Towards Datatainment – Working with Data Gods

This post and all other business-focused posts are now being held on the Seven League blog.

 

The following is an article I’ve just written for Leaders in Digital Sport‘s “Leaders” bulletin. The data gods in question are Gavin Fleig and Ed Sulley of the MCFC Performance Analysis team.

Do you know a bit about football? Or a lot? Can you remember the score of a game 5 years ago when you were standing in a stadium on a wet wednesday night and your team lost 3-2 after going down to 10 men in the 63^rd minute?

That made-up match might not be the right details, but if you’re a real football fan then you probably remember something similar – as well as a selection of other random details from times past and fixtures forgotten.

Does that make you a statto?

Most people will react badly if you call them a statto. It has become a pejorative term – as portrayed by Skinner and Baddiel‘s Fantasy Football League comedy sidekick or maybe long before that – and although there’s a grudging respect for the feats of memory that sporting stattos display, there’s also a lot of laughter behind the communal hand.

The stereotype is that stattos remember facts. Data. They don’t necessarily remember meaning or turn that data into knowledge or, better still, understanding, let alone the peak of the information pyramid: wisdom.

So data’s got a bad rep – it’s the domain of stattos and geeks.

But I have seen football data that delights.

Over the last 6 months I’ve been working closely with the Performance Analysis team at Manchester City FC. Gavin Fleig and Ed Sulley have been informing and enthusing me (a footballing muggle) with their explanations and demonstrations. They are sporting data gods. 

Every movement of our players in a match, in the gym and on the training pitch is recorded, converted to data, analysed and presented back with expert understanding. It forms part of the whole picture that informs training and performance improvements. No-one here is called a statto and yet they all accept their statto status.

The reason why I’ve been working so closely with them is because I think they hold the solution to a challenging problem for football. It is this: How do you bring together people with a lot of knowledge of the game and those who only have a little? The fan from Manchester and the fan from Minnesota. The established and the uninitiated, or, as I like to call them, the veteran and the newcomer. Of course a Minnesotan might actually be a lifelong soccer fan, it goes without saying, and there are definitely people in the UK with low levels of knowledge (myself included) but generally speaking the average Mancunian knows more than the man in Mumbai.

Frank Skinner, Ed Sulley (Statto) and David Baddiel. (ok, so it’s Angus Loughran who played Statto)

You could argue for hours about what makes a ‘fan’ vs a ‘supporter’ or ‘follower’ but leaving those definitions aside, I believe there are a couple of major challenges for any club that wants to increase the number of people who care (to whatever degree) about the activity of the club, but have a range of experience of the game: 1) veterans do not necessarily welcome newcomers and 2) it is tricky to talk to fans with varying levels of football understanding.

I don’t think that Gavin and Ed have the solution to problem no.1. But I think that data, digital media and performance analysis might hold the solution to no. 2.

In my view, broadcast media is bad at addressing multiple audiences. It picks a target audience and aims at it. In the case of football coverage in this country, the BBC’s MOTD and Sky’s MNF address fans who are closer to the veteran end of the scale. There is little explanation for the layman.

Digital media, on the other hand, is perfectly suited to explain the game to the uninitiated. We have restrictions about what we can do with match footage, but we can use performance data, put it in an application that allows the user to choose a comfortable level of knowledge and let them play with the information. What’s more, they can do it at their own convenience and at their own speed.

This is what takes data into a new sphere – we can move from data journalism through data visualisation and, hopefully, to the holy grail of engaging an audience: data entertainment, or (forgive me) datatainment.

Gavin and Ed have been able to show me how the data can tell a story which gives me greater understanding of the strategic context of a game. I now appreciate some of the chess-playing tactics of managers and the expertise it takes to carry out those plans.

Over the next few months at City, we’ll be starting on a journey towards datatainment. We’ll begin by launching prototypes that start to make the performance data transparent and, with the help of the audience, we’ll progress towards making truly entertaining data-led products.

I can’t wait for the audience’s reaction and I’ve no doubt it will be voluble whether in praise or pillory, but if even just one stato and one newbie both get information and perhaps enjoyment out of our prototypes, then we will have started off in the right direction.

Advertisement

LiveRugby app – second screen or settling an argument?

This post and all other business-focused posts are now being held on the Seven League blog.

