{"id":239,"date":"2020-06-17T13:51:44","date_gmt":"2020-06-17T20:51:44","guid":{"rendered":"http:\/\/hardythoughts.com\/?p=239"},"modified":"2020-06-17T13:51:44","modified_gmt":"2020-06-17T20:51:44","slug":"starting-to-rise-nationally","status":"publish","type":"post","link":"http:\/\/hardythoughts.com\/index.php\/2020\/06\/17\/starting-to-rise-nationally\/","title":{"rendered":"Starting to Rise Nationally"},"content":{"rendered":"\n

The central point of my post on Monday was that the states with rising numbers were accelerating in such a way that their daily increase would soon outweigh the daily decrease in non-rising states, and that as a result the national numbers as a whole would soon begin to rise.  There is some evidence in the data that this has now happened.<\/p>\n\n\n\n

The evidence comes from looking at the 7 day average of the percentage growth rates for each group and for the nation as a whole.  This is the same type of data that we looked at on Monday.  One of the nice things about this measure is that it isn\u2019t very noisy – it doesn\u2019t change direction a lot.  Daily case numbers go up and down a lot.  For example every Sunday is lower than the Thursday before it and the Thursday after it.  This is what gives the daily case graph their familiar up and down zig zag pattern.  If we\u2019re interested in whether the numbers are rising overall or falling overall, that up and down pattern is noise.  Knowing that the numbers fell from Thursday to Sunday, or rose from Sunday to Thursday, doesn\u2019t tell us anything interesting about whether cases are generally rising or falling.  The amount of change in the weekly cycle overwhelms the amount of change from week to week, making the latter harder to see.<\/p>\n\n\n\n

By looking at the 7 day average, we cancel out a lot of the noise in the weekly cycle.  It\u2019s not perfect, because the weekly cycle isn\u2019t perfect, but it makes it a lot easier to see the general trend.  This is the dotted line in graphs such as this one:<\/p>\n\n\n\n

\"\"\/<\/figure>\n\n\n\n

The dotted line shows the general trend more clearly than the solid line does.  It still isn\u2019t fully smooth, it still goes up and down a bit, but it\u2019s a lot smoother than the solid line.  Just how smooth a line you want depends on what you\u2019re looking for.  If we were interested in the correlation between the number of tests on a given day and the number of new cases found that day, then we might not want to smooth the line at all.  Because we\u2019re interested in longer term trends, we want a line that doesn\u2019t bounce back and forth much.  Ideally we want a line that only changes direction when there\u2019s a longer term shift and not just because there was one day with a particularly high or low count.  That\u2019s what we get by looking at the 7 day average of percentage growth rates.  Here\u2019s a graph with both the daily and the 7 day average percentage rates for the nation.<\/p>\n\n\n\n

\"\"\/<\/figure>\n\n\n\n

The dotted line in this graph is even smoother than the one in the previous graph.  Over the period from April 19th to June 12th there is only one day that isn\u2019t a smooth transition between the day before and the day after in the 7 day average line. Except for that one day, the line represents a consistently decreasing trend. With that one exception, every day is no higher than the day before it.   However, since June 12 the line consistently rises – every day since then has been at least as high as the day before it.  Given how smooth this measure has been up to this point, this looks like a change in the longer term trend.  It\u2019s possible that it will start trending down again tomorrow, but I don\u2019t expect it to.<\/p>\n\n\n\n

My reasoning is much the same as it was on Monday – the rising states are rising faster than the others are falling.  Some of this is just based on the underlying math. A state\u2019s daily cases can rise as high as its population, at least in theory, but can\u2019t fall below zero.  If one group of states is rising and another group is falling over a long enough period, then the rising states will rise more than the falling states can fall.  However, another part of my reasoning is based on the way the virus spreads.  Viruses want to grow exponentially, to snowball, until they saturate the population.  This virus is a long, long way from saturating the US.  Keeping it from snowballing requires exerting constant pressure on its growth rate by social distancing, wearing masks, and similar activities.  As we do less of that, we expect more growth.  Over enough time, growth will overcome any gains we have already made.  To prevent that, we have to actively do something to decrease the spread – social distancing, contact tracing, vaccinations, \u2026 something.  Even the most optimistic predictions for a vaccine don\u2019t have it available for several months from now and there doesn\u2019t seem to be much appetite for social restrictions, so my best guess is that we\u2019ll see continued growth until the numbers get too high to ignore.  Once we decide to take action, it will take at least another month to bring the numbers down.  So I expect this next phase to last at least through the end of August and likely well into the fall.<\/p>\n\n\n\n

