliable. In other words, we are mapping mentions and not measuring sentiment. But, the data do give us a rough sense of who is interested in (or at least talking about what), and where they are doing it from. It allows to begin to counter myths (e.g. that Mancunians don't support Manchester United), develop new insights about places that we don't necessarily have good data about, and most importantly, have some guesses as to which team the Queen might support.

See also:
A broader take on how information augments place (a second paper on the topic can be accessed here)
Other examples of our Twitter mapping (racism, flooding, earthquakes)
The code behind this visualisation (made freely [CC-BY-NC-SA] available on Github)

The Online Manchester Derby

The Manchester Derby, a heated football game between Manchester City and Manchester United, has been hotly contested since 1881. United have recorded the most wins overall, and indeed the most success in general. However, City recently (by some measures) became the world's richest sports team and have invested heavily in order to become outdo their local rivals.

But how has this rivalry taken form online? Has the new found wealth (and relative success) of Manchester City given the team more visibility than their more successful rivals? And, is the commonly repeated assertion that City fans are from Manchester, but United fans are from elsewhere reflected in online visibility for the two clubs?





The maps show that within Britain, Manchester United are far more visible than Manchester City. There are only 18 postal code areas in which there are more references to City than United, but 92 in which the opposite is true. The main cluster of areas in which City outdo United are unsurprisingly located in the Northwest, but interestingly within Manchester itself it is United and not City that are characterised by the greater degree of online visibility. So, at least in online world, it remains that Manchester City can't quite yet claim to be the only real representatives of Manchester.

Recapping our Predictions for the World Cup or Why Floatingsheep rocks!!

Now that the dust has settled after the world cup it is time to reflect on our predictions. Our language sensitive ranking system based on geo-coded references to football/soccer correctly identified three out of the final four countries in the world cup. The only error was predicting England over Germany. Although one can suspect foul play from the English member of the Floatingsheep collective, there is a methodological explanation as well.


Our searches are limited to land, as there are very few placemarks in water. While this works in a general sense, it does exclude references to football by a country's aquatic citizens. While this is miniscule in most cases, the case of Paul the Octopus suggests that Germany may have a sizeable underground (or better phrased, underwater) population of football enthusiasts that were missed in the data.

Or maybe Mark just cooked the data so England would win. Still you could have done much worse if you used our predictions.

After all, using our system we did make about $17,000 dollars via offshore betting. Unfortunately, we have already spent it all on gumdrops, Botox and aquavit. Clearly we're just not cut out for life in the fast lane.

More World Cup Than You Know What To Do With

We're now two weeks and 48 games into the World Cup and transitioning from the opening round to the round of 16. Last week, we gave you our alternative rankings of the sides in this year's World Cup, based on the number of Google Maps references to "football" in each country, as a percentage of the total amount of content.

And while our rankings certainly caused a shake up in terms of where teams stood in the pecking order of the footballing (or is it soccer?) world, they didn't seem to have too much effect on the outcome of the matches themselves. Frankly, it's probably a good thing you didn't bet the farm on our upset pick of South Korea over Argentina (Argentina won that game 4-1). However, were you to have used our rankings to bet on Serbia (#15 by FIFA, #8 by FS) over Germany (#6 by FIFA, #14 by FS) last week, you would have gone away quite rich. Likewise, with just .09% separating them, it's no surprise to the Floatingsheep collective that our #1 and #2 overall teams, Algeria and England, played to a draw.

After recognizing the flaws in our system, however, we've come back with a new way of ranking the sides. Like any map or statistical analysis, we were forced to exclude some things in favor of others, and much to our detriment. Because our original rankings used references only to the term "football", and not local linguistic variations of it, our rankings were highly skewed. For instance, the world's #1 team, Brazil, was ranked dead last of the 32 teams in the World Cup by our rankings.

So we come back with a new set of rankings, based on the local variations on the word "football" - from "calcio" in Italian to "futebol" in Portuguese, we've taken a finer grain approach to our newest series of rankings, seen below.

