Comment by gus_massa | Hacker News Reader

Comment by gus_massa | Hacker News Reader

Parent Full thread

[spoiler alert]

With mask/max10/closeHR/CloseUni you can get a good enough result for the first wave until the summer holidays. Perhaps you can reopen a little more.

(Max100 also work, but the hospitals are too saturated and I get too worried.)

The problem is the second wave after the "cold weather" arrives. Be prepare to hit "full lockdown" as soon as it start increasing. Open when a little when he wave disappears, but you may need to close again.

Continue with that until the vaccines arrive. With this simple strategy I got 32.3K deaths and a cost of 441B https://koronahra.cz/results/60e63272fecc84001a33a31a (Not in my first try.)

Someone got 6.4K death and a cost of 268B, that is much better.

Someone got 2.3K death and a cost of 342B, that looks like a good tradeoff.

Someone got 40.1K death and a cost of 141B, that is very cheap if you don't care abut people.

The graph in the finish scree has the result of other players. It's interesting to see that there is a lot possible improvement. (Perhaps also luck, some events are randomized.)

[spoiler alert]

I found that fully closing schools works better than closing high-risk, and you can leave business fully open for significant periods. My best is 2,582 deaths at a cost of 385B. https://koronahra.cz/results/60e6550661cdc10019e509ce

Masks and max 10 stay on the whole time. I close unis at +100 infections, close all schools at +200, close high-risk at +400, and essentials only at +800. Ease up on the restrictions when infections go down. I ignored those breakpoints and left things as open as possible during the warm weather period, since you can safely do that as long as you kept the initial outbreak under control.

Edit: 2,112 deaths at 380B, just targeting +75/+150/+300/+600: https://koronahra.cz/results/60e65b5dfecc84001a33a3a3

Edit 2: 1,691 at 405B using the 2112 strat with slight adjustments based on intuition: https://koronahra.cz/results/60e65cc061cdc10019e509e6

I still don't understand why (UK) schools weren't closed sooner (for the general population, excepting children of key workers). I pulled my kids out a week before the first lockdown, it would have been perhaps a week sooner but it's illegal to do that in the UK.

It's well known schools are a major contributor to spread of diseases (and I did read some papers on flu spread in schools prior to making my decision to check the verity of the common wisdom on schools being Petri dishes of disease!).

Here in Argentina the government of the City of Buenos Aires says that the transmission in schools is very low (they are opened, but they include mandatory mask, temperature check, cleaning with alcohol everything, weird staggered enter and exit time to avoid contact between different classes, and every time a student sneeze the class get closed for a week or something like that.)

The National government and the government of the Province of Buenos Aires that control the ¿20? satellite cities [1] (with a total of x3 more population than the city) disagree, so they closed the schools.

After some time, the city closed some the classes of the oldest kids, and the province opened the classes of the smallest kids. And perhaps both made additional opening and closing, I lost track now. (If you have some spare popcorn and can read Spanish or are fine with an autotranslation, you can use Google to get more info. It's heavily politicized, so take each news source with a grain of salt.)

Anyway, ... it's not so clear how much is the spread rate in schools, and it's also not clear that they are using the correct coefficient in the model in this game. My guess is that we will need like 10-20 years to get an accurate model, and meanwhile we should try to make some educated guest.

I think the model in this game is accurate enough to decide between schools and business in real life, but it's useful to get an intuitive feeling of other stuff, like that you should not wait until the hospitals are at 100% of capacity to start a lockdown.

[1] https://en.wikipedia.org/wiki/Greater_Buenos_Aires

I got 3600 dead at 1.2T. I guess there’s ways to improve on that by coming down hard on the first wave, where basically 70% of my deaths happened.

I find it hard to impose a lot of restrictions when there’s basically 10 people per day infected, but suddenly it’s 500+/day.

There was/is definitely a lot of people prone to what I call the Seatbelt Fallacy - 'why do we need seatbelts [airbags, ABS, etc.] when deaths in car accidents aren't even that high!??' - which becomes "we didn't need lockdown, hardly anyone died" ...

Reply on news.ycombinator.com