COVID19 "Experts" and Fake Curve-Flattening

Adam Larson (aka Caustic Logic)
April 16-17, 2020
rough, incomplete
adds April 20, 21, 23

Note 4-21: Apologies for this post's rambling sloppiness - I think I should leave this as-is and perhaps re-write the points more clearly somewhere else. It won't be at 21st Century Wire, that's for sure. Ideally, maybe someone with more reach than I who gets these points - especially Mather's graphing trick/error - could raise them anew to help fight this contagious idiocy.
---
So I've been making some enemies jumping into the COVID19 subject and controversies. A depressing number of otherwise intelligent, skeptical people now suffer from an excess of skepticism, and it's a moderate public health danger. I've argued the points so many times already, here short form: question are fine, but ask honestly. The wisdom of the current lockdown reactions of nearly all world governments are validly debatable, but two important points:

1 it's not a global conspiracy where millions of nurses, doctors, patients and fatalities are faked or exaggerated by compiling other deaths, etc. From Communist China to Theocratic Iran, hyper-capitalist US and UK and virtually every government on Earth - everyone who would agree too much death at once and a crashed medical system is no good, but neither is a wrecked economy - all consulted all their best experts on health, science, and the economy and decided on what they did. Maybe all of them are wrong, but …

2 … this suggests the danger is real, and the measures these armchair experts call terrible mistakes or an evil plot have a reasonable basis anyway. The rules should be followed absent some rather good reasons, as long as they're reasonable and not abusive (and you might not be able to easily tell which ones are and aren't).

Critics tend to focus on the actions of the home government they distrust, and taking that in isolation, find it easy to see it as the start of a horrible dystopic future, for which they had their elite "experts" fake a simple flu up into a compelling panic. They try not to lose a step over every other regime with all their varied interests doing the same thing at the same time - the Iranian scientists and the Russians etc. are also controlled, somehow. And all their doctors and fake patients involved in staging the "crisis" there … all their news media, etc. Everyone's in on it - they guess.

These independent minds aren't swayed by "science-ism" and official "experts" - but to back up their hunches, they cite other "experts" - some 20 of them or more known to oppose the status quo, who explain things like how COVID19 doesn't seem to kill anyone - people just drop dead from other things after allegedly getting it. That's reportedly been happening with thousands a day worldwide, close to 1,000/day just in the UK, but maybe the numbers are faked, or whatever.

And they say it's not that many; spotting early on how the low numbers (existing then) looked like normal tallies for a whole flu season, some predicted that would be the end of it, and it'd be like a normal flu season - which I hear kills thousands a day frequently, so hospital SHOULDN'T be extra-swamped. They don't seem to believe the reports that hospitals have been and are overwhelmed in many areas, and the fear is how much worse it could get seems ill-founded, to these skeptical minds. And of course not that many would die (not) from the virus - most are old, ill, were about to die anyway in the following weeks to decades. And they warn therefore POLICE STATE RESIST, etc.

Anyway, I've found this annoying and worth combatting, so I'm a "gate-keeper" and all that. Fine. Here I'll share a look at two expert (mathematician) views involving predictions that popped out for being cited at me recently. I close the gate on them (as if they'll actually be stopped by MY gate), and here's why.

First, on sources: I'm using what seems to be the standard running tallies provided by the WHO that everyone else uses, as tracked by Bing - https://www.bing.com/covid This has been called Bill Gates' fear machine and not trustworthy, but it seems to use the same numbers everyone else does, and it's handy. There is an odd issue I've noted where most graphs (but not all) are missing the date April 1, showing instead a sharp spike between May 31 and April 2. This error is variously ignored, crudely repaired, or not present in the graphs I use here. It doesn't seem to affect anything past that spot on each graph, but it raises some question... (the example here is from Brazil's graph) And of course there numbers aren't complete or gospel, include false positives, exclude false negatives and, unconfirmed cases, etc.  To some minds they also include lies or exaggerations, and I admit I can't rule that out. So, grains of salt and all that, we proceed.

