Posted by Numbers continue to trend downward, but not as fast as the model predicted. The model lost significant power over the weekend and now has an r-squared of 0.51 – basically a coin flip.
But… like any good analyst, I had a secret, backup model that I built. I’ve been running it the past 2 weeks and it has been gaining in power and finally overtook the previous model. This model is built using a 5th Order Polynomial – aka a Quintic Polynomial. These models are good when there isn’t necessarily symmetry in the data and can have multiple inflection points.
And right now, the model is suggesting a Wave 3 coming soon. Check out that surge.
Let’s be clear that I’m not certain a 3rd wave is actually coming. But it would make a lot of sense. We had Labor Day. The return to college. Offices and businesses opening up. Weather getting cooler. Conditions are actually ripe for a 3rd wave.
One mitigating factor might be the smoke. We’re in week 2 of heavy smoke cover. I don’t know about you, but we’ve been basically homebound for the past 2 weeks. We’ve done absolutely nothing (not that there’s anything to do anyway). However, if the smoke keeps people from getting out and about, it could drive infections lower. The one mitigating factor – the precautions that you take for COVID – e.g. lots of ventilation – are the exact opposite of what you do for smoke – limit external air intake. It will be interesting to see how these factors play off each other in infections.
So, what should you be on the lookout for in a “wave 3” situation:
- It’s going to be several weeks before infections would hypothetically increase. So, don’t expect a big jump in the next few days.
- We should see the recent declines start to slow down and plateau. If we see a plateau in the next week, get your mask and Lysol ready!