As Scott Adams might say — two movies on the same screen! While the right was suspicious of COVID-19 modelling as part of a sinister Deep State plot to take down a roaring Trump economy, the left is now eyeing the same models as a sinister Trump plot to set himself up as a hero:
“The most cynical interpretation of all this, one I can’t quite bring myself to accept,” wrote Hayes, “is they rolled out the model showing 100k deaths after they knew it would be less than that so they could anchor everyone to that # and take a vicotry(sic) lap when ‘only’ tens of thousands died.”
(And whether you’ve realized it yet or not, the world of climate change science will never be the same again.)
LSDNBC has specialized in recruiting paranoid delusionals to be their resident experts in everything.
There’s a conspiracy under every rock and under every knuckledragger.
Chrissy Hayes is jealous of Madcow’s idiocy, he’s trying to keep up!
you’re right. in the future, computer models will be known as “computer models.”
they will enjoy the same status as stock picks made with dart boards.
Why are there no computer models for picking winning lottery tickets?
Hahahaha.
[Stands up and salutes]
Actually I could guarantee you a win, just buy every combination of numbers.
If you could afford to buy every combination that remains separated by more than one, you wouldn’t be waiting all that long to claim the grand prize either.
So as an intellectual proposition, the prediction of winning lottery numbers is more advanced than most of these things. You just can’t afford to cash in.
Please watch this very related video from Scott Adams: https://youtu.be/VoG_SRo8SJs
You would think that before making such a sweeping statement he might have checked to see if there were any other independent models out there that were similar, or better yet, even more doomsday.
Then again, I am sure he knows his audience. Just play to the crowd and give them what they want.
I my career I, and people who worked for me, used lots of models. Models can be very useful, but you must remember they may or may not predict the future accurately.
You must always validate your model against what has already happened (history matching) and against what is happening now. If your model doesn’t match it could be due to:
– bad input data
– bad or incomplete real data you are matching to
– over simplified model
– bad, or incorrect assumptions made in building or applying the model.
– etc.
Models help you make decisions. Sometimes their accuracy is mediocre or even poor, but they can provide the direction to take for your decision.
As you hear about model results how many reporters ask:
– what are the error bars on your model?
– what are the assumptions you’ve made in the model and how are you checking these assumptions?
But in this age of gotcha journalism I don’t think you’ll hear those questions, let alone get any answers
Don’t forget the hours of validation that should go into testing the model before you use it. I use geologic models a lot in my work. One of the first tests is “mining waste is free, how much should I mine?” to see the total potential volume of good stuff. If the volume you get out is greater than the volume of the mountain that’s being modeled, you’ve found a problem in the yield calculation.
Similarly, run a climate model for a million years, then look at the hottest temperature and lowest temperature available. If the seas and frozen from top to bottom, or the atmosphere is hot enough to boil lead, you’ve gone outside of what physics says can happen, and the model isn’t close enough to reality to be usable.
After that, focus on 1 cell. Make sure it does what you calculate it should what particular tests are applied to it. In Mining, the tests are economic and chemical. If your single cell tests, for all types of cells, can’t pass the basic math tests then there’s something wrong with the model and it shouldn’t be used.
As the social carnage from the cure our government’s have adopted gives people pause, maybe they will see that the exact same thing has been playing out in the climate modeling field. Only worse because the modelers in that field actually doctored the input data as well as the models. The evidence leads a reasonable person to conclude that it is deliberately fraudulent on their part. With remedial measures implemented by useful idiots.
We can only hope that models will be understood as nothing more than models and that climate modeling posing as science will be recognized as what it is – modelling.
And maybe some of the SDA crowd will recognize that this is a two headed beast. Pandemic models are models too. The mass extinction of elderly people and those with co-morbidity was predicted using the GIGO principle. The fact that China is covering it up could just be a face saving gesture and not part of a nefarious plot.
(And whether you’ve realized it yet or not, the world of climate change science will never be the same again.)
Agree. Totally agree.
Although the right should carefully document the inexpertise of many of the COVID experts. (models changing, the media reversing positions 180 degrees, WHO, etc).
If you cant trust the experts on coronavirus, why trust em on climate change? Greta might have to find another job.
You forgot to mention ” The Best In The Galaxy ”
Just go away… Don’t go away mad, but just go away, with your mumbo jumbo.
Thanx NOT.
In small part, Adams is right.
Seems what ever Trump is doing, works for him every time.
Like the song goes, ‘… can’t please everyone, may as well please yourself’
Wonder if he plays chess.
So when Canada has vastly lower deaths than our models show the liberal lamestream media will accuse Trudeau of cooking the numbers right?
“Our protective measures worked.”
I am not so sure this experience will forever change the public perception of climate science. What’s more likely is that the divide will sharpen. Those who always talked in terms of “follow the science, factual evidence” etc will double down. Those who were skeptical of it and see reason to find fault with COVID-19 projections will take that path. Then it will turn into a debate that once again nobody can really answer definitively, were the projections and the precautions taken rational or irrational? Who really knows? My gut feeling is that the total lockdown has been overkill and not as effective as its backers tell us, since in actual fact the disease seems more highly targetted on vulnerable groups whose protection could have been enhanced (if at all possible) on a much more localized and specific-to-their-situation basis. But others will say, we all stayed apart, and look, it seems to have worked fairly well. Ironically, maybe it prevented us from dying of other diseases too, and getting into car wrecks that never happened.
Will be much the same with climate change, partial successes as seen by their lobby in making changes in society can then be presented as cause for the avoidance of worst case outcomes that may never have happened anyway. Without a control experiment (an identical world with a different government structure) there’s no way of knowing these things, people will tend to form opinions based on wider generalities (trust or distrust of academics and the media, for example), as well as the small slice of the pie they can see, taste and devour. I’ve worked closely with large climate data sets and there are few things quite as complicated or difficult to assess as “uncontaminated” in an ever changing society and world. Observing protocols change and often researchers don’t know as much about that as they should, or think they do.
I have recently finished a rather massive study of the Toronto weather data that began to be observed way back in 1840. The only place this is currently available on the internet is here:
https://www.netweather.tv/forum/topic/93113-toronto-180-a-north-american-data-base-of-180-years/
Eventually I hope to have a dedicated website for this material. In doing this research (which flows from earlier studies I had done as far back as my student days) I became more and more aware of observational protocol changes, and how those might interact with constantly varying urbanization changes around the site. One interesting quirk that I discovered was that the concept of a “trace” amount of rain and snow was unknown to the observers for the first few years, then they caught on to it, and it became more and more