Football in a pandemic

Had a quick gander at the football news this morning and here’s my report:

Football wants to start back up on June 1st. The reason June 1st is important is because there are some 180 or so players with contracts that expire June 23rd (source: BBC football daily) and they need to finish this season before then or they will have to renegotiate some contracts.

Anyway, the big plan here is to test players and staff at regular intervals in order to maximize safety. Wolves (that’s the Wolverhampton Wanderers if you’ve forgotten) have already started this program, testing 10 staff.

These schemes have been met with quite a few questions. The UK Government has to approve of sport returning and the shadow minister for sport wrote a letter saying that the topic should be debated in Parliament because we are after all, still, a democracy and in a democracy you debate things. Safety of people and the players (players are people too!) are of the utmost concern here. Transmission rates for the virus are extraordinarily high and while footballers themselves are unlikely to die from the disease (Office of National Statistics in the UK shows that 92% of deaths have occurred in folks 60 and older) they don’t want to pass the virus on to their children, parents, or other loved ones.

The players for their part are split: some eager to return and some concerned. Not wanting to contract the virus is a valid concern. There are a lot of unknowns about this virus. For example, what are the long-term health implications for people who recover? Viruses are funny little things. Some of them linger for decades before unleashing a new hell on our bodies (Hepatitis C, Chicken Pox, etc.) so you should understand if no one wants to just go ahead and contract this virus. But of course also, the players don’t want to be the site of some fresh outbreak which kills a bunch of people. That would be a heavy burden. Football already takes extreme precaution to prevent fan deaths; 96 died at Hillsborough which led to many measures to make stadiums safer for the supporters.

Even the Premier League doctors are questioning whether football can be made safe. A group of four docs working for PL clubs wrote in to the chief medical officer of the League and expressed concerns in a series of questions.

This is, of course, a bit of an existential question. Clubs could (I’m going to go ahead and say that a lot of clubs will) fold without the broadcasting revenue promised to them. Players could find themselves out of work as well. And of course, I’ve been glib about this in the past – suggesting that if a club like Bournemouth were to fold, someone would surely buy it – but it’s not guaranteed that the club would be bought back or that some players would play football again. And that’s a huge loss for the fans and folks who work at these clubs.

What’s also unknown is what will happen to the Premier League itself. Many clubs have been itching to start up a European Superleague for decades. If the Premier League starts to break apart, and especially if other leagues start to collapse, some clubs could band together and form a 20-team league of the richest in the world. While that seems like a doomsday scenario you can bet your life that Arsenal would chomp at the bit to join.

That is, if there will be an Arsenal. As a club the Gunners are hugely dependent on matchday revenue. And with an owner who – despite increasing his personal wealth by over $200m thanks to the stimulus program – is extremely unlikely to put his hand in his pocket to put money into the club. Though I expect he’d do it in a pinch to save his investment. So while other clubs like Bournemouth need the TV money, Arsenal needs both. And while I don’t expect Arsenal to fold, it is a remote possibility. And while I would expect some other rich person to see the massive Arsenal fanbase and lick their lips at the prospect of getting this club cheap, there’s always the small chance that something catastrophic could happen.

That’s the football news, folks. There is absolutely zero reliable transfer news, which is as it should be because the club just begged the players for money and spending money on transfers after that would be hilarious and probably cause a player revolt.

I’ve had a series of what I think* are odd interactions over the last week. People, folks, family members, loved ones, and others telling me that COVID-19 isn’t that dangerous. And specifically that the death rate is 0.24% (2.4 deaths per 1,000 infections). I’m not sure where people are getting this number; it could be some propaganda machine, it could be that they are taking one small sample like Hong Kong (4 deaths in 1000 confirmed cases or 0.4%) and deciding that’s the actual death rate, they could just be “doubling” the seasonal flu death rate (0.1%), they could just be jerks, they could just be using cruise ship data (and using misreadings of that data), they could be misreading the stats**, a combination of things, or something else.

The fact is that we don’t know what I would call the “lab death rate” is – the percent of mortality if we put 100,000 demographically diverse humans in a lab, infected them all with the exact same strain of the virus, gave them all the exact same treatment, and measured the outcomes. I feel like it would be probably ~1%, but the problem is that the real world doesn’t operate like that.

Whenever I mention case fatality rates, people love to point out the outliers. But as you know in any statistical study there will be outliers. One poster on Twitter mentioned that Finland did a sample study for antibodies and they found that nearly 3% of the people in the study had antibodies for COVID-19. Extrapolating that out to the entire population gets you a case fatality rate of 0.2% in Finland.

