No worries, JP. The GOP has no designs on banning cherry-picking.
I'll run the maths for you here.
Let me know where you cannot keep up.
Total deaths:
NY:68714
Iowa: 9510
Raw deaths per 1M-
NY: 3532
Iowa: 3014
NY looks LOTS worse than Iowa, right?
But look at the March-June data.
If you review the running Case Fatality Rates for the first months of the pandemic, they were well north of 5%.
March: 8.2%
April: 7%
May: 5.3%
June: 2.84%
(those are the CFRs for JUST the month period, using a 7-day crude lag for CFR, which means using April 1-30 for deaths and March 24-April 23 for leading cases for that month)
Every month since June 30, 2020, CFR monthly numbers dropped to mostly between 1% and 2%. Why? Because doctors figured out how to treat patients, and dexamethasone became Standard Of Care treatment for severe cases, which was not known or available during the first 3 months.
So, how do we correct for that, since NY was hit first and hit hard? Adjust their deaths from March-June using that 1-1.5% CFR that became the norm for when others got hit. Early CFRs were easily 4x to 5x worse for NY and others "first in line".
So, how many deaths did NY and Iowa have from March to June 30?
NY: 32152
Iowa: 715
Let's use the 4x ratio, so if they HAD dexamethasone for those first patients, their CFRs would have been 1.5% or so, meaning 3/4ths of those patients would have lived.
Adjusted deaths for Iowa and NY during that period:
(these are 1/4th of actual, based on the CFRs being inflated about 4x-5x; I used 4x)
NY: 8038
Iowa: 179
So, now you can take their total deaths counts, subtract the March-June 30th numbers off, and add back on that 25% estimator that uses a 1.5% CFR.
New adjusted total deaths numbers
NY: 44600
Iowa: 8973
Now, your deaths per 1M are
NY: 2292
Iowa: 2844
When we correct for early CFRs, particularly for states hit very hard early on, we can see that they actually fared a lot better than the crude data show you. This would be the "least biased" comparison you can make, because CFRs are nominally similar across all the time-frames.
The DanGoodspeed charts I've linked before ignored those early 3 months, which is really not a bad metric because CFRs AFTER that date are nominally consistent; they do, however, fail to make the adjustments above for the first 3 months, which is how I would recreate those plots, for every state.
You are Welcome.
P.S. You can take an even more granular approach, by separating out each monthly CFR and correcting those 4 months from March-June with the close CFR numbers, and you'll get an even better "least biased" correction. Feel free to do so. I used numbers off WorldOMeters for this.