Mather/Peerless Reads: Fake Not Facts
On April 5 I was informed by an exasperated Patrick Henningsen (21st Century Wire): "growth stage is over, growth rate hump and curling over now, so 'crisis' is over." Therefore, no justification for the ongoing repression of the people of the UK, contrary to what I kept arguing. He cited this video: https://www.youtube.com/watch?v=VBUHmwHN96U
"Coronavirus Fact Not Fear - Chart Update 1st April 2020" by "Peerless Reads" - which I see is run by "Andrew Mather, a 60 year old British mathematician, financier, author, techie."

I don't think I understand "curling," but mostly followed what Mather says. The UK government's prediction of 500,000+ death there, shifted to 5,7000 if people comply with the Orwellian lockdown - that was all baldfaced scare tactics, he concludes after a calm look at the facts. Comparing reported trends between "Hubei (finished), Italy, Germany, Norway, and UK charts without lockdown," he finds all of them will level off, or already have, at a certain rational point where the virus wears out, gets bored, stops being contagious, or manages to have all active copies die or fall completely through the cracks between people that would naturally emerge.

Mather's projected final deaths in UK:
* 4,508 if it spread and killed naturally as it did in Hubei, China
* 6,064 if it killed as it had so far in Italy
* 10,269 if it killed as projected by some in Italy.
And after killing that many, the virus would get bored or whatever and go back to being mundane. That's the natural progression, Mather somehow decided. Therefore, "we were always headed for 3,500 - 10,000 (deaths), nothing like the 530,000 we were told to be scared of."

Well it's been two weeks. We can't check what WOULD have happened, so far, with no lockdown, except to say it'd be worse than what we see. First, a scaled-down example in Washington State (where I live) with my own inexpert guess at a reasonable spread without gov. Inslee's orders. Whatever the scale, that's the idea - distancing and lockdown seem to be helpful at keeping the levels lower.

Current UK deaths: 12,868 13,729 as of the 16th, and rising some 600-800/day (most days - it's ranged from ~400 up to nearly 1,400 in a day). Apparently, the UK never was headed for 3,500 to 10,000 deaths in total (nor to 5,700). Luckily, preventive measures have kept it from exceeding those predictions by a ridiculous margin - yet.

Here's the graph shown - dotted lines start at prediction time, April 1, and note that's where they start leveling - in his imagination. The curves before should be based on real numbers, presumably graphed right, but … they seem to have dramatically varying shapes compared the ones I'm seeing - they turn hard left, then hard right in a way that makes me wonder. We'll come back to that.

As for the comparisons, which of these nations did and did not use lockdown is worth more review. China - Hubei included - enacted strict lockdown and other containment measures. People were chased inside by drones, dragged off to quarantine centers in vans, etc. This is NOT the natural progression to compare to; this is actually more like the model of a lockdown response we'll only want to emulate in part.

This is quite similar to what Mather takes as the natural progression. But he might have noticed the boot imprint on the Chinese lines (gold, gray, gray in his graph) as something he shouldn't presume with the others - those would show the actual natural flattening he imagines, not the Communists' barely-truncated version, with that barely-needed boot. Also note there's been a recent spike of cases not shown here, linked to Chinese workers returning from Russia - another thing Mather doesn't seem to predict as natural. I mean, does the virus get bored of being bored after a while? (and add 4-17: China just reported they had under-reported deaths in Hubei hugely - add some 1,290 further deaths, a 38% increase of what's shown here, and what Mather used as a baseline.)

The numbers of cases Mather predicted wind up not too far off from what we see, but these follow on lockdown measures, and might well reflect them, even enormously. His assertion the same would have happened naturally is what hasn't been proven. And as I'll show, he underestimated its lethality, so a similar number of cases in the UK led to a higher death rate already in excess of his highest final prediction (in fact nearly 50% over his high end), and still it's still rising quickly.