 

An iPhone app that has live rugby union data for the 6 Nations and that promises to fulfil my second-screen needs. I had high hopes of you, LiveRugby app, and lovely though the data detail is, you don’t overcome the challenge of entertaining me with data. Or even engaging me. Datatainment delight is still a way off.

Let me go back a step.

I’ve always been more of a rugby follower than a football follower, so I was excited to see that there was a new app, using Opta’s incredible wealth of live data, which would be delivering an experience on the iPhone for the 6 Nations competition.

It’s nice enough. The interface feels all ‘chalk-boardy’ and it’s pretty straightforward to use. I think the interesting stats, the Opta stuff, is too far down the IA. I didn’t download this as a ‘general information about the 6N app’… I downloaded it specifically for in-game data updates so a) make it easier for me to get to them b) set a default auto-refresh that i can change the frequency on, or can turn off – having to hit the refresh button is another user-friction feature.

Oh, and “Don’t have a tv? Never mind watching the game. Use our LIVE commentary!” says the promo copy. Really? You mean I  should read less-than-twitter-length updates? In the Ireland v Wales game that has just finished I counted 71 updates across the course of the match. Bearing in mind it was a fantastic game and so much was happening every 60 seconds, there’s no way you’d want to stick with the app updates. Even if you couldn’t get near to a tv or radio, if you’ve got a smartphone you can still get to the BBC mobile site (there they go, impacting the market again!) for an experience that’s more like the excellent TMS coverage.

I think there’s a fundamental problem with the concept.

Is it trying to be a second-screen app? or a data-resource app?

Use case 1: I’m in the pub and I want to reference back to a match from last weekend and prove an argument about the distribution of rucks, the success of kicks, etc etc. In this case, the app will be great. It still makes the information more difficult to find than I want, but it’s all there.

Use case 2: I’m watching the game / listening to the game and I want to use the app as a second screen to add to my experience of the match. This is where the app stumbles.

There is a problem with the presentation of data. It’s passive. I don’t like passive data. Passive data is difficult to work with. If you can turn it into information, that’s better – but to do that you will need to make some kind of comparison. This app does that – some of the key information, like the fact that Scotland made 68 tackles (all successful) against England’s 153 (of which 14 unsuccessful) is, indeed, presented. But again, it’s done in a rather passive way.

Data – Information – Engagement – Editorial – Entertainment

So you’ve got the data, you’re informing me… but I still have to work hard. You’re a long long way from entertaining me. It’s the editorial layer that’s missing. England were forced to defend by Scotland’s attacking efforts – but a combination of what Gavin Hasings called ‘schoolboy errors’ combined with some resolute English defence meant that Scotland couldn’t convert their opportunities into points. That’s the story here – that’s what the app should be pulling out – that’s what editorial intervention would add.

But if this app is trying to offer passive data, or to inform the pub argument after the event – then it succeeds.

If it’s trying to enhance my in-game experience, then it fails.

There is, of course, the fundamental point that bringing the appreciation of data into live games is difficult. Jonathan Davies on the BBC Sport coverage has a man beside him trying to put together interesting insight into the performance but there’s very little time to do so. Little production time and then little air-time – so his insights have to be very simple and ‘top level’. Of course Davies and the BBC experts could go into great detail, but there just isn’t the time.

Second Screen vs Speed

The same is true for users of second-screen apps. Football and rugby are both fast-moving games and there isn’t time to be digging 4 clicks down into an app and then hitting refresh to get passive information. Sunset and Vine may have done it in cricket, and many US sports have much more time within the flow of the game.

But if the second screen is going to work in football and rugby, then data and editorial will have to work much more actively together to be able to enhance the in-game experience and engage, and even, dare I say it, entertain the audience.

Stop Making Football Data So Much Hard Work!

This post and all other business-focused posts are now being held on the Seven League blog.

 

This is a quick, and ranting, post.

This morning on of our excellent Performance Analysis team here at City pointed me in the direction of the ESPN Gamecast centre for the match on Sunday where City beat Arsenal. It’s nice, don’t get me wrong. But only up to a point. Then I put my “I’m not that much of a football fan” hat on … and I can’t help but spill out reasons why it doesn’t work for me.