With these new language-based virtual rankings, a number of countries have improved their position. While Algeria, England and Cameroon remain #1-2-3 in our rankings, traditional football powerhouses Germany, the Netherlands, Argentina, Spain, Portugal, Italy and Brazil all do substantially better when taking into account their local terms for the game we've all been adjusting our sleep schedules to watch for the last two weeks.

Of course we continue to face methodological issues with this newest set of rankings. Because we don't have a complete set of data including the Arabic, Danish, Greek, Serbian, Slovak and Slovene words for football, we were forced to use the generic "football" for each of these. While Slovenia, Slovakia and Denmark take a hit from this oversight, Algeria, Greece and Serbia have all managed to come through it unscathed, still ranking considerably high, given their disadvantage.

Based solely on picking the team with the higher ranking in our new, language-sensitive ranking system, we're going to make the following predictions for the knockout round portion of this year's World Cup finals:


Although England has the toughest row to hoe in facing the #4 team and, potentially, the #7 team in our rankings, we're going to go ahead and pick our highest ranked team to go ahead and win it all. Based on the matchups, expect #5 Uruguay to make it to the finals, with the Netherlands at #6 edging out the #10 Spaniards for third place.

If you're a betting type and you win big, we'll collect our share at a later time and date. If you happen to lose big, we sincerely hope you won't hold it against us...

Alternate World Cup Rankings

Now that the World Cup is into full swing we figured that we should revisit some of the earlier work that we did on the cybergeographies of football (see here and here). In the table and map below, we've calculated the proportion of all geotagged content in each of the 32 countries competing in the World Cup that mentions the word football.

(click on table for closer view)

And for those more visually inclined, here is the data in map form.

We are able to see that the amount of online interest in football (or more specifically, the propensity of people create content mentioning football tagged to a specific part of the planet) rarely correlates to a country's FIFA ranking. Brazil, for example, which is the top ranked team in the world is at the very bottom of our rankings. Only 0.02% of content in Brazil makes reference to the word football. This is an even lower percentage than North Korea! Of course, our earlier post on the topic did find a lot of content mentioning the Portuguese word futebol so this is almost certainly a linguistic issue.

Will these rankings go on to replace the official FIFA rankings? We'll just have to wait and see. But, it is worthwhile noting that yesterday's unexpected Swiss victory over Spain and the Uruguayan victory over South Africa can both be explained by these cyber-rankings. So we're investing the vast fortune (otherwise known as a deficit) we've made via floatingsheep.org t-shirts on a few side bets. Hmmm...perhaps South Korea to win against Argentina? You heard it here first!

Perhaps most interestingly, Algeria and England are first and second in the rankings (with 2.2% and 2.1% respectively). So, we'll have to see if Algeria live up the their reputation in the cyber-rankings in the match between the two sides tomorrow.

Football (or is it soccer?) in nine and a half languages

On Monday we created a map illustrating the geography of virtual references to the words "football" and "soccer". In today's post, we've added eight more languages into the mix: German, Portuguese, Dutch, Spanish, Japanese, Korean, Thai and Chinese. The map below visualizes which of these various ways of referring to "football" are most visible at any particular location in the Google Maps database.

What struck us most was how the map reproduces expected patterns (based on language groups) with very few exceptions: most points in Korea reference the Korean word for football more than the same word in any other language. The same thing is true in Japan, Thailand, Brazil/Portugal and every other country associated with the languages that we conducted this batch of searches in.

Ultimately, Australia wins the prize for having the most homogeneous footballing cyberscape. There is only one place in the country with a reference to football in a language other than English: A reference to Fussball (German) somewhere around the vicinity of Alpine National Park in Victoria. Perhaps there is some sort of odd colony of football playing Germans (is there any other kind?) in this National Park (would any Aussie readers mind checking up on this for us?).

