Great Expectations

On Sunday 15th July the 2018 World Cup Final kicked off after a month of sweaty Russian VAR-checked action. France won it beating a resurgent Croatia 4-2 after some highly questionable refereeing decisions.

But you know that.

The real question (i.e. the question we’re going to waste our time on now) is:

Was it what we expected?”

And the answer, should you be too lazy / busy / entirely disinterested to read further is: Mostly.

Netherlands v Northern Ireland, 1978 FIFA World Cup Qualification
Netherlands v Northern Ireland, 1978 FIFA World Cup Qualification. Not relevant, but aesthetically great.

* * * * *

Russia 2018. The FIFA World Cup. 32 teams supplement their training camps with borscht, visits to towering Soviet era monuments and general confusion because “…isn’t Russia meant to be cold? What about those hats?! I bought three!!!”. Fans from 32 nations debate with each other and the rest of the world about how well they will do, have done, could do next time and should be doing year-in-year-out. The media feed on the fatty morsels of ‘expectation’ served with a crusty side of pessimism (or a lovely optimism salad, depending on their agenda and appetite for extended metaphors).

As results billow in the post-mortems are quite clear on whether expectations have been missed, met or exceeded…

Germany, to get slightly ahead of ourselves, spectacularly missed expectations.
But you don’t need two separate tables of data to tell you that. Though we’ll get to that glory soon…

Expectation.

Expectation is riddled with context.

There is, for example, almost no expectation that I will be the President of Tuvalu in the year 2050. This is for many reasons, not least because Tuvalu does not make room for the role of President. However, if we were to find ourselves in 2050, the office of President having now been passed into local law, its election process based on a knock-out Super Mario Kart competition that only I and 73 other people have entered all of whom are known to be rubbish on Rainbow Road… then the expectations are very different (because I OWN Rainbow Road).

Before the World Cup, the expectation that Germany would go out in the Group Stage was lower than an ‘legend level’ limbo bar (a stick on the ground). After losing to Mexico in the opening game there was negligible expectation, but possibility. About 75 minutes into the final round of games with Sweden leading Mexico 3-0, South Korea giving nothing away, a complete lack of energy from the Germans and 15-20 minutes left on the clock… then the expectation became tangible.

Of course it doesn’t always cook so slowly. Expectations for ‘Sbornaya’ (the hosts) rocketed after they beat Saudi Arabia 5-0 in the opening game. And rightly so. Having looked likely to fumble themselves to a devastating early exit during their last competitive outing at the Confederations Cup, they proved that home advantage means something when it matters. And those expectations, of course, were well surpassed…

Great Expectations.

The expectation pre-World Cup was that Brazil would lift the trophy.

I say this not as my opinion, or as a general feeling from the masses. I say this with certainty. This is because I’m simply stating what the market believed. The market, in this case, being sports betting.

What’s useful about the odds offered by bookmakers on sporting events is that they essentially measure the expectation of what will happen in a given match and give it a quantifiable ‘score’. If something is very likely to happen, then it will have short odds meaning that if you bet money on it, you won’t win very much back. If it’s very unlikely to happen (either statistically, such as there being 50 corners in one football match, or based on cold hard information, such as Patrick Thistle making the Champions League final before the end of the decade), then it will have long odds.

Bookies become very rich because they put a lot of time and effort into determining the odds they offer to ensure that, over the long term, they are right and ‘things-that-aren’t-meant-to-happen’ don’t happen. Because when they do it costs them LOTS OF CASH. And it’s this reliance on data plus the crucial driver of money-at-stake that make the odds for a sporting event very reliable as an indicator of what’s meant to happen. AKA, what the real expectations are for the outcome. Something the partisan press, your mate Barry or ‘a feeling you had in the shower’ are less able to do.

On June 18th, the odds on which country would win the 2018 World Cup looked liked this:

  • 🇧🇷 Brazil: 4/1
  • 🇪🇸 Spain: 9/2
  • 🇫🇷 France: 7/1
  • 🇩🇪Germany: 8/1
  • 🇧🇪Belgium: 10/1
  • 🇦🇷 Argentina: 12/1
  • 🏴󠁧󠁢󠁥󠁮󠁧󠁿 England: 16/1
  • 🇵🇹 Portugal: 22/1
  • 🇺🇾 Uruguay: 28/1
  • 🇭🇷 Croatia: 30/1

Info taken from The Week article based on Oddschecker data.