But even if the numbers aren't so wrong, it's the visual graph most people will walk away with. That has colors and shapes and tells a story you can see. And it's confusing or even deceptive, and also makes it hard to make out what the predicted numbers are. Do notice those rises can't be read the normal way, but more like a Richter scale in reverse. I don't know if he did this to be tricky, but each rise is exponential, at least according to the numbers on the left edge: the top bar represents a spread of 900,000 (100,000 to 1 million), while the one below spans just 9,000, the one below that 900, then 90, 9. The bottom line shows a rise from 1 case to 10 cases. The top line looking all flat there should actually rise 100,000 TIMES as high as it does to be proportional to the bottom line. Obviously that calls for a different scale that does NOT create a false appearance of sudden onset followed by inevitable and swift flattening. This is what we want to happen for real, not in an optical illusion on the computer screen.

I tried a rough re-proportioning just from the screengrab. It's not easily do-able, but just for a basic idea what it should look like, see below - the top line is way taller than the one below and so on. But the difference evolves over the box, so it just keeps getting flatter as you go up and hence - the virus is getting bored (hence, my proportional numbers on the right aren't right). From there down I squeezed lines down, but fudged that too, to keep things even visible.

Adding: I watched the rest of the video - it compares other nations on the same exponential scale, showing the same "flattening" once the numbers get high enough to get more compressed. "Curlover" is the "curvature" or "flattening" that he says "says the growth is slowing down." But the flattening we SEE mainly means the growth is being shown more squished. I tried again to reproportion the thing better, with each line at least partly reflecting its own vertical distortion. I just skipped the bottom two rows, and again fudged the two above that. Far from exact but closer, I decided all lines stop about 1/4 of the way up from 100k to 1m, so below the marked 325,000 line. By this, his projections come out something like:
- Italy: 295k
- UK 220k
- Norway 170k

Again, the numbers there are per 100 million (so divide by 100 for the more common per million). By Mather's prediction, vs. as of today (April 16) per https://www.google.com/covid19-map/:
* UK (pink/magenta) should be at about 1,700/million and still rising, set to level around 2,2000/m over the next two weeks. vs. current cases 1,552/million - not that much off, but rising faster than predicted, even with current measures, while this was a prediction for without them.
* Italy (green): fully leveled at about 2,950/million predicted vs. 2,741.46/million current - it progressed slower than predicted, but still is not leveled off. Is something slowing it down? But yet it's not leveling off completely, on its own or with the help? What gives?
* Norway (blue): fully leveled at ~1,700 vs. 1,276 and still rising, just slower than predicted. Same questions as with Italy.
* Germany (red): no projection shown vs. 1,630.19

So he's not so far off from current numbers, but denies these are low due to lockdown measures - AND he decided they'd be final numbers in a closing story, rather than early points in a crisis that's still unfolding dramatically. Here's how this illusion-based claim (green boxes) - and the next "expert" claim (blue) - compares to actual running trends in the UK:

(add 4-17: record deaths in the UK just now: +1,708 deaths in a 24-hour span (now 14,576), +10,216 news cases. Also US deaths hit a record: +6,801 in 24 hours.)
(add 4-20: now past 120,000 confirmed cases and 16,000 fatalities - adding a lot of notes and Forsyth prediction)

The 40 and 70-day markers there take us to the other expert view I looked into...

An Israeli 40-Day-40-Night Theory
https://www.timesofisrael.com/top-israeli-prof-claims-simple-stats-show-virus-plays-itself-out-after-70-days/
Times of Israel reported on 14 April the findings of Isaac Ben-Israel, who is: "a prominent Israeli mathematician, analyst and former general," "head of the Security Studies program in Tel Aviv University and the chairman of the National Council for Research and Development" and "who also heads Israel’s Space Agency." Wow. Another mathematician, but one presumably dedicated to the safety and well-being if the nation he serves, and highly competent. And here he slams economic closures as disproportionate responses to a virus that apparently gets bored and stops bothering people in a certain country after a few weeks. Maybe he's hoping the idea catches on in Iran.

He "claims simple statistical analysis demonstrates that the spread of COVID-19 peaks after about 40 days" (and thus, 40 nights) and then "declines to almost zero after 70 days — no matter where it strikes, and no matter what measures governments impose to try to thwart it."