Sweden and Poland are interesting cases: a diverse mix of references to the sport in English (both "football" and "soccer"), German and Spanish, with a small smattering of Dutch and Portuguese. Of course, if we had searched in Swedish or Polish the results would likely have been otherwise.

English appears to be the dominant language for references to the sport in most parts of the world with no direct connection to one of the languages in which we conducted the search (e.g. in Iran, Finland and Russia). We should also point out the the French word for football is "football," so it is difficult to distinguish between references made in English and French using this keyword.

This map is about more than just a sport. We are interested in using this method to study and map cyberscapes in a range of languages. This map was just a first step to test some of the boundaries of the method. We will eventually be mapping a range of other terms in a lot more languages in the near future. Suggestions are welcome.

p.s. This may be a dagger in the heart of many calcio loving Italians, but despite having won the World Cup four times we simply forgot to do a search in your language. Ci scusiamo. We don't know what we were thinking.

Football vs Soccer

We've decided to wade into an age-old debate. Should football/soccer be called football or soccer? Consulting Wikipedia doesn't really help, as there are twenty one pages of argument debating this very question (they finally settled on the term "association football").

In order to usefully respond to these debates, we figured that we should try to develop a useful understanding of the geography of "football" and "soccer." So our foray into this argument will be through a couple of simple maps. The first shows the relative usage of the word "football," while the second relative usage of the word "soccer."

References to "Football"

References to "Soccer"

Some interesting patterns are evident on these maps. Much of Arabic and Anglopone Africa stands out with a large number of references to both terms (in fact, "football" and "soccer" are some of only a small number terms that we have thus far encountered that have a comparatively high number of hits in Africa). A quick sample of the hits for football and soccer in a few different African cities reveals a range of geographic associations with the terms: bars that show football, soccer pitches, conversations that were had about football, recorded memories of a soccer game that took place, and of course references to the forthcoming World Cup in South Africa. Yet it remains unclear why Libya, Sudan, Tanzania and many other countries on the continent score so highly.

In some cases, a large number of hits for either "football" or "soccer" does seem to be correlated with on-the-pitch successes (either historical or contemporary): see Uruguay and Iraq as examples. Uzbekistan is another interesting example. Surprisingly, the country is home to a relatively large number of virtual references to "football," despite not really being the first country that springs to mind when thinking of the sport (a la Brazil, Germany or England). The country has a low total number of placemarks, but it is likely the relative prominence of FC Bunyodkor (an intriguing case of billionaires, politics, and sky high ambitions) that has sparked a lot of online interest in the footballing side of Uzbekistan.

None of this really helps us resolve the football vs. soccer debate. So we made one final map illustrating the contrast in the number of references to both terms.

What does this map tell us? There are only a few places in the world in which there are more references to soccer than football (Brazil, South Africa, parts of Japan and South Korea, and Canada). Interestingly, Ireland, the US and Australia all refer to association football as "soccer" and yet have far more references to "football." This is likely due to the fact that all three countries play their own regional versions of the sport: Gaelic football, American football, and Australian rules football.

Stay tuned, as our next post will look at how linguistic differences in virtual references to football/soccer in Google Maps are reflective of the differences in language use in the offline world (or perhaps not?).

New York Yankees and Real Madrid

Today brings another look at the geography of interest in sports teams as represented in Google Maps. Note the concentration of references to the New York Yankees in New England. While it might seem strange that there are so many in Boston (home of the Red Sox), it is important to remember that we are simply counting any references to the keyword in question, including negative ones (which surely exist in abundance in Massachusetts).

References to the New York Yankees

And for the rest of the world, we offer a map of references to Real Madrid (a football/soccer team for the unenlightened, NOT some kind of MTV reality show). Apparently they are very popular in the U.A.E. and Spain (although the latter comes as little surprise).

References to Real Madrid

Sometimes it is best (and easiest) to simply let the maps speak for themselves....so that's it.