New Zealand Forced To Release New Vax Data: Here’s The Analysis

There’s been an update on the New Zealand vax data. The government ponied up some official numbers. Some are excited about this. Alas, like the previous leaked data, I don’t think much can be said with the new release. Once again, this result will please nobody.

New Zealand, Scoops McGoo reporting:

BREAKING: the New Zealand government today released gross mortality statistics by covid vax dose (including unvaccinated) under Official Information Act request

Combined with @BarryYoungNZ’s leaked data, we now see mortality in each dose group (including 0 doses)

In that tweet are some pictures of “excess” deaths. Which we will not trust. No, nor should you. As I have said umpteen-squared times, every “excess” death is the result of a model calculation, and unless you know the ins and outs of that model, it strengths and its weaknesses, you must not trust it. (Blog, Substack background.) Which is not to say it is wrong. But mind my warning.

We’ll instead look at the counts provided by the NZ government. I asked Scoops a question about this data, which I’ll cover below, for which I don’t yet have any answer.

The data is laid out fairly nicely, considering this is a government job. But it is incomplete. You will recall in our last analysis of the leaked data we lamented we didn’t have the 0 dosed population, and so lacked a comparison group (Blog, Substack analysis of the leaked data). Here we have some of that data, but are still lacking a comparison, as you’ll see.

The Data

The NZ government provided monthly death counts beginning from July 2019 until recently, broken down by age buckets 0-20, 21-40, 41-60, 61-80, 81-100, and 100+, each broken down by the last covid vax dose (0 – 8), and by the number of Days between last vaccination and death.

Here’s the problem, as I asked Scoop. We see at June 2021, when the vax was just being rolled out, that for the 0-20 age group there are 94 reported deaths for those with 0 doses, and “<5” deaths for those with 2 doses (there isn’t anybody at this date in this age group of just 1 dose who died).

But we do not know how many people were in the 0-20 age group, by bucket, dose, or in total. We just have the 94 and “<5″. To make a fair (or fairer) comparison, we need to look at 94/”number of 0-20 with 0 doses” and “<5″/”number of 0-20 with 2 doses”.

I’m making these numbers up, but suppose there were 1 million in the “0-20 with 0 doses” group and, since the vax was new, say, only 1,000 in the “0-20 with 2 doses”. If so, then the rates are 0.000094 and anywhere from 0.001 to 0.004, which is higher and which makes the vax look worse. But again, this is a wholly fictional guess (NZ only has 5.1 million total persons).

Now to analyze this, and to make it easy, I assumed “<5” = 2, since I also assume if it were 0 they would have reported 0 (as they did elsewhere). This doesn’t affect the analysis much at all.

I next summed over the date-age-dose “Days between last vaccination and death” bucket since the time period is only a couple of years, and since these buckets in “Days between” are kind of screwy. They are: “N/A” (for 0 shots), “≤30”, “≤90”, “≤180”, “≤365”, “365+”. I do not think there is any double counting here.

For example, one way to count this is that if somebody died with 30 days, they also logically died within all the other categories except “N/A” (less than 30 days fits in the less than 90, etc.). Yet take September 2021, the 41-60 age group with 2 doses. The number of dead for “≤30” are 37, for “≤90” are 28, and for “≤180” are 12. They aren’t double counting that way, since the numbers don’t add up. They must instead be counting each bucket separately. This makes them look independent or exclusive.

But they may be double counting using deaths from previous months more than once. Take somebody who dies in September, and so in that month is in the “≤30” bucket, but in October this same person might also be included the “≤90” totals for that month. I don’t think so, though, judging by the plots below. But if I’m wrong about that, then ignore this entire post. The Summary the NZ government provided doesn’t explain this.

This is not person-level data, only aggregates. We don’t know who went from 0 to 1 dose, and when, and from 1 to 2, and when, and so on. All we have a bucket counts. Which is a real pain in the keister. Why they can’t release individual-level data I do not know.

With all these caveats out of the way, here’s what the data looks like.

The Analysis

Let’s start with the 21-40 age group. Here’s a plot of monthly total deaths for each dose.

The red lie is 0 doses, which starts off in 2019—which is before the covid panic begins—and then drops around July 2021, after the vax is rolled out in earnest.