As the article notes, Ben-Israel "is not a medical expert," and cannot explain why his findings should be true; "Asked to explain the phenomenon, Ben-Israel, who also heads Israel’s Space Agency, later said: “I have no explanation. There are all kinds of speculations. Maybe it’s related to climate ("no matter where it strikes"?), or the virus has a life-span of its own (all active copies die on some cue?)”  A "leading doctor dismisses his claims" - and that's Gabi Barbash, a hospital director and the former Health Ministry director general, and that's besides all those more relevant experts consulted by the Israeli government, the Iranian and Chinese ones, virtually all governments that exist, on balance firmly disagreeing with his premise. I'm not buying it either.

"... research he conducted with a fellow professor, analyzing the growth and decline of new cases in countries around the world, showed repeatedly that “there’s a set pattern” and “the numbers speak for themselves.” He "claims simple statistical analysis demonstrates that the spread of COVID-19 peaks after about 40 days" (and thus, 40 nights) and then "declines to almost zero after 70 days — no matter where it strikes, and no matter what measures governments impose to try to thwart it."

He was interviewed about this on Israel’s Channel 12 (Hebrew), where Prof. Gabi Barbash, a hospital director and the former Health Ministry director general, strongly disagreed, arguing that the death toll would have been far higher if world governments had not taken the preventive measures Ben-Israel was griping about. That mammoth blind spot seems to have also skewed Mr. Mather's outlook.

The article cites "countries, such as Singapore, Taiwan, and Sweden, which did not take such radical measures to shutter their economies" as those where the mathematician saw the pattern clearest. That lets us narrow-down some double-checking.

Taiwan: The first case was announced on 21 January 2020. March 1 is 40 days in – Bing graph has a sudden start only on 3-17, at 77 cases, 1 death. By 3-31 it's at day 70 and should level off to zero, but it's risen to 322 cases, 5 fatalities. Maybe it slows down a bit late? We can check 15 days later – 4-15 it was 395 cases, 6 deaths, and rising but only very slowly - only 7 cases added in a day.

Taiwan and the PRC - the two version of China - are perhaps the best examples for a curve that flattens, whatever the cause. They both show roughly what he describes. But neither is a typical case, even among those few he cited.

Singapore offers another example similar to Taiwan - an island near China with some heads-up and aggressive early controls that kept the number of cases to manage quite small. Day 1 is Jan. 25, with 3 cases confirmed at once. 40 days in is 3-5, 305 cases and no deaths yet. It should be peaking and then leveling to zero by about day 70 days - that's 4-4: 1,189 cases, 6 deaths. If it's close to zero new cases/day by then, there should not be some 2,500 added in 11 days by 4-15, as there was.

Sweden: with its notably libertarian stance … 1st case 02-02. 40 days in is 3-13, when they had 620 cases, no deaths yet. It should be leveling, and another 30 days in it should be nearing zero new cases/day. 70 days is 4-12: 10,483 cases, 899 fatalities (up 332 from the day before - a bit slower than surrounding days). And it's still rising sharply at 70 days. 3 more days on, it's hit 11,927 cases, 1,203 deaths and still rising - now with day 74 still not fully tallied, there's 12,540 cases and growing fast, 1,333 fatalities and set to keep rising for a while.

Update 4-21 on tallies to April 20:
* Taiwan, day 87: 425 cases, 6 deaths: still mellow, without lockdown. They had good luck, but were prepared for bad luck too:  "Taiwan hasn’t needed lockdowns to fight the coronavirus, but it simulated one anyway" https://finance.yahoo.com/news/taiwan-hasn-t-needed-lockdowns-092940133.html
* Singapore, day 86: 8,014 cases, 11 deaths - "Singapore extends lockdown after sharp rise," etc.
* Sweden, day 78: 14,777 1,580 - "anger in Sweden as elderly pay the price for coronavirus strategy" - still the price is not as high as it could be - they have many controls in place, Swedes police themselves fairly well as to hygiene, distancing, etc., and they may have good luck there, like the deadlier strain(s) not arriving (yet).