So, here’s the 13 things that jump to mind. [Warning, but the end, it’s really winding me up]

1. I want it full screen
2. I expected a ‘play’ button to show me stuff. Maybe it could just put one tab after another – morphing along to show me the different things. that would be fun and it would show off the stuff for me like a little preview.
3. Speaking of play buttons – where’s the video that should be linked into the gamecast – do i have to go somewhere else? this is espn, right?
4. I clicked players and then silva and it looked like his heat map was all over the pitch – but then i wanted to compare it with the arsenal opposite number – only i don’t know who the opposite number is so i had to ask someone. i also had to ask what RM meant beside the name. Once I’d been told it was Right Midfield, i then compared silva’s heatmap with Alex Song’s… but I was aware that this might not be a good comparison. Is it? Isn’t it? and how the hell would i use this thing to find out? Why are you making this difficult for me to understand things.
5. I want to compare with previous competitions between arsenal and city.
6. I see no point in having a tab for ‘goals’ when there was only 1 and this screen shows me nothing.
7. I want to be able to zoom in on a section of the pitch to see in more detail – then average position seems interesting, but they’re all clustered together and i have to be a surgeon with the mouse to identify different players.
8. Where’s the colour key for the bottom? why are some names in grey?… oh yeah, because they were subbed. there’s probably a colour key somewhere… but why are you still making this difficult?
9. I want to see a timeline of the game… to see it developing as it goes – and i want to be able to see it speeded up (or slowed down as i desire)
10. I’m never going to read the text on the right. Why not compress it into a timeline so that key moments can be highlighted … i need it to show me key sections of commentary, not give me every blow (unless i want to dig that deep)
11. The Report, Quotes, Commentary, Match Stats tabs are useless.. i barely even saw them.
12. The whole thing is presented down the page after reams of nav… just give it to me in a nice pop-out… or at least in a clean page.
13. Can I embed this on a blog? can i share it with friends?
14. STOP MAKING ME WORK SO BLOODY HARD TO USE THIS THING – I WANT TO ENJOY IT… HELP ME ENJOY IT! I am a lazy sod, so just analyse and interpret on my behalf and then entertain me. ok, thanks, bye.

Ok. So, now that I’ve calmed down…

I suspect ESPN are well aware of all the options and all the complexities of user experience. This is, actually, one of the nicer implementations of this kind of data… But as I’ve previously banged on about, I still think it has a long way to go.

As time goes on and our work at City looks into the possibilities of datatainment it is becoming clear to me that the audience perspective is the essential factor. I may not be the target for ESPN’s gamecast, but despite football’s estimated 3.5bn fans, I still think some of them and the other 2.5bn people on the planet would appreciate a bit of an easier ride.

Either that or I’m just lazy.

Strictly Come Datataining – More sequins in sport & the Entertainment Layer

This post and all other business-focused posts are now being held on the Seven League blog.

 

He looked at me quizzically.
“There are times when you come up with ideas and you’re just not sure whether they’re great – or just shit,” I said.  “And sometimes the more you say them, the more they sound like they could be a good idea – but you always wonder if they’re shit and no-one’s telling you. But with this idea, either people I don’t know are being terribly polite, or they’re quietly laughing, but some of them have definitely made encouraging noises.”

More data in Strictly? More sequins in Sport?

I was, of course, talking about datatainment. Hugo Sharman seems like a very nice chap. I don’t think Deltatre North‘s head would laugh in my face or behind my back – but he was looking quizzical. “Why don’t you tell me, when we get to the end of this, whether you think we should be contributing some ideas for datatainment,” he said, because I don’t think he could work out whether it was a wanky term for something that’s already being done (and that Deltatre do very nicely for lots of clients including the Uefa match day centres) or whether I meant something else.

The truth is, I don’t know. Yet. But I’m feeling confident that we are going to find something new. It’s all about the entertainment filter.
When you look at football match-day centres, or many other sports for that matter, you are confronted with a lot of data visualisation. If you’re a fan and you love your stats, then it’s entertaining. But for me, it makes me have to work too hard. That’s because I’m not a football fan. When it comes to cricket or rugby, I love the challenge of understanding the information – but it sometimes is a challenge. And there’s definitely a kudos in being the one who knows Dan Carter’s kick conversion rate – but a little like I do wish those annoying IBM ad buffers would shut up, I also get a little bored of stats knowledge being something that divides the audience – splitting it into the know-it-alls and the novices. I have to work too hard to get enjoyment out of data for football … I want it made simple, easy, enjoyable. Perhaps this is the effect of sub-optimal data vis. Maybe better design would make it naturally more engaging. But I think it’s more than that – visualisation can tell stories – but I’m not sure it can entertain in the way I mean.