Expectations…

But that was the 18th June. And the world was a very different place when, on 15th July nearly a month later, 11 Frenchman took to the field in Moscow to take on Croatia. On that day the odds broke down like this:

🇫🇷 France @ 6/5 | Draw @ 2/1 | 🇭🇷 Croatia @ 3/1

France were *expected* to win.

Odds from Odd Portal’s averaged odds for the World Cup.


A word on odds.

For the rest of this article, despite what I’ve done above, I’m going to use decimal odds (HK style, like the best sweet and sour dishes) simply because it makes it very easy to compare lots of numbers at a glance, not least if you’re not used to looking at these things. Fractional odds, as used above – 4/1 for example, ‘four to one’ – are the usual go to which is why they’re quoted above, but as things get tighter, they become a bit (or a lot) trickier to follow.

When it comes to decimal odds (and to clarify, and for the purist, while we’re using Honk Kong odds, there are slight variations used in the UK, Europe, US and beyond),  then the closer to 0 the odds are (the lowest they can be, 0 would literally just get you your money back) the more likely it is believed that result will happen. So 0.5 is expected to happen more than 1.5. Got it? Good.

I’ve used Odd Portal’s averaged odds for the World Cup to ‘write’ this article. This is a store of the match odds (i.e. for the result after 90 minutes) in the minutes before each game kicked off averaged from a number of bookmakers based around the world.

According to their data the most sure-fire of results came in at 0.21, for 🇪🇸Spain to beat 🇮🇷Iran and 🇧🇷Brazil to beat 🇨🇷Costa Rica. Both did. If you’d put £10 on those games at those odds you’d have found yourself with the princely sum of… £12.10 at the final whistle. A profit of £2.10. Get in the Um Bongos.

Once the odds get out to around 3.0 – and admittedly I’m going to get a bit unscientific here – you’re reaching ‘less likely’ territory. And by the time you hit 6.0 you’re well into ‘unlikely’ territory. But of course it happens.

The longest odds to win in Russia 2018 were the 18.48 odds for 🇰🇷South Korea to beat 🇩🇪Germany in the last round of group games. If you’d foreseen that (or, more likely, your finger had slipped when placing the bet), then you’d have collected £184.80 profit from your £10. Comfortably enough for a flight to Frankfurt to console fans of Die Mannschaft.

To get the most from this article (and I’ll admit, like much Nerdius content, it’s a tough squeeze) that’s all you need to know. That and that life IS NOT too short to engage with this stuff.


Now let’s get back to context. And, because they had nothing to do with the World Cup and won’t sully our data, let’s talk about 🏴󠁧󠁢󠁳󠁣󠁴󠁿Scotland. Because a) they never do so this won’t really age and b) they’re the team I was born to lament.

There are, I think, four different contexts when discussing the expectations of a football (or I guess any) team:

HISTORICAL EXPECTATION: What normally happens?

We lose to the minnows and put up a fight against decent teams.

GENERAL EXPECTATION: Where should we be?

In the World Cup final! We have more people than Croatia or Uruguay and the ability to invest in facilities. Plus we managed to produce Kenny Dalglish, Denis Law and many others, so we can again. We just need to get our boys playing competitively abroad and sort out the domestic structure and it will come.

ADVANCE EXPECTATION: What’s likely to happen next week, next month?

We’ll lose and it will be gutting.

IMMEDIATE EXPECTATION: What’s likely to happen later today?

We’ll lose but will still go to the bar.

The self-deprecating Scotland fan chat aside, these four contexts often get muddled when people talk about expectations. Indeed, I’d contend that if we want to have a useful conversation about whether teams have failed to meet or surpassed the expectations set for them, then the only thing that really matters is the ‘immediate’ context.

You can, to use a footballing cliche, only play one game at a time. And you can only beat who’s in front of you. Not only that, but expectations live in the very real here and now, not in some fantasy parallel universe where there’s been more investment in the game, where a world class player turns out to have a Scottish grannie and where Scottish clubs were subsumed into the English system a la Swansea and Cardiff. This is true for the fans of course, but even more so for the players and their support staff who work hard and don’t want to come up short.