Not mentioned by either of these experts: the record on other non-lockdown nations: Brazil, Turkey, Belarus. Have a look at these things I've been working on (40 + 70 day marks not added):

Brazil (4-17 update) is the only county considered here that's in the southern hemisphere, so NOT even in its flu season. It spreads badly anyway, but to let it get this bad in April is probably an epic failure by the "Trump of the Tropics."

Turkey (4-17 update): see here for a post I did on the Greek border issue (blue boxes above)

Belarus (4-20 update) was faced with the same temptation as every state: keep things normal and hope this passes with minimal damage. They also have a low population, less global travel, less early cases, decent management of the few they cases they knew of, and hardly any problem until recently. Even now at 42 deaths total, it might seem laughable to think it'll ever get out of hand. But then see where all these other countries got to once the numbers start rising that fast, and recall this is another nation of millions. They will have at least 3x this many cases, maybe 10x. The deaths will follow, and they'll go deep onto the hundreds, likely low thousands, depending how good their medical system is (not sure, but might be good - former Soviet republic, etc.)
Add 4-20: they may be not reporting new cases? Something wrong with the tallies not changing ...

Above, the UK chart shows 40 and 70 day markers that absolutely do not correspond to leveling of rates, even if we shift day 1 to some later point. New York state was mentioned; Prof. Barbash cited it as showing Ben-Israel wrong, but the latter  "noted the latest indications from New York were precisely in line with his statistics that indicate daily new cases figures peaking and starting to fall after about 40 days." Checking that: 1st case 3-01. 40 days to 3-31 = a rise of 180,457 cases, and 8,627 deaths. And then it's slowing and starting to level? A bit early to say for sure, but ... no. 3 days later (April 14) Ben-Israel says he's seen a decline, but it's not there, and 4 days later the state has 213,779 cases, 14,030 dead, and still climbing petty sharply. The last 24 hours alone saw +18,847 in active cases, +1,473 deaths (a 24-hours span jut hours ago had included 3,196 deaths). So day 70 is still  up in the air. With any luck and enough people taking it seriously, it may be close to zero growth by then. God knows they'll need it by then, but they probably won't get it - it'll probably just be hitting that "40-day" peak, and the "70-day" wrap-up will take another 100-150 days. And hopefully by then there will be a vaccine and better treatments, making the remainder of the lockdown program unnecessary (or less useful, anyway).

Back to Mather - and wait, why is a mathematician name "Mather" - pure luck? He had UK infections set to level around 2,630/million, Italy at 4,000/m, Norway at 1,400. This was his varying natural point where the virus would get bored. Well it's not done yet in New York and it's at (New York State: pop ~20 million vs. 222,284 cases = ) 11,114/million, nearly three times the highest rate he predicted, for Italy. And looking at the curve, you can see that's not even close to the final answer either. And this is WITH CURRENT MEASURES, weeks of lockdown adding up so it's not 2, 3, 5 times worse than this - as it would likely be by now if we sat around waiting for the virus to get bored, or sat around waiting for the predictions of these "experts" to come true. (And an upside to the quick growth of infections is these guys get proved wrong quickly - they could be confusing people for longer, and causing more damage than they are.) So to summarize … shit's serious, and those guys are wrong, expert or not. Take that into due consideration as you weigh policies against the risks. Questions do need asked in this time of unprecedented upheaval, but they need to be asked rationally and honestly.

Follow-up: comment left at Mathers' video, in case he doesn't allow it: Could have checked the typos, but EH.

If I might ask, why do your curves play out on a chart where each lines jump ten-fold in value from the line below? Se see a rise from 1 to 10 mapped the same vertically as a rise from 100,000 to 1,000,000. It's not clarified that this APPARENT flattening the higher it gets is separate from the real flattening you suggest is shown. I aske because they seem to be the same thing to me (the illusion=your reality, and since higher levels appear flatter and flatter, the illusion is somewhat OPPOSITE of reality) and because your predictions are being proven woefully wrong 2 weeks in. 14,576 UK deaths SO FAR and rising fast (+1,708/last 24h), and that's WITH lockdown, while you predicted 3,500-10,000 TOTAL, WITHOUT it. Maybe I'm just confused, so you can allow the question to stay public and then provide the answer. Sub-question: does each line have its own steady value, or as it seems, does the curve play out evenly across the chart?