There is some good news, though.\u00a0 It looks like there is a good chance that they have found an effective treatment for some of the worst outcomes of COVID, those that require the use of oxygen or a ventilator.\u00a0 The drug is Dexamethasone, and a steroidal anti-inflammatory.\u00a0 To be clear, this isn\u2019t a treatment for the virus itself, it\u2019s a treatment for the specific types of respiratory problems that the virus can cause.\u00a0 It is of no use in preventing these problems and might even be harmful if used before they occur.\u00a0 However, it\u2019s widely available and not expensive, so it could go some way toward reducing the fatality rate.\u00a0 You may have seen headlines about this, but if you want a little deeper look, here is Derek Lowe\u2019s take on it<\/a>.\u00a0 <\/p>\n\n\n\n

Finally, here is a projection for the next 10 days.  I decided not to make any adjustments based on anticipated growth, so my expectation is that this will end up being a bit low over 10 days.  However, I don\u2019t expect the differences to be large over that time frame.<\/p>\n\n\n\n

Day<\/td>Cases<\/td>Deaths<\/td><\/tr>
6\/16<\/td>2,137,579<\/td>116,960<\/td><\/tr>
6\/17<\/td>2,159,527<\/td>117,614<\/td><\/tr>
6\/18<\/td>2,181,478<\/td>118,269<\/td><\/tr>
6\/19<\/td>2,203,430<\/td>118,924<\/td><\/tr>
6\/20<\/td>2,225,382<\/td>119,579<\/td><\/tr>
6\/21<\/td>2,247,333<\/td>120,234<\/td><\/tr>
6\/22<\/td>2,269,285<\/td>120,889<\/td><\/tr>
6\/23<\/td>2,291,237<\/td>121,544<\/td><\/tr>
6\/24<\/td>2,313,188<\/td>122,199<\/td><\/tr>
6\/25<\/td>2,335,140<\/td>122,854<\/td><\/tr>
6\/26<\/td>2,357,092<\/td>123,509<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n

Be safe, and do whatever you can to help slow the spread in your own areas.<\/p>\n","protected":false},"excerpt":{"rendered":"

The central point of my post on Monday was that the states with rising numbers were accelerating in such a way that their daily increase would soon outweigh the daily decrease in non-rising states, and that as a result the national numbers as a whole would soon begin to rise.  There is some evidence in […]<\/p>\n","protected":false},"author":2,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":[],"categories":[5],"tags":[10,8,7,9],"_links":{"self":[{"href":"http:\/\/hardythoughts.com\/index.php\/wp-json\/wp\/v2\/posts\/239"}],"collection":[{"href":"http:\/\/hardythoughts.com\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"http:\/\/hardythoughts.com\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"http:\/\/hardythoughts.com\/index.php\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"http:\/\/hardythoughts.com\/index.php\/wp-json\/wp\/v2\/comments?post=239"}],"version-history":[{"count":1,"href":"http:\/\/hardythoughts.com\/index.php\/wp-json\/wp\/v2\/posts\/239\/revisions"}],"predecessor-version":[{"id":240,"href":"http:\/\/hardythoughts.com\/index.php\/wp-json\/wp\/v2\/posts\/239\/revisions\/240"}],"wp:attachment":[{"href":"http:\/\/hardythoughts.com\/index.php\/wp-json\/wp\/v2\/media?parent=239"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"http:\/\/hardythoughts.com\/index.php\/wp-json\/wp\/v2\/categories?post=239"},{"taxonomy":"post_tag","embeddable":true,"href":"http:\/\/hardythoughts.com\/index.php\/wp-json\/wp\/v2\/tags?post=239"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}