Nice match centre, shame about the result

Talking to Hugo, I suggested it was the difference between watching the Young Musician of the Year and X Factor. But that’s not right – too much genre change and skill variance alongside the entertainment increase. Maybe it’s the difference between Come Dancing and Strictly Come Dancing.  No, we do not need more sequins in football. But when they re-invented the ballroom dancing programme  a few years back, the BBC did a remarkable thing – they managed to popularise the artform in a way that no-one had done since Strictly Ballroom had pastiched and parodied in the great 1992 Aussie film. You didn’t have to know a passo doble from a samba (I can’t even spell them right), and by pairing celeb novices with experts, they gave it the common touch. No wonder John Sargeant and Ann Widdecombe did so well. The judges are a mixture of amateur and expert – but even Craig whassisface, the most pernickety of the lot, explains the detail of the moves.

They opened up ballroom dancing and welcomed in newcomers who knew nothing – surely something that football (and maybe all sports) could do better – and especially something that data visualisation could do better in sport.

By the time I got to the end of my ramble/rant at Hugo, the quizzical look was gone and had been replaced by a smile. “I have to admit, I hadn’t thought of it that way, but I think you might be right – maybe there’s an editorial entertainment layer there that we are missing.”. So maybe we are onto something.

In a way, the editorial layer I’m talking about is something that tv punditry alongside performance graphics has been doing for a while. Sky are doing more and more of it, the BBC does plenty (and Deltatre provide that) and even the woeful ITV Rugby World Cup commentary tries to give it a whirl with their touchscreen. What I don’t want is a situation, as ITV have, where subject experts, former and current players, give poorly-enunciated opinions while pressing the odd player image on a touchscreen. I want the data, but I want the opinion, expressed entertainingly, and maybe even with some interactive element. And I want it now.

Cristiano Ronaldo: the Datatainment star for Sky Sports

This post and all other business-focused posts are now being held on the Seven League blog.

 

OK, so the use of “thrilling” music might be over-egging the pudding somewhat, but there’s some lovely super-slow-mo and data capture from bio-mechanics and a variety of scanning-gizmos in this programme on Sky Sports.

The Castrol Edge site has more, or there’s the twitter hashtag too.

Four parts:

Part 1 – Body Strength: http://www.youtube.com/watch?v=s7vYfKfI87U
Part 2 – Mental Ability: http://www.youtube.com/watch?v=21AFrsXa7E4
Part 3 – Technique: http://www.youtube.com/watch?v=1vyqz9OtTJw
Part 4 – Skill: http://www.youtube.com/watch?v=FbNQ_kCE6V0

Defining Data Visualisation, Data Journalism & Data Entertainment

This post and all other business-focused posts are now being held on the Seven League blog.

 

After my blog post from last week, Data? Entertainment? You need Datatainment, and a comment about datatainment seeming very close to data visualisation, I thought I’d try to define my thinking on how these terms differ.

• Data + Visualisation is the process of making the data visible – the representation, such as Nike + or the use of epic, relies upon the data itself being informative or entertaining, but the process of visualisation does not apply an editorial filter. Of course, any design belies the application of some kind of filter – but with visualisation there is an attempt to represent the ‘real’ events or facts.
• Data + Journalism is using that visibility to investigate an issue or point – it may not necessarily visualise the data for the user, but will use the techniques of visualisation in the journalistic process.
• Data + Entertainment is where you use data as the primary source of entertainment. You might choose to make the visualisation of raw data entertaining or perhaps use data visualisation as part of the process of entertainment – but there’s definitely a strong editorial control which is focussed on entertaining the audience rather than exposing data.

Let me try to give examples:

Data Vis: Showing the number of check-ins on location based services foursquare and facebook check-in by MCFC fans over the course of the season.

Data Journalism: Using the visualisation to explore the patterns of LBS behaviour on match-days and therefore the stress on local transport infrastructure.