I expect it'll take him forever to admit that he saw the comment - or he'll delete it, possibly. I also did another graphic in more detail: here I get his magenta line and the actual plots to match at the stat and at April 1, and extend the red over (and up!) from there, at the same ratio. You can see how the actual numbers wind up looking different in the first part, and after 4-1 you can see the same distortion increasing across his projected rise. Also (it can be confusing) I split each row into its own internal scale, divided in 10 chunks of 9, 90, 900, etc. and mapped actual by-day numbers, connected into that orange line. You can see how each level show a natural section of the curve, but curving differently after bending at each sudden ten-fold decease in verticality. And the gap between red and pink should mainly reflect the difference between the numbers used - actual cases (red) vs. cases per 100 million (magenta).
 

There are tiny numbers in there to show what it's passing and we can see how deceptive this is. If you take visual rise as the indicator of trouble, you'd see a drastic climb from 2 to 8 (cases per 100 million), or from 10 to 100, and then see a slightly mellower climb from 100 to 1,000 then 1,000 to 10,000 - each over a similar time to the drastic climbs - then an even mellower slide through a TENFOLD increase over a longer time, 10,000 to 100,000. After 100,000 it's barely worth charting it's so uselessly horizontal. Passing from 100,000 towards 1,000,000, that'll take quite a bit longer and look super-flat, despite being another tenfold increase in nothing like tenfold the time. You could roll a baby carriage up that slope with no effort, lay down on it and take a nap without rolling downhill at all. And you can wonder why all these hospitals are griping over a piddling 900,000 further cases.

Speaking to his viewers without waiting in another comment in this later video:
https://www.youtube.com/watch?v=Tugj3jmv5n0
So Mr. Mather created an optical illusion with his charts; they use an exponential scale to artificially flatten numbers more the higher they go. See the labels at left and note a dramatic climb from 1-10 (per 100m) occupies the same vertical space and similar time as a leisurely hike from 10,000 to 100,000. Passing 100k now, the climb to 1 million will be virtually flat, good job.
But that was not leisurely in the real world, because 9 is not the same as 90,000 or 900,000. He acts as if this illusion reflects something real, but it's not lining up. He predicted 3, 6, maybe 10,000 deaths total, even without lockdown. But even with it, it's near 16,000 UK deaths already and still rising about 800+/day, because the infection rate is NOT "curling over" except in his special magic system. But we don't live there. Alright? It's a hoax. Quit being dumb, people. http://libyancivilwar.blogspot.com/2020/04/covid19-experts-and-fake-curve.html

https://twitter.com/MichaKobs/status/1251340173608194049
Kobs: see logarithm plotting for Italy - that's the scale and the effect of it - I might have known that's what you call this trick. Linear: that's what people expect to see, apparently think they ARE seeing.
The later video has a PDF form of all the slides (I guess his others do too)
https://peerlessreads.s3.us-east-2.amazonaws.com/CV19FNF_UKIndepth_2004010d.pdf
Methodology explained there:
slide 10: "Here’s what I had to work with. Raw data from the WHO DSR painstakingly typed in. And for this, I typed in every day for a few key countries (Italy, Germany, Norway, UK, US and Canada) rather than data every three days as suffices normally for chart updates."
11: "Noting each country’s population in millions, I then scale up or down to cases (total) per 100m population" (still not clear why he converts the numbers to /100m - just to make it harder to follow? FWIW he uses 66m for UK, lower than the 67.8m figure I use for /m calcs)
12: "Subtracting the previous day’s total from the next day’s figure, I get the cases declared on that day."
(Then something about plotting the numbers, choosing the scale, what he intends it to show - no details given here) and finally:
13: "The end result is some rather jagged charts as you see below" (all sharing a tendency to easily plateau, especially the higher they get)
Skipping ahead, he frets over the UK curve still creeping up and down despite his distortions, mumbling about an end point he didn't like, and concludes in slide 21 "I ignored the end point for good reason: it’s an outlier, a data point that can’t be relied upon and which disrupts a perfectly sensible picture." Yeah, tomorrow's will do the same, etc.