But let’s work that backward for a minute and see if the number makes sense. Ok, if 3% of the USA had the disease, that would mean that 10 million Americans have had or do have the disease (we have 1.3m confirmed cases). And if that many people had the disease and if the case fatality rate was a mere 0.2% (10,000,000*0.002) we would have an expected deaths of 20,000 or so.

We know that case fatality is 80,000 as of today so, and using the Finland infection rates, even if 10 million people have had or do have COVID-19 that would put the case fatality rate at 0.8% (0.008) or 8 times as deadly as influenza. That would also mean that if 1/3 of the US got infected over the next year with an 0.8% CFR we will see about 960,000 deaths. And by the time we reach “herd immunity” – if that’s even possible, we don’t know if the virus will mutate or if people can get it more than once – we will see 2.7m deaths in the United States.

What I find frightening, however, is the actual death numbers and how closely they align with the 0.8% CFR I’ve dreamed up above. The Office of National Statistics in the UK releases a weekly death report. Their latest report shows that for the week ending 24 April, their total deaths count was 21,997, that’s a 52% increase in deaths from the standing average for the last five years. And of those 12,000 additional deaths, 8,237 had COVID-19 mentioned on the certificate.

Some people will point to the ONS data on “respiratory deaths” which is significantly lower – 1,900 – than all deaths tagged with COVID-19 but again, that’s completely unfair because if you get COVID-19 and you had an inflamed liver and die of organ failure, you wouldn’t have died right then without the COVID.

The USA had 2.7m deaths in 2019 and if we double that number, which is in line with what we are seeing in the UK caused by COVID-19, and we do not find a cure/vaccine, or take measures to flatten the curve, we could see 2.7m additional deaths in the United States for 2020. That’s oddly in line with the 0.8% CFR but probably a coincidence.

I will mention that the deaths so far have largely been among people 60 and older. Not just largely, 92% of them in the UK are 60+. I say this only for completeness. This isn’t me saying “oh well, it’s just the olds” I care about all life in the abstract and in my personal life I have dozens of older folks I love deeply. It’s me mentioning this stat because it matters in the response: how we deal with this going forward.

We don’t know exactly what the case fatality rate is because we don’t have enough testing – though the USA’s testing of 1.3m is incredible (though 39th in world testing on a per capita basis) – and because we haven’t been testing for antibodies (to see who’s had the disease without knowing). Official numbers worldwide show extremely high case fatality rates, 14% in the UK and Italy, 7% globally, 5% in the USA, 4% in Germany, and 2% in South Korea.

There are only a few ways forward on this.

• The South Korean/German model – rapid and widespread testing followed by contact tracing for positives, strict quarantine for infected with follow-ups by medical personnel to determine whether the infected needs hospitalization. Hospitalization at the first sign of trouble. Provide universal healthcare and make sure folks have food and money to pay rent. This seems to reduce death numbers and leads to the fastest possible re-opening of the economy. Apparently the Germans are even going to start playing sport again this weekend (still can’t believe it’s going forward).
• The UK model which seems to involve strict quarantine and yet the virus is still spreading at what looks to be a stable rate though that could be because of better testing. They also have universal heath care and seem to have also found a way to keep unemployment numbers low and to provide the unemployed with a social safety net (ish). They are, however, starting to send mixed messages and telling people to go back to work and cutting the furlough benefit soon. We don’t know what the death rates will be from this model, but they are loosing restrictions now and we will have to wait a few months to get a clearer picture.
• The USA model – testing like crazy (total number of tests, test per capita are 39th in the world) but loose quarantine restrictions and various rules for each state which has some places just opening things back up now. So far we have the highest total number of tested and deaths. This seems the most reckless and overall death numbers will probably be unbelievable.
• Another way? Quarantine for the elderly and constant testing for their caregivers?

We will eventually return to something we call “normal”. What that will look like I cannot predict at this point. I can say with some reasonable certainty that the virus will eventually become part of our everyday lives, like the Flu. And hopefully, like the Flu, we will get some vaccines for the virus. If not, well, I guess wearing a mask at all times, washing hands constantly, and playing “is this seasonal allergies or COVID-19?” will become the new normal.

*Do I really have to say “what I think”, ever? Isn’t it clear that everything here is “what I think”?

**The death rate in South Korea is 2.4% which is 0.024 numerically and a lot of people (normal folks) struggle converting whole numbers to percents and back.

Qq