How I forecast things - a thread on how my brain works.

I get asked this a lot, and I've had quite a few DM's ranging from that is amazing to what the hell are you talking about that's garbage recently so let me try and explain.

Its not essential reading - just interesting.

I'm semi-retired. I have quite a few qualifications, none of them medical. I'm pretty bright, high on the autism scale and when I was younger I could hardly talk let alone communicate but I overcame that in my 20's. I play host to a wide range of divergent personality types.

My main strength is being able to store data and find patterns in it. This doesn't make me a mathematician. It's more complex than that and I have always likened it to guessing the number of Jellybeans in a jar - Google it if you like.

The obvious way to guess would be to count the number of Jellybeans in a jar is to count the beans in the bottom layer, then count the layers and multiply the two. You will be fairly close. But there is another way and this is how my brain works.

You can cheat on the puzzle by taking everybody else's count and averaging them all. It's an odd, almost inexplicable solution but if you average out say the guesses of 100 other separate individuals you will be almost spot on.

My mind is capable of doing this on its own. I can make 100's of estimates all disparate and not influenced by each other and then average them out. There is a word in maths for it - it is "Probability" - yet that requires complex mathematical calculations that I don't need to do

So my own word for it is "LIKELIHOOD". By basically making a series of "guesstimates" in my mind I can average them out and provide a solution that is the most likely outcome. And oddly enough the Universe seems to like likely outcomes because more often than not I am correct.

The Jellybean example is simplistic. Let me give you a more complex real world example relating to the Covid pandemic. And I am not just talking about statistics and forecasting case growth or hospital admissions.

The problem with Science is that people have different viewpoints based on the same data. It's not that they're trying to mislead people (well ok not always) but it's simply that their experience leads them to form certain conclusions since their minds have predefined pathways.

These pathways run like rivers through their logic. An epidemiologist may come to a completely different conclusion than a virologist or immunologist using the same data because the river flows in a different direction for each of them. They're all correct in their own way.

So I chose to apply the Jellybean theory to the conclusions these exceptionally bright people make. I chose to follow a number of people on Twitter, each a specialist in their own field from around the world. I store their profiles in my head like "Top Trumps" cards.

Each specialist has a weighting in each field & from that I build a "Credibility Score". I apply the same method to publications in journals like Nature, Cell, BMJ etc except this time I am storing data analyses. I don't actually "need" to understand the detail just get the drift

I can speed read and store information. I can do this faster than my PC allows me to scroll because of course I actually NEED to see the information clearly to be able to store it. (I'm not psychic ).

Once I have done this I pick a topic and then "Jellybean Jar" it to predict the most likely correct solution based on the data from various papers and the reaction to that data from around the world on twitter rated by the credibility score of the people talking about it.

I don't actually do all of this consciously (yeah I know that's weird). I just leave it running in the background. It can take, hours or days and then suddenly I will find answers popping into my head and I write them down.

In effect these are not my conclusions they are yours.

And that's it. There's other stuff I can't get answers to such as Palm Sugar & Bamboo Root consumption, or viral spread and mutation based on temperature and environment so I just work on basic pattern matching. But when you guys eventually start discussing it I will see answers.

So what I am saying is I cheat.

When you get my sudden ridiculously odd eye-rolling who in Gods name is this idiot tweets that's just the answer popping into my head.

When you get a REALLY odd question from me that's generally just me assessing your credibility.

However if I follow you then its because I rate you among the best I have come across and that isn't just your intellectual capability that is also your personality and honesty and general likeability.

And when I tweeted over the weekend about E484K and how dangerous it is - it's not me that has figured that out - how the hell could I - I'm no expert.

Its YOU lot that know it - you just don't realise it yet.

Hope that helps.

PS if you want to see a prior example of this look at how batshit angry I got on Sept 21st to 24th. I'm not linked to SAGE or in some whacky secret scientific or political or intelligence group. I don't have some sectret psychic ability.

I was just reacting to you lot.