Rt is a key measure of how fast the virus is spreading. It’s the average number of people who become infected by an infected and infectious person. Rt = 1 means one infected person infects one other person on average. If Rt is above 1.0, the virus will spread quickly or logarithmically. For example, if Rt = 2, the progression of infected persons from that one index case goes 1,2,4,8,16,32,64,128,256,…etc. When Rt is below 1.0, the total number of infected persons will start to decrease.
https://rt.live/ is a really cool site that tracks the Rt number, by State, and over time.
Here are some examples, but you really should go to the website and look for yourself.
These are the current data for every state in the Union. In green are the states with Rt below 1. In red are the states with Rt above 1:
This was 3 weeks ago:
This is current and highlights the states without stay-at-home (shelter in place) orders:
Rt numbers over time per state:
You can also superimpose these curves with the curves of new cases:
Persistent low Rt numbers as in the case of New York since early April, seems to correlate with “flattening of the curve” of new cases. The reverse also holds true in some states like Ohio where a recent uptick of the Rt number between April 12-19 corresponds with increased numbers of new cases and no flattening of the curve.
I signed up on https://rt.live/ to get updates by email and twitter. I recommend that all those making COVID-19 related decisions, whether nationally or locally, follow these data carefully. They may provide a guide for cautiously reversing stay-at-home orders in some locations.
Please remember: our health and our economy are one and the same.
R&R - Reachout and Rebuild 
Someone asked how Rt is calculated. That is a key question. Read more here.
Estimating COVID-19’s Rt in Real-Time
Kevin Systrom – April 17
In any epidemic, Rt is the measure known as the effective reproduction number. It’s the number of people who become infected per infectious person at time t. The most well-known version of this number is the basic reproduction number: Rt when t=0. However, R0 is a single measure that does not adapt with changes in behavior and restrictions.
As a pandemic evolves, increasing restrictions (or potential releasing of restrictions) changes Rt. Knowing the current Rt is essential. When Rt is >1, the pandemic will spread through a large part of the population. If Rt is <1 the pandemic will slow quickly before it has a chance to infect many people. The lower the Rt: the more manageable the situation. In general, any Rt <1 means things are under control.
The value of Rt helps us in two ways. (1) It helps us understand how effective our measures have been controlling an outbreak and (2) it gives us vital information about whether we should increase or reduce restrictions based on our competing goals of economic prosperity and human safety. Well-respected epidemiologists argue that tracking Rt is the only way to manage through this crisis.
Yet, today, we don't yet use Rt in this way. In fact, the only real-time measure I've seen has been for Hong Kong. More importantly, it is not useful to understand Rt at a national level. Instead, to manage this crisis effectively, we need a local (state, county and/or city) granularity of Rt.
What follows is a solution to this problem at the US State level. It's a modified version of a solution created by Bettencourt & Ribeiro 2008 to estimate real-time Rt using a Bayesian approach. While this paper estimates a static R value, here we introduce a process model with Gaussian noise to estimate a time-varying Rt.
If you have questions, comments, or improvements feel free to get in touch: hello@systrom.com. And if it's not entirely clear, I'm not an epidemiologist. At the same time, data is data, and statistics are statistics and this is based on work by well-known epidemiologists so you can calibrate your beliefs as you wish. In the meantime, I hope you can learn something new as I did by reading through this example. Feel free to take this work and apply it elsewhere – internationally or to counties in the United States.
Additionally, a huge thanks to Frank Dellaert who suggested the addition of the process and to Adam Lerer who implemented the changes. Not only did I learn something new, it made the model much more responsive.
If you want more math, check this out: https://github.com/k-sys/covid-19/blob/master/Realtime%20R0.ipynb