Historically, the Netherlands have an excellent team. Best team never to win the World Cup basically. But their current team didn’t even qualify. It’s pointless to base expectations on what Cruyff et al did, just as it is to pine for the team who made the final 8 years ago and finished third at the last outing. They’re old. They’re gone. The new lads may as well be wearing clogs.

In terms of the General expectation based on, say, ‘available players’, the top 10 nations by raw numbers are: China, USA, Germany, India, Brazil, Mexico, Indonesia, Nigeria, Bangladesh, Russia. This is from FIFA’s ‘Big Count’ a survey of all 207 member nations on the numbers involved in the game. Sure, economics come into it too, but 🇧🇩Bangladesh – who have the 9th biggest football playing population in the world – are currently ranked 194 in the world by FIFA, while 🇨🇲Cameroon (49), 🇨🇮Côte d’Ivoire (68) and 🇸🇳Senegal (27) all have a lower GDP per capita. Just some top line stats of course, but its enough to tell you they tell you nothing.

Advance calls on what might happen are useful and based on the current ‘real’ scheme of things, but things can change quickly: Injuries, a great or terrible performance in the lead-up, a sacked manager, an owl roosting in the ground. It all matters.

It’s therefore only the ‘Immediate’ context, the judgement call made in the final moments before the game with all the info at our disposal that can be said to really represent whether we should be overjoyed, or over-angry at the end result. Of course injuries or strange refereeing decisions happen mid-game to change the course of history too, and so expectation can change mid-game, but we’ve got to draw a line somewhere. Or at least, I’m going to.

And so, after all that, let’s get to the data.

Below is a table with all of the results from the 2018 World (that’s 64 games) and the odds from just before kick-off for a win for either team, or a draw. One caveat, as mentioned above the odds are for the 90 minute result. In the knock-out rounds I have taken any game that went to penalties to be a draw. If a game was won in extra time then I’ve counted it as a win (though this is only relevant for one game).

Note: Actual result highlighted by cell colour. ‘Favourite’ result (if not actual) shown by orange text. See the Expectation Key for a traffic light style totally made up system of my devising for how ‘likely’ each result was. 

Some things of note:

  • 39 games met expectations, 25 games did not
  • Of the 25 games that did not meet expectations 14 still had results that were pretty likely (i.e. it was seen as a close contest)
  • Of the 25 games that did not meet expectations, 6 could be termed disasters / heroic (more accurately were over 4.0 odds wise)
    • 🇦🇷Argentina’s draw with 🇮🇸Iceland
    • 🇲🇽Mexico beating 🇩🇪Germany
    • 🇯🇵Japan beating 🇨🇴Colombia
    • 🇸🇦Saudi Arabia beating 🇪🇬Egypt
    • 🇪🇸Spain’s draw with 🇲🇦Morocco
    • 🇰🇷South Korea beating 🇩🇪Germany
  • From expectations of TOTAL 1.14 Germany games went 25.12 (and was 28.32 until Tony Kroos score his last minute free kick against Sweden). YIKES
  • Almost all upsets occurred in the Group stages. While 8x games were ‘unexpected’ results out of the 16 knock out games, only one – 🇪🇸Spain v 🇷🇺Russia – was a clear upset (3.16 for the draw prior to Spain being eliminated on penalties)
  • You may also note that England were actually favourites to beat Belgium in the group game, which was the only anomaly I found when cleaning through the data. However Belgium were more open about fielding a weakened team and England had been looking nifty. Further research confirms this was the expectation of the masses.

Of course, the real question we want to uncover is which teams at the World Cup over performed and which under performed. And for that, this table is far more useful…

Note: This is a simple traffic light system. Red = didn’t meet expectations, yellow = did, green = exceeded expectations (all based on the data from the previous table, of course).  In the first column I’ve given the teams an overall ‘colour’ rating based on their game-by-game results. In the second column is the ‘Advance’ expectation (unscientific, my view based on general chatter about each team). Under EQ I’ve made up a fancy ‘expectation quotient’ that means NOTHING but looks confusing and thus swanky (it’s just exceeded:met:missed). In the final columns I’ve marked ticks for if teams exceeded or failed to meet expectations on at least one occasion. It’s simpler than I’ve made it sound. 