These are total counts and not rates! One reason the 0-dose group death counts drop is because many of these people dropped out of the 0-dose group and moved into 1 or more dose groups. There were fewer people in the 0-dose group left to die after the vax rolls out. WE DO NOT KNOW HOW MANY FEWER.

The number of deaths in the 2-dose group jumps up in January 2022. Which makes the vax look worse. But we do not know how many people are in this group. Surely a lot of people who used to be in the 0-dose group are now in the 2-dose group.

This is a huge problem for any analysis here. These curves are interesting only in isolation. They cannot be compared because we do not have the number of people in each dose group.

So can anything be said? I think so. Look at the black line. That is the total monthly deaths for everybody, regardless of dose group. Ignore the drop off in January 2024, because that is when the data ends.

Compare the black line from all dates in 2019 to all dates after, which is when the panic hit. Perhaps there is a slight uptick, perhaps not. If that is an increase after, say, July 2023, it won’t be the coronadoom itself, because attributed deaths to it petered out by then (most attributed deaths were in 2022). But that peak in that month is not unusual, compared to a similar peak before covid in 2019.

In any case, trying to discern cause from these plots carries vast uncertainties.

Incidentally, since the black curve looks like that, it makes me think I got the “Days between” aggregation right.

Now let’s look at the 0-20 age group.

Same story in the dose-comparisons as before. It can’t be done, since we don’t know bucket sizes.

But that dramatic decline in total deaths (black line) jumps out. And pay attention the y-axis, which now has a different, and smaller, scale/range than before.

We know from everywhere that covid barely affected the young: deaths counts from it were low. If I read this official chart right, to date there were 13 total deaths in kids 0-20. Overall. “Normal” death totals, from any cause, using 2019 as a reference, are about 90 a month. By 2023 it was down to about 10 a month. You can say it was the vax, but covid wasn’t killing that many, and it was killing none in 2019. And the vax was not there in 2019.

In other words, this is likely and largely some behavioral change.

Deaths in the young, especially teenagers, are often mishaps. Are kids staying inside more often? Driving less? I suppose this can be counted as a good, but the whiff of Safety First! comes seeping up from the graph. A people without any sense of adventure would be mighty dull.

However, that is mere speculation. What is not is the count itself. Another possibility, which doesn’t seem like it could be large in effect, is that people in the 0-20 bucket aged out into the 21-40 bucket after July 2022, and that New Zealanders aren’t making babies to replace them.

I have no experience with New Zealand youth and parenting culture, so if you do, you make a guess.

Here’s the 41-60 age group:

Similar story and limitations as the 21-40. Note the y-axis scale change.

Here’s the 61-80 age group:

Now we have something. Note the y-axis scale change. That black line clearly goes up. A likely culprit is the coronadoom. The peaks coincide with officially attributed deaths. And the official states do say it was mostly the old who died of it.

It’s tempting to say the vax is doing little, since we have very large bumps in the 2-dose and 3-dose groups, the latter accounting for the bulk of deaths around July 2022. But in order to say that for sure, we need those population counts by buckets.

Here’s the 81-100 age group:

Again, we have something. Note the y-axis scale change. That black line once again clearly goes up. As before, some increase is surely the coronadoom.

But here the evidence is a bit stronger the vax isn’t doing anything remarkable, especially in the 3-dose group. It’s at least likely that for the old, the vax did not do much. Yet we cannot tell with anything close to certainty with this data.

Finally, the 100+ group:

The y-axis range is small here. Well, there aren’t many people over 100. But the story seems similar to that of the nearest age group.

What about vax harms? Well, you could say the vax was causing those deaths in the 2-does and 3-dose groups (and maybe higher). But a more likely explanation is the coronadoom. In any case, we can’t tell what is killing anybody, since there is no cause of death.

This data only allows hints. And only vague ones at that.

Like I said in the previous NZ analysis, what we need is person-level data, with age, dates of shots (if any) and dates of death (in any). And officially attributed causes of death.

Does anybody have that for anywhere?

Subscribe or donate to support this site and its wholly independent host using credit card click here. Or use the paid subscription at Substack. Cash App: $WilliamMBriggs. For Zelle, use my email: matt@wmbriggs.com, and please include yours so I know who to thank.