Read on Twitter
Yesterday marked a full year from when I did my first daily update and I don't think I've missed a day since. It started as a very simple pivot table of the NHS death data and a simple line chart. I had an unused 8 year old twitter account and 0 followers. https://twitter.com/RP131/status/1248613329527955458
This thread is a (rather longer than planned) look back at my past year. It won't be of much interest to many but I've written it as much for myself to help remember a rather unexpected and unusual experience.
Nine days after my initial update, I added my first attempt at an animated chart: https://twitter.com/RP131/status/1251865018292936704
30-Apr was the first deaths dashboard to try and explain why the figure announced at the press conference never seemed to add up. https://twitter.com/RP131/status/1255903197815869449
My focus for May and June continued to be deaths data. The gov't dashboard was just a single page with a few CSV files so I had to scrounge data wherever I could and interpolate it from charts in weekly reports. The animations were getting a bit better. https://twitter.com/RP131/status/1271792051944857602
By early June I was finally starting to get the message across to focus on date-of-death views instead of the crazy orange yoyo of the weekly reporting cycle. One of my first properly successful tweets with 300K views and over a thousand likes. https://twitter.com/RP131/status/1268208783471321088
July I cracked and started looking at 'case' data. I'd previously considered this to be more a reflection of the rollout of the testing system but decided it was time to pay more attention to understand the justification for Leicester local lockdown. https://twitter.com/RP131/status/1278848516253790208
Also in July, PHE started publishing full date-of-death data which removed quite a lot of the guesswork and allowed us to confirm our theories about the days the national total was lower than the NHS figure. https://twitter.com/RP131/status/1280944050984685569
August I knocked up the rp131-poster tool that automated my daily twitter updates. This considerably reduced the daily effort for me, but also opened the door to a proliferation of content that became quite difficult to contain (sorry!).
12-Aug was a good day when England switched to use the 28-day death rule. I've never been sure if I played any part in that whole thing but I was pleased with the outcome either way. https://twitter.com/RP131/status/1293694922147729410
The content continued to grow, generally in response to requests/suggestions and in September the table views were a suprise hit. https://twitter.com/RP131/status/1302995461817741317
I also had to introduce the pinned index as things were getting a bit hard to navigate. I think this was successful as my monthly profile views jumped from 200K to just under a million. https://twitter.com/RP131/status/1304206990470713346
The 'doubling' charts proved to be both popular and controversial. While the +ve test numbers did eventually reach these levels in January, I still disagree with how such a blunt fear tactic was used. Could probably write a whole other thread on just this. https://twitter.com/RP131/status/1310234864294277122
October brought the tier system and a whole load of colour-coding: https://twitter.com/RP131/status/1316007146232975360
Bit of a welcome deviation when I decided to read the source code for the NHS Track and Trace app. https://twitter.com/RP131/status/1311580249021444097
The animated charts showing daily report contributions were well received. https://twitter.com/RP131/status/1313498170891137024
November saw the arrival of the bubble chart. Copied from the gov't report after November lockdown I've always liked it as a way of showing week on week changes for larger datasets. https://twitter.com/RP131/status/1333083972709081088
Couldn't help myself: https://twitter.com/RP131/status/1332285717930778631
Later in November I started to dabble with excess death charts. https://twitter.com/RP131/status/1329867832352518147
In January I bit the bullet and moved a load of the more detailed content onto an external site. I realise it made it a bit harder to find/share but I think it was a necessary step due to how much everything had grown. https://twitter.com/RP131/status/1353021507862208512
There are many others I could have mentioned. Interactive bubble chart, MSOA university testing, LA weekly change charts, age breakdowns. Then more recently there's the vaccination coverage and closer focus on hospital data. This is just what stood out to me from a quick review.
I've said many times that I'm really not sure how I ended up doing this. When I try to explain it to people outside of twitter then I tend to struggle to articulate both what I'm doing any why I do it. The simple answer though is that people just kept telling me it was useful.
Twitter stats tells me in the past year I've had 6.5 million profile views and my tweets have been seen 35.5 million times. The content outside of twitter has had 315K page views from 16K unique users and I get about a thousand people visit every day to view the detailed content.
My follower count has rather plateaued in the past 2 months but I'm still amazed I ever got a 'K' suffix in the first place. I suspect if I'd renamed my profile to brand it for covid I could have grown more but after I'd had a few mentions on Radio 4 I didn't want to change it.
I'm not sure what I'll do next. I intend to keep posting at least until all restrictions on our lives are lifted. I expect I'll restructure the content soon as I don't think we'll need the same level of detail if the numbers continue to shrink.
I think it's clear now that no amount of data will make the gov't shorten the roadmap. The slogan should really have been "dates unless data makes us delay them". The focus now should be on what opening up does to the data and whether the models predicting another wave are true.
One final milestone to end with. In August my wife persuaded me to post the buymeacoffee link. The response was totally unexpected and the continued support has been amazing. Thanks so much to everyone! https://twitter.com/RP131/status/1295768488292880386
