Are restrictions working?

I just found a site that has organized states by severity of lockdown, then graphed that against COVID death rates.

The graph in question is this one:

The graph’s X axis is lockdown ranking, with the least locked-down on the left, and the most on the right. The y axis is the death rate.

As far as I can tell there appears to be no real statistical difference between states that are locked down and ones that aren’t.

It’s also baffling that Texas has been open for a while jow and has some of the lowest infection rates in the country.

Anyone care to speculate what’s going on with this data?

Sure, it was written by a financial journalist. That’s most of the story in a nutshell.

Let’s start with the Dakotas. How did they rank on the death rate according to this author? (I’ll give you a hint, lowest and seventh lowest).

Good call! Imshould have looked more closely at the ‘death rate’. It’s not a rate at all, but the total number of deaths as far as I can tell.

I looked at a list of states ordered by population, and it closely matches their ‘death rate’.

That is not just selective editing, but a complete lie. Or, the authors don’t know what a rate is.

So everyone ignore this as being at all pertinent to COVID, but it’s a good example of how to lie with numbers.

It’s not even that. At least 6 states have lower death totals than either of the Dakotas, yet North Dakota tied Alaska for 1st place.

The rest of the study is a hot mess as well. They measure restriction rankings for that scatterplot by current restrictions. A month ago, they had Texas at 33rd. Today they have them at 5th. The latter is what they used for the chart. And the chosen restrictions and their weightings have “I pull this out of my ass” written all over them.

And of course, it completely ignores States “super spreading” events like Stugis S Dakota. Covid exploded across the midwest following that event. So, while S Dakota might be “average” for deaths, in part that is because “they” helped raise the average in surrounding states.

Just for shits and giggles, this study (PDF) uses WalletHub’s openness rankings, but does so over time and uses daily deaths per million as the death rate instead of whatever arbitrary thing WalletHub was using.

Still, the increasingly strong relationship between high levels of openness and high DDPM (Daily deaths per million) suggests that lockdowns have been effective in both reducing DDPM in highly infected states and in preventing new spikes in deaths. Additionally, this trend implies that states that are more open are susceptible to higher COVID-19 death rates. This will likely be the pattern until better treatments are developed or herd immunity is achieved.

Yet another factor is that any study should be controlling for differences like population density. The Dakotas should have extremely low rates, based on their low population density: The high rates there are all the greater testament to how poorly it’s been managed there.

I think there is more than population density going on, though. Texas’ numbers are baffling, and Texas has some big cities. Miami is not exactly small either.

Looking at a map of Covid infection rates, it sure seems to me like there’s a north-south bias. You see that in other conntry data as well. Climate must play a part to,some degree. We know that Covid lasts a lot longer on surfaces when cold, for example. I wonder if it survives longer as an aerosol when floating around in colder air. The super-spreader events at meat-packing plants might support this.

There is just a lot we don’t understand about how the virus is spreading and mutating. It’s a complex system.

Then again, in the US, you’ll also see a fair bit of north-south bias in politics, and hence in both the restrictions put in place and in compliance with them.

But sure, average temperature is probably another thing that a good study should try to control for. I didn’t mean to suggest that population density is the only possible confounding factor-- That was just the first one that came to mind, because it’s relevant (in the same direction) for all infectious diseases.

Well, yes, but right at the moment the results we’re seeing are the opposite of the ones most people would predict from this fact – cases in the South, with the exception of Florida (the southern part of which is in more or less a climate zone of its own), are still trending downward, despite lifting of mask mandates in several states, lower vaccination rates across the region, and a generally more nonchalant attitude toward restrictions of all sorts. Meanwhile in the upper Midwest they’re still going up, and they were also going up in the Northeast, at least until recently. I guess you could make a plausible argument that fewer cases earlier = more cases now (and this might explain the differences between, say, Michigan and Wisconsin right now), but it seems to me that the more likely explanation is that factors other than human behavior and policies are at work.

I would mention what it is going on in Arizona, while mask mandates have been lifted, all stores I have gone in Arizona are having close to 100% mask use. Virtually all stores and people going to them are ignoring what the stupid governor has done.

Sure, but the title of this thread is an area in which we seem to have gained some knowledge. Yes, restrictions appear to be working.

It seems reasonable to me that the states with the least restrictions/most early spread are seeing declines now because most of the people who were going to get infected already got infected and now they have post-infection immunity. At some point you burn through most of the people who don’t care about taking precautions (or are actively hostile to precautions) and your rate of spread is going to slow down because those super spreaders have self-immunized.

Good point. I don’t think legal health mandates, in U.S. states, have more than a marginal influence on behavior. If you want to study impact of masks, or restaurant closures, you would have to measure on the ground behavior.

It’s disappointing that we don’t have better evidence one way or the other.

Yes, if only there were any evidence that masks, social distancing, closure of mass events, etc. had a positive effect on reducing the spread. It’s too bad that we’re stuck with guesses and opinions.

Is there anything close to definitive? There are small studies that show certain things work in small cases but is there any hard evidence that the lockdowns have been effective? It’s really hard to pin down with so many variables: population density, weather, age, demographics. There have been spikes around the holidays and (I think) the Superbowl; that would seem to indicate that anti-lockdown behavior leads to infections. I would have thought that we’d have pretty good comparisons between areas that have locked down and not but I have seen any.


Masks, avoiding crowds, and distancing in general all have been demonstrated to be effective, over and over again. Population density, weather, age and demographics has nothing to do with any of that (density might have some unavoidable aspects simply due to many people living in a large building). They only become major factors when people don’t do the right thing. If everyone wore masks, avoided crowds and distanced themselves, we’d likely have hundreds of thousands (or even millions) fewer deaths than we have today.

Any study of effectiveness is going to be complicated by the general abundance of the virus in the population masking what is going on. However, the Australian experience, based on a generally well-maintained zero community case rate, is that when there is a breach from quarantine into the general community, a three day lockdown, targeted mandatory quarantining of contacts and cranking up more stringency of masking etc, then you can stop this highly contagious disease from spreading very effectively. It seems to stop the spread of covid within a week or so.

It does work, but has to happen within a regime of lots of testing, serious contact tracing and largely consistent community support.

Thanks all. I don’t want to make it look like I’m against restrictions; I think wearing a mask is a no-brainer and I have no empathy for the my-rights! people. I support most measures that are expected to help (like masks). It’s easier to support measures with proof that they work. At the macro level it seems like the only countries that have made this work are the island nations. What are the non-island countries or states that have done better than others?