Yes, it's best to focus on the 7-day rolling average.
Which of course, is horrible.
So how are people getting the virus? According to OKC- County Health Dept, contact tracing during July, the 4 hot spots for the virus have been faith based venues, restaurants, office settings and daycare. It indicated that bars gyms, and warehouses are still among areas of spread.
Only the restaurant part concerns me. Not often, but sometimes I go to a restaurant for dinner but always go at a non-busy time such as no later than 5 pm. After being seated the waiter asks me to have my mask on when I leave the table, due to city ordinance. So on one occasion someone going to the restroom would walk past me without a mask on.
I think we're all suspense in coming months to see how well reopening of schools, universities and sports play will go. Whatever the outcome, it will surely be useful in gathering a better understanding of the virus.
We never bent the curve. we only kind of flattened it sort of a little. Then we did stuff and didn’t do stuff that caused our numbers to shoot up for two straight months with no government response besides sundry mask ordinances.
There won’t be a bent curve until something is done to make it bend. Do you see Governor Stitt doing anything?
A flattened curve at an unsustainable number isn’t good either but we haven’t peaked.
It isn’t always a straight line, and all data points increase every single week. Are they completely out of effing control or just totally out of control? Who cares? The situation is out of control.
Are you happy with these numbers? What do you think is going to happen when school starts?
I never said we peaked. I said that we’re probably getting close to peaking. We’re starting to plateau. Clearly we have bent the curve, but whatever you want to yourself. Go run a derivative function on the 7 day average and tell me what you see? It will show that your rate of change has slowed. Whether that’s real or not is impossible to know given our issues with reporting numbers lately. I will say that is likely real, as our curve is behaving very similarly to nearby states.
Ok, I guess we have a different view of the word bent. I see what you are saying and I am impressed by your mathematical and science knowledge but I take little comfort from what you are saying. The situation in Oklahoma is spiraling out of control. New cases and hospitalizations rise weekly. And now, deaths are rising. Our health department continues to report bogus recovery statistics that the CDC does not recognize, and there are ill-timed gaps in reporting. In some cities it takes several days to schedule a test and several more days for results. And, our governor is in 24/7 Trump re-election mode and is completely abdicating his primary job, which is to protect the citizens of this state
I do not need to survey a complicated math equation to say any of the above.
Certainly hope you're right, as the last time it looked like rate of change slowed (A), we saw a huge leg up until (B) happened.
Sheesh, I need a refresher on derivatives. Have you already calculated it? I believe I can add that to my spreadsheet and plot it over time to see how deaths & cases change.
![]()
These large endpoints, like today’s numbers, can really negatively affect trend analysis. Any derivative you calculate including today’s numbers will have a very large affect. I know the common way to analyze this data has been to apply a 7 day running average which smoothes the curve; however, our tests are delayed up to 10 days in some cases. I wonder if a 10 day running average would be better.
Even better than running averages, I wonder what the data would look like if you binned it instead. My suggestion is to pick a day. Any day. Doesn’t matter. I propose Friday’s as the numbers decrease over the weekend and are captured in Tuesday’s. Take Friday, and sum all the numbers for the previous week. So you’ll bin the numbers by week. Week 1, week 2, etc. Plot that curve. I bet it will look nicer than the 7 day running average.
Is it your goal to make it “look nicer?” If so, why? Whose interest does that serve?
We already have elected officials asleep at the wheel and a huge volume of people wandering around, claiming this is a hoax. Do we really need to get all academic with graphs when it should be enough to state that our cases, hospitalizations, and now deaths, keep rising beyond a sustainable level?
Some people don’t understand that factors like demand and testing availability and data entry issues and many other things may affect the daily reports. So some people may look at a “messy” graph that sometimes jumps all over the place and decide “look at that mess, how can they claim anything that is that variable could be accurate” and dismiss it out of hand. By making the graph less volatile, it may make it look ‘more accurate’ to folks like that.
At least I interpreted the comment to mean “make the graph look better to look at” rather than “make the numbers appear better”.
You can affect trend lines with algorithms based on sensitivity of changes. People need to quit trying to create spin without complete knowledge of how the reporting is done. Anybody can clearly see we are in a state of emergency exacerbated by the head in the sand positions of a certain segment of our society and "leadership". Reality is a hard master. It doesn't care what you believe it only cares what is. Time to believe reality and not myths.
Did we really have a July 4th peak?
No absolutely not. This data is messy. There are lag times, testing errors, and a host of other problems. Plotting these data as we have been can lead to misinterpretations of the data. This is common issue in data analysis for noisy data. Smoothing doesn’t necessarily help with this. It kinda helps, but you can still draw bad conclusions from these graphs.
What I’m asking is whether there are other ways to view the data that may help us better understand what’s really happening out there. I’m not trying to fit a narrative.
And yes, it’s absolutely necessary to get “academic” with graphs and stuff. That’s what we should be doing with this data. If people can’t understand the basics of trend analysis, scientific methods, etc, then that’s on the education system....which is probably part of the reason why we’re in this mess to begin with.
Again, you are ignoring the fact that your basic thesis does not change what is actually happening. We don’t need to look at a graph to see that. Regardless, two separate graphs were supplied that seem to undermine your claim.
Ugh...again, you fail to see where I’m coming from. You’re taking the referenced graphs at face value. You’re taking them to be accurate which they’re not. It’s impossible to know what’s really going on with time-based data when the data is entered at incorrect times and/or values.
What I’m proposing is not a thesis. It’s a hypothesis. I have no clue what my suggested graph will show. It may look identical to what’s been published today. It might, however, be a smoother representation of long-term trends, and we may be able to better interpret what’s going on. This could help spur better conversation on these boards.
These numbers are important but don't forget that this is epidemiology and not statistics. You can't crunch the numbers and apply graph smoothing or directional analysis unless you understand what the numbers mean. Instead, just go back to the OSDH graphs which update cases based on the date of onset. That removes most of the errors in the data:
https://looker-dashboards.ok.gov/embed/dashboards/70
Delete
Here are the weekly totals (defined as Monday-Sunday) since the start of this thing, up through and including today. Data is from CovidTracking, which seemed to be the most reliable data I could find (not sure how they're handling the weird multi-day report compared to other sites, though).
Derivatives on data like this will be quite sensitive to how you define the width of the bin and I didn't want to bother with the assumptions involved with fitting polynomials to the data, but summing the data to be weekly should help negate the day-to-day biases. In this case, here is the change in weekly totals:
So, there's some movement, but nothing I would call an obvious or substantial slowing down, yet.
Finally, here is our weekly change in cases normalized by the total number of cases (which is less meaningful than it appears given "recoveries", but I digress). Other than the large numbers early on due to the very low number of existing cases, our rate the last month or so has been remarkably steady at about 0.19 (that is, we're adding about 20% of our total cases week-over-week).
Sorry for the poor graph quality. Blame Google Sheets, lol.
There are currently 182 users browsing this thread. (0 members and 182 guests)
Bookmarks