Datatainment: During the pre-match show at City Square, profiling the fan who’d checked-in at the Etihad Stadium the most over the course of the season. Or, bearing in mind that we have a huge number of people who visit us online during a game, perhaps we might also represent the global locations of our fans – maybe using foursquare’s ‘swarm’ approach to identify key hubs of City fandom outside Manchester so that we can bring those fans closer to the match day experience.

There are better ideas of what to do with Datatainment for City, but I won’t write about them here as I’d rather talk to the fans and see if we can deliver them in the real world.

Data? Entertainment? You need Datatainment

This post and all other business-focused posts are now being held on the Seven League blog.

 

[In which I justify the creation of a horrible new term that describes where data and entertainment meet – in this case, through football and it’s fans.]

Datatainment is a new thing. Linguistically it’s a bastard son of Data and Entertainment, with it’s irritating cousins Gamification, Informercial and Advertainment. But I think it’s more than just a stupid word dreamt up to try to kick-start a dull meeting. Could it be a new genre? How does it differ from data journalism? And what does it have to do with football?

Though the journalist in me wants to write a fully thought-out report – the blogger will throw up some thoughts and see if you’ve got anything to say.

Way back

Many years ago I did some work that pioneered online data journalism. I only realised this the other day because, at the time, we didn’t call it that. I was just working on the ‘specials’ team for BBC News Online and 3 of us were dedicated to the Kosovo Conflict special report. One of my jobs was to get the overnight press release from the allied command which detailed the locations and targets of overnight bombing raids. Then I’d make sense of the data, and go to work with Max Gadney, then one of the designers, to turn it into a map which represented the data, but also registered the nature of the target. You may remember, this was the conflict where many claims were made about laser-sighted accuracy – and some of those claims were later found to be overblown. Schools and hospitals got hit – and the data showed it over a month of intense bombing raids.

Little did I know that, many years later, data journalism would become a ‘thing’. A thing many others have studied, researched, honed and blogged about. Something that you can do a course on or can feed with a well-focussed FOI request.

Data Journalism – it’s not Entertainment. It’s not Datatainment.

In my role as Head of Digital for Manchester City FC, I was meeting with a major global hardware and entertainment brand yesterday. When asked what was my next move in innovation – my answer was clear – to make the most of the data. We will lay down a data architecture, and we will invent data products to sit on top. We will use it to entertain and engage. We will use it to involve a wider audience. We get 40,000 people in our stadium, but we get at least 4 times that on our online Match Day Centre (MDC), but the big difference is that the capacity of our stadium is limited. Of course, the capacity online is not.

I’ve whinged about the use of ‘gamification’ as much as the next man. It’s fitting that my spellcheck on this machine has just corrected that word to ‘ramification’. The inimitable Margaret Robertson (@ranarama) has written, among others, just how mis-placed the awarding of points misses the point of how games and game theory can be employed to the benefit of a project. And that’s to say nothing of the badgification (!) that we’ve seen with foursquare and, seemingly, every other venture in the last year. But at least it’s not ‘greenification’ that I heard used at the Power to the Pixel conference keynote last year.

So, what the hell and I doing creating another of these sodding hybrid words.

Datatainment. it’s better than ‘datafication’.

Let me tell you a story. At the drinks after last year’s Story conference I met the chief creative at an award-winning Brighton agency who’d done lots of work with the BBC, Channel 4 and others. I thought to myself ‘excellent, I bet there’s some great stuff he could do for me at Manchester City… I wonder if he’d like to come and have a look at what we’ve got and talk about some opportunities.’

But when I told him that I worked for a football club, the disappointment was palpable. The light in his eyes died and I could see him thinking of a way to get out of there. But I am persistent, if nothing else. I gave him some facts: We have roughly 40000 people who turn up every week to the same place. More and more they turn up for a full day out – and the MCFC experience is one that really moves things beyond turning up a five minutes to 3pm, watching the game and going straight home. Our 40,000 season card holders are passionate and engaged with our brand (forgive me football fans, but you know what I mean). They’re representative of the population in terms of smartphone penetration or computer usage. AND, the vast majority of them carry the seasoncard around with them – a seasoncard that uses an oyster-card-like RFID chip to identify them and store some basic information. Oh, and because I thought it might come in handy, these 40000 peoples’ cards now also have personalised QR codes on them and, just because I could, they have augmented reality markers on the back too. Is that enough to do something cool and interesting with? No? then I told him some more…