Some more things of note:

  • 🇪🇸Spain failed the most. But only because they managed to fail twice and still make the KO round, so they could fail again
  • 🇪🇸Spain’s 4x failures comes in just ahead of 4th placed 🏴󠁧󠁢󠁥󠁮󠁧󠁿England, who failed 3x times (and never exceeded expectations game-by-game)
  • It’s therefore, as in England’s case, very possible to far exceed advance expectations while actually failing to meet expectations game-by-game.
  • Three teams did just as expected and went out in the Group Stage (🇵🇦Panama, 🇹🇳Tunisia, 🇵🇪Peru)
  • 🇰🇷South Korea, 🇨🇷Costa Rica, 🇦🇺Australia, 🇮🇷Iran and 🇸🇦Saudi Arabia over performed, but still went out in the Group Stage
  • 14x teams met or exceeded expectations in every match (which means 18 teams dropped the ball at least once)
  • Of the finalists, 🇫🇷France ‘met expectations’ throughout (except a dodgy draw with 🇩🇰Denmark), while 🇭🇷Croatia were a right mixed bag
  • As such, any of the top handful of teams (🇫🇷France, 🇧🇷Brazil, 🇪🇸Spain, 🇩🇪Germany) could only ever really meet expectations, or flop
  • 🏴󠁧󠁢󠁥󠁮󠁧󠁿England, 🇭🇷 Croatia and a handful of others show that you can underperform on a given day but still go far as long as you can take a decent penalty (or save them), or at least slip up at the ‘right time’.

* * * * *

And breathe…

So them’s the numbers. Have a poke around, berate my colour coding, marvel at 🇲🇦Morocco, 🇮🇸Iceland and 🇸🇳Senegal who managed to miss, meet and exceed expectations in their triplets of games – the holy trinity (for literally this article and no where else).

But what all this faffing about with odds really tells us is that while the media’s colouring of expectations aren’t quite accurate, the bare game-by-game results data is only useful in making the kind of technical-but-tedious point that won’t get you invited to many parties (obviously examining the data and finding weird coincidences (like the fact 🇧🇷Brazil’s results when stated as ‘.’ for didn’t meet and ‘-‘ for met  – no need for ‘*’ for exceeded – paints an appropriately sad face: .—.) is totally normal).

Expectations in football are bigger than that.

🏴󠁧󠁢󠁥󠁮󠁧󠁿England (to use an example close to my house) were seen to have had a brilliant World Cup. Which is on balance fair. But when you step back and look at it they actually failed to meet expectations in three matches our of seven, and one of those was in the semi-final. So they really *should* have done better. They really *should* have been in the final (where it would have been expected to all go a bit ‘Taillebourg‘).

But then they won a penalty shoot-out.
And that was HUGE for England and their fans.
And it made a Radio 1 DJ point this out.
Which is wonderful.

Football is the most popular sport in the world precisely because it doesn’t always follow expectations. Whether it’s Denmark ’92, Leicester 2016 or The Miracle of Istanbul; the ‘Hand of God’ (or his second), the Panenka or the Milla wiggle. It’s the moments beyond the results that defy expectation but create their own world where the prize for your commitment is something wonderful you’d never quite see coming.

This World Cup had Ronaldo’s free kick against Spain and Neuer slipping against South Korea. It had ridiculous strikes from right backs, Panama’s first ever World Cup goal and it had the first teenager to score in a final since some lad called Edson Arantes do Nascimento. But it also had Senegal and Japan fans cleaning up the stadium. And it had this Brazilian fan in tears of joy along with the rest of us…

When you take them together it’s all of these intricate, all-encompassing narratives swarming together that fulfil the expectations of fans around the world. Sure, we never bring ourselves to  expect these things from great sport – after all it’s the knowledge that this could be another drab 0-0 that makes the drama possible – but it’s the expectation they can happen that lives in every kick and every chant and every stupid VAR-bothering replay on a pitch-side TV that looks like it should be showing a rotating graphic of what’s on the stadium carvery and an exhortation to ‘book early for Christmas’.

Because while it’s easy to kill an afternoon pondering the black and white reality of ‘success’ with stats and facts and data (and Lord is it easy to complain and goad as required), it’s the expectation that there’s so much more to sport that keeps us coming back.