Remedial maths (and football)

We’ll say that me being ill was the main contributing factor to my dozing off during England’s latest draw against Slovakia – nothing to do with the performance, or the ITV coverage…

Whilst we’re in the knock out stage, further progress depends on a number of factors – injuries, tactics, player selection and performance on the day.

But possibly the most crucial factor of all could have nothing to do with Roy and the rest of the team.

I am of course, talking about, who will be broadcasting the match.

For years, I and a number of friends have had the feeling that England are more likely to lose when they’re playing on ITV rather than the BBC.

We hadn’t looked into it in any detail, it’s just how it felt from experience – I think Argentina in ’98 is the one that kicked it off for me (so can’t blame Adrian Chiles), and even lightly mulling it over now, I’m remembering Uruguay at the last World Cup.

So, what better way to aid my recuperation from some weird virus that’s mangling my joints and giving me wonderfully fluctuating temperatures, then testing the theory?

My wife thinks I’m quite the catch really…

Thanks to the likes of England Stats and another great site with details of football coverage in the 90’s I managed to put together a spreadsheet – Phils Frolics amazing spreadsheet of wonderfulness – Eng results (.xlsx) – covering all England games at international tournaments since Italia 90 (the first tournament I remember being an event).

Now, you can’t just go throwing stuff onto a spreadsheet willy-nilly and expect good answers, so I also devised a plan…

Approach

To do this in as scientifically reliable way as you can for such an endeavour, we need some parameters and explanations for what data I’ll be looking at.

The first thing to say is I’m only looking at the recognised international tournaments (Euros and World Cups) themselves, not the qualifying. Not only would that be a ballache to find all the data for, but it would introduce more broadcasters, and skew the data away from the focus on actually competing in a tournament.

As well as looking at the basic stuff like broadcasters and results (win, lose, draw) I thought I’d try to capture as much data as possible so included things like group/knock out stage and what tournament it was.

I also thought that there may be something to be said for what’s riding on the outcome of the match e.g. the opening group games are relatively low pressure, as there’s two games left to try to get through. But if you lose that, the second one becomes crucial in a way it wouldn’t be if you’d won. Ditto if you won the first two, the third doesn’t matter as much. So, I’ve included a column called ‘crucial’ which essentially covers games where the outcome directly impacts on progressing to the next stage.

By this definition, all knock out games are included, which led me to remove the third place final of Italia ’90 from the data as there was nothing really riding on it, so it’s a bit of an outlier that doesn’t sit with the rest of the games.

The results of drawn matches that went to penalties are classed as either wins or losses (usually the latter, as it is England I’m looking at).

I’ve also taken out matches where they were shown simultaneously on both channels – hopefully if such things happen again, my work will help inform your channel selection!

Disclaimer

As you’d expect with data collected from the internet by an ill person, I would assume there are errors in there which I’ve not yet found – so don’t take this, or indeed anything I say/have said, seriously.

Findings

The basic data shows that:

  • BBC have shown 22 games – 12 wins, 3 draws and 7 losses
  • ITV have shown 21 games – 5 wins, 11 draws and 5 losses

Which, in percentage terms, looks like:

Overall both

So, despite my hypothesis, ITV have actually had less (23.8%) losses in the matches they’ve shown than the BBC (31.8%)!

Though, if you want to see a win, the BBC are significantly ahead on 54.5% – almost double ITV’s 23.8%.

Also – if you enjoy the spectacle of two teams failing to best each other, ITV (52.4%) is the place to watch a draw rather than the more decisive BBC (13.6%).

So, slightly surprising, but there’s more to it that just raw numbers.

ITV do have slightly better record at Euro’s compared to World Cups – 30% of games they’ve covered were wins vs 18.2% at World Cups. Though they’re still worse than the BBC’s 50% at Euros (and 58.3% at World Cups).