Add 4-20:
Another try at visually explaining the lie:

UK Column, April 3: Mather hoaxes Mike Robinson and Patrick Henningsen
https://www.youtube.com/watch?v=5KmyHP5szGY

Letting them know:
https://twitter.com/CL4Syr/status/1251488775466659840
https://twitter.com/CL4Syr/status/1252212050102587392

https://www.youtube.com/watch?v=EEjgPLsSJgc
Another Mather video explains curves for dummies as if it's all you need to read his graphs - all curves shown ae labeled and graphed in "linear" mode, and he emphasizes how utterly simple this all is - nothing tricky to it at all! Just graph the data - a curve is a curve - and then his important graph is algorithmic, NOT linear, something he never did explain, leaving you to see it as a linear-literal curve reflecting real-world changes.

I'm convinced this mathematician (and Mensa member, I hear - a genius) did this all on purpose, not accidentally. He'll know an algorithmic generated curve when he sees it, having just (as he says) manually entered the much higher numbers. I left another comment at that video (in reply to one of his) calling him a sick person who should stop his joke now that 16k UK residents have died and it's rising.  He keeps replying to fawning praise comments with shy thanks, urges to keep spreading his message of "sanity" - I'm sure he's seen mine, but has no reply so far.

4-21: a more readable (still confusing) comparison graph - note the gap between lines with magenta higher than orange just reflects the different units (cases per 100mil, calculated from 66 million rounded off UK pop) vs. plain old cases (out of what I'd call 67.8 mil, and would put his converted points a hair lower). That's what I plot in orange - the original, unconverted WHO case numbers.  Gap aside, they should and do meet right at his scale cutoff lines (same base data), then vary between - his follows the overall distortion and mine maps that level's averaged scale with a traditional linear curve, on five scales, for five separate curves telling very different stories, as shown by the numbers if not the shapes.
These overall slope shapes could be called rope up the stairs (orange)  vs. snake up the stairs (because it has a spine, muscles - and if snake is leading … it's leading in the right way.)

Hoax Cancelled? New York's Numbers Break Mather's Scale
add 4-23: Mather had predicted low deaths in the US as well before the virus got bored. I'd have to dig for the number he gave, but at 48k already and growing steadily, it's sure to have surpassed that by far. New York State has led in cases and deaths from nearly the start, at time having more than the rest of the nation combined. He seems to have no clue how viruses spread - seeing the numbers grow exponentially in New Yok City, he finds it implausible and probably a sign of fakery.

April 22: "Coronavirus or Cuomovirus? Twilight Zone New York under Governor Cuomo"
https://www.youtube.com/watch?v=sDa5HEgYaPU
In this video he's just talking on a bench outdoors (rebel!) presaging the terrible, exhausting lie he was about to reveal about NY governor Mario Cuomo's "fantastical battle." A whole new class of lies emerged, painting the virus as 5-10 times worse than it had been anywhere else (according to his optical illusions, anyway). He says he "cannot afford to be wrong" so has been taking his time.

Next, "Coronavirus Cuomovirus Cuomo and New York Virus Fraud" (also April 22)
https://www.youtube.com/watch?v=OKDkS3coKJM
then not talking at all, just showing the horror. The text reads in part: "Two [sic] tired to discuss what I'm discovering and revealing" and at the end "I've pushed it as far as I can" According to this data, NY had something 30x worse than coronavirus "the Cuomovirus?" He finds a span where NY had more cases than the rest of the US combined especially unlikely. (note the new charts have a label the old ones didn't: "LOG SCALE" (logarithmic, like the numbers always said))

Another comparison shows how close are the total for NY state, NY city, and the rest of the US. It looks odd, but this is on the high end of the scale - those differences are larger than they appear. At the end it's about 750k for the whole country (current 4-18/19), with well under half - 325k - being in NY (a bit higher than I see - 264k even now). About half of those (~150-160k) are in NYC (I see 146k now). That's 19.5% of all confirmed US cases - they may have higher confirmation than usual, but still may have 2-10 times as many cases un-identified.

