⁣I looked at every batch in the VAERS database for Moderna's Covid 19 vaccine, and divided the batches into their alphabetic groups J, K, L, M, A, B, C, D, E, F, G, H.

I then looked through the batches in each alphabetic group. If a batch was misspelt, then I didn't treat it as a separate batch - rather I added its death count to the corresponding batch that was spelt correctly.

I ended up with a list of correctly spelled batches for each alphabet group together with the number of deaths associated with each batch.

A box plot was then created.

The box plot turned out to be even more linear than before.
The plot also showed a greater spread at higher toxicities - suggesting that higher toxicities produce more variation in outcome.

If we take the number of deaths for groups D, E, F, G, H as equal to 1 x then -
the number of deaths for C group is 2 x
the number of deaths for B group is 5 x
the number of deaths for A group is 6 x
the number of deaths for M group is 8 x
the number of deaths for L group is 10 x
the number of deaths for J group is 11 x
the number of deaths for K group is 13 x

It certainly seems that toxicity differs significantly between these alphabetic groups.

General rules of thumb are -
1. batches ending in 20A are more toxic than batches ending in 21A
2. As the alphabet ascends toxicity descends
3. For toxicity, K > L > M
4. For toxicity, A > B > C > D
5. For toxicity D = E = F = G = H

I have published the spread sheet with all data online so you can see and confirm my data.