Looking at all the ITV losses helps show where my hypothesis came from. These were:

  • Romania in ’98 – second group game, meaning we needed to beat Columbia in the third game (shown on BBC) to progress to the next round.
  • Argentina ’98 – the next round. I still can’t talk about this one.
  • Portugal ’00 – first group game, which was followed by the jointly broadcast match against Germany, which we luckily won (and BBC got largest audience), before losing to Romania (on BBC) and failing to progress to the knock out stage
  • France ’04 – first group game, followed by the 3-0 win over Switzerland (also ITV). Defeat of Croatia saw us through to the knock out stages, where we lost to Portugal (on BBC) on penalties.
  • Uruguay ’14 – an awful tournament, and this must-win game followed the opening match loss to Italy, so at least a point was needed. They also screened the irrelevant (for us) Costa Rica draw that followed, helping cement their association with England’s failures.

The first four of these were games I watched between the ages of 18 and 24, when I was most likely out watching with mates in the pub or at one of our homes in a big group. This probably explains where the theory came from – a bad run for ITV in an era when I was ‘growing up’ and shaping my idea about the world.

The last was just a crap tournament all round. No-one came out of it well.

We also need to think about which games the broadcasters are showing – usually the deal struck means one gets to show two of the group games, and the other then gets first dibs at the first knockout round. So ITV’s huge proportion of draws could reflect their tending to have two of the group games where usually a win and a couple of draws in enough to get you through.

Looking at the ‘crucial’ games discussed above, we get the below table:

BBC ITV
Win 46.2% 25.0%
Lose 46.2% 50.0%
Draw 7.7% 25.0%

This shows that for the crucial matches – the ones most intensely watched/felt, once again, BBC come out on top for wins, and ITV do indeed have a (marginally) worse record for losses. Saying that though, the actual underlying numbers are low as ITV only having shown 4 crucial games exclusively, compared to the BBC’s 13.

Again, the figures reflect that the BBC is more likely to be showing a crucial knock out game due to their having first dibs in the post-group stages – they’ve exclusively screened 6 in the knock out stages versus ITV’s 2.

Conclusion

So, what have we learned from this, other than I shouldn’t be left unsupervised during the day?

As my stats lecturers would say, you must be careful about assigning causality.

Do England lose because they’re on ITV?

Realistically, no, the channel through which images are beamed into our homes can’t be said to have an effect on the outcome of an England match, (and the numbers, suggest the BBC actually have a worse record).

Do ITV lose out because England aren’t very good?

I’d say so, yes.

Since Italia ’90, England have only competed in 13 knock out games at eight international tournaments (they failed to qualify for ’94 and ’08, and failed to progress beyond the group stages in ’92, ’00, and ’14).

Of these, ITV has exclusively shown only two games – that pesky Argentina match in ’98 (which is probably the ultimate source of the anti ITV theory among my generation) and Belgium in ’90. This gives them a 50% success rate compared to the BBC’s 33% (six exclusive knock out games, two wins, four losses) – which isn’t bad!

The real lesson, I believe, is that ITV get given/choose crap games. They’ve had a ridiculous number of group games (19 out of 21), and of them, not many (2!) were actually crucial because of either the number of games to go, or earlier results.

Of these two, one was at the last World Cup and the other the recent Slovakia game, which was borderline crucial for me as we had 4 points in the bag, but a thrashing might have made progress less likely depending on results elsewhere.

The results of these were a draw and a loss. Not a wonderful record, but also not large enough to be considered a trend, but recent enough to taint memory.

Now ITV have the prospect of hosting the Iceland match on Monday (their third England game of the tournament so far!) which might go some way to restoring their reputation, unless the sons of Óðinn continue their smiting, having quenched their thirst  on Ronaldo Tears.

Who knows, if we make some progress, I might do an update post in a few weeks – not least because I’ve already found numerous flaws in my working on this so I’ve had to rewrite it more times than I think is necessary for what was essentially a welcome distraction from god-awful daytime TV when I was ill and I can’t be arsed to do any more quite yet.

Advertisements

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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 )

Google+ photo

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

Connecting to %s