I find that level slightly surprising, but exponential virus growth works best in dense urban areas, so it's not that strange NY city should lead the world and, once it's done that, to lead by a wide margin. It was by far the worst place on Earth to avoid COVID19, but will be getting better, as other places get worse. And Mather might be alarmed to find governor Cuomo may have collaborators; Gov. Phil Murphy oversees New Jersey, which is 2nd worst-off in the Union, part of a 2-state hot-spot. At almost exactly 100,000 cases for 8.88 million = 1,126,126 (1.13%). New York state has an average of about 1.35% confirmed.
Finally (?) "CvFNF ChartUpdate 200423a - Cancelled" 23rd April: there will be no chart update.
https://www.youtube.com/watch?v=_rxGexSEns8
He was beyond exhausted and planning to stay that way. He already gave us everything we need to "break three governments" - as he puts it the British one, and one headed by Justin Truedeau, and one by Bill Gates. In a comment, I said:
I get exhausted every day, but some sleep fixes it. You … I was wondering how long you would keep up this running joke. … (I explain again his hoax) ... But finally New Yok's numbers wouldn't fit - you were going to come explain how that was a huge lie, thought better of it and developed exhaustion. Please just stay exhausted, thanks."

In response, Mather's first reply to me: "wow - a troll - we see so few these days. Thank you for taking the time to reassure us they're not extinct" - nothing to rebut the facts.

What caused this natural exhaustion? I guessed New York fared far worse than anywhere else, so even his plotting trick could barely conceal the continued rise. It might even push past his 1 million mark into the 1-10 mil band he didn't even include previously. I converted some 1/3 of the daily totals to his units - cases per 100 million people. For a state of 19.5 mil, you times the case number by 5.128, so case #1 puts them at the 5 line. By that, on April 13 they broke the line of 1 million per 100 million, or 1% of New Yorkers confirmed infected. That was before he even looked, around the 20th, so he'd be dismayed and informed all at once. Here I compared my own plotting of nearly half the days (orange, labeled) on his diminishing scale, atop the first set of curves for UK, Italy, etc. This time I broke each line down about how he does, so I can plot it better (note: my graph or points got a bit messed up at the 100k line, so I fudged a smoother curve there only).

Even here it LOOKS like it's flattening just a bit higher and later than anywhere else. Mather could say it's mellowing, no help from lockdown, but ... when some 1.35% is confirmed as having it, maybe 3-10 times that many have had it, so many know someone who's died or came close or they can see hospitals overwhelmed, they might know it's taking about 7-800 dead per day, nearing 18k total. That's close to what all the UK has suffered in a little longer span. It's been steadily worsening, finally in recent days decreasing in its rate. People living in that (and there are a lot of them) just might balk if anyone tries to tell them "see, it's nearing exhaustion, just about to stop spreading. Why did you panic?" The people are closer to exhaustion than the virus, but most of them appreciate that - thanks to lockdown - it's not twice this bad or worse, like it could be.

Or - seeing it look flat on the computer screen isn't as convincing when you live the reality that 900,000 is just NOT the same thing as NINE.

postscript: maybe I provoked him, maybe he'll do the above, but he couldn't stay down even with this much blood potentially on his hands. But he had to come back with a longer video woking in New York ... ATM I'm too exhausted to even look, but …
CvFNF ChartUpdate Restored 200422a
https://www.youtube.com/watch?v=FXPBkwi1uvQ
One among the cult-like comments shows just the effect Mather hoped to have in the UK and now in New York City and the whole world with his epic hoax (and there might be legal proceedings, we'll see):