You can add location information to your Tweets, such as your city or precise location, from the web and via third-party applications. You always have the option to delete your Tweet location history. Learn more
we can also see what happened in states with no lockdowns. it's pretty striking. there is just no way to slice that data that makes it look like SD or SIP work. p values on not working have 4 zeros before the number... it's more consistent with them doing harm.pic.twitter.com/PXbeQzQzFu
many have posited that this inverse relationship is driven by worse outbreaks driving stronger distancing reactions. i suspected this too. so i tested it. i ran US state case counts to see if higher counts drove stricter policy. nope. r2 is minimal. this is noise.pic.twitter.com/NOQvwQhtw0
the only real predictor of policy in US was politics. man is that what i was NOT hoping to see. blech.pic.twitter.com/xsK874EQSr
the rest of world looks just like the US, only more so. r2's are even more inverse to SD/SIP hypothesis this distancing dog simply does not hunt. it sounds like it ought to work, but the data is stunningly clear and literally unanimous more SD correlates to more deaths per mmpic.twitter.com/obZ7favZxs
another check we can use on this data is consistency. if SD were working and some countries were engaging in more than others, we'd expect to see changes in relative rankings. but we don't. it's been extremely similar for 3 weeks. SD is just not a meaningful driver here.pic.twitter.com/SeKswaDTQL
we've seen this all over the world where curves in deep distancers like belgium peak the same time as sweden. there is just no evidence that this behavior flattened or even altered curves. so it's time for it to stop.
the price paid per unit time is an all time peacetime record for humanity. i doubt anything else even comes close. we bought nothing. we may have actually bought more harm. the "models" were wrong. it's clear to see now.
the game will be for those that did this damage to claim they saved us. "it would have been so much worse!" that's just wrong don't fall for it. these "leaders" took the global economy for a drunken joyride and wrecked it. demand accountability.pic.twitter.com/gOo7nwyJUQ
you can see more on methodology and why i did it this way here. i'd be grateful if you'd read it before asking questions it may answer.https://twitter.com/boriquagato/status/1252267723712638977?s=20 …
many have provided insight, prodding, useful questions, and good ideas while doing this analysis. my thanks go to all. any mistakes are entirely my fault and not theirs.
as this now covers enough time to be as valid as it's going to get, i'm unlikely to update these further. states are starting to open again. tracking how that goes is going to be the far more interesting issue going forward, so i expect to focus on that.pic.twitter.com/ML4JNaMVTk
How about cases per capita vs mobility? You charted deaths per million against mobility, deaths are a function of cases, so cases per million should also show the same.
cases are a difficult metric to use because i think they are mostly data artifacts and you wind up measuring the rise in testing (and the variance of that rise), not the disease curve. i'm not convinced it's good enough data to really work with if you can help it.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.