In the last couple of months we’ve launched a tiered system of membership to the Club – and for the first time that includes a beginner-level membership – Blue. One of the main audiences that will be most interested in what this has to offer is the overseas audience – the audience who could never get to the ground but still want to be involved. And these Blue membership cards *also* have the RFID, QR and AR elements built in. So where we say 40,000 people have cards today… by the time the ex-pat and overseas audience gets involved, we’ll be talking many more. So, suddenly you’ve got a massive, engaged, excited audience focussing on the same series of events and all with personalised, data-driven, tech-useful clever cards.

Now, if you can’t do something exciting and creative with that lot… Can you imagine the amount of data coming off those cards? location, activity, frequency… and that’s without even adding any other levels of gaming on top. but I didn’t need to labour it. By this time he got the point and could see that maybe football was worth taking a closer look at.

What’s interesting is that this kind of data is just the beginning.

In terms of football data, what we get is phenomenal.
– opta stats – public and professional level telling you everything from a player’s possession rate to his average sprinting speed. and there are also the prozone and venatrack systems available. All this match, and training, performance tracking feeds into the Performance Analysis team – and some of it also goes live into our web-based user-facing Match Day Centre.
Of course there’s also the data we generate from tracking site usage – english and arabic – and on our social and media presences, twitter, Facebook, youtube and flickr.

Then there are things like Fanvision – which has a major impact on user experience in the US market where getting stats on a game, during a game, creates a remarkable added-entertainment layer to American Football and other sports.

Statos have always been present in sport… celebrated in some, shunned in others, but always present. F1 – motorsport in general – massively data dominated. Cricket has the home of sport data – Wisden. Golf is basically a data-driven sport.

And then there’s our partnership with EA SPORTS for FIFA12. Do EA have any data we could work with, do you think? 3m people bought the console game in the UK last year, 10m worldwide. Over a 100m people have purchased it. And another 8m play the online version of the game every year. And do EA have stats to play with … data you could use as prediction engines… data you could use to entertain? It’s already gearing up to be a fun partnership.

That’s Datatainment

So I’ve gone from data journalism to data entertainment – datatainment.
Is this anything new? perhaps not.
If you define data as pieces of information – then you could argue that quiz shows have been examples of datatainment for years. But you’d be wrong.

I can’t remember who said it – and I’m probably misquoting – but there’s a difference between data, information, knowledge and wisdom. At which point it’s beholden on me to also quote the saying “Knowledge is knowing that a tomato is a fruit. Wisdom is knowing not to put it in a fruit salad.”

Data is less than the information that’s shared or tested in a quiz show. Data is the raw, unadulterated fact. Moulding those facts into stories – like @mattsheret does so well for last.fm in his role as Data Griot (data storyteller) – is data storytelling. Using facts, like the data from the bombing raids over Kosovo in 1998, and plotting them on a map to expose the detail of whether those laser guided missiles actually *were* as consistently accurate as claimed… or the homicide map of New York… or David McCandless’s work through @infobeautiful and the Guardian …  or using spreadsheets and databases to bring up anomalies in MPs expenses… or the other work showcased at Design of Understanding… these are some of the many excellent examples of data journalism. But when it’s not journalism – when it’s used to inform but, primarily, to delight or inspire or to bring someone closer to a subject they love and give them that warm glow of passion and fun… I submit that this is something else. It’s data entertainment. And that’s what I want to bring to football with @MCFC.

Are there other examples? I’m sure of it. And I’d welcome your thoughts on the nuances I’m missing or the examples that prove or disprove the case.

I may not be allowed to return to London – at least not uninjured – after @matlock @jaggeree and @ammonite promised to do me harm if I kept using the datatainment. But I’ve compounded that now so I will just have to bear with the ‘deadarmtainment’ or the ‘chineseburntainment’ that have been promised. I suffer for my art. Actually, speaking of art – the rather lovely @artfinder app for ipad is, well, art and entertainment, isn’t it? what do you think? Arttainment?

—-

John Kearney’s comment below prompted an attempt to clarify my thinking: Defining Data Journalism, Data Visualisation and Data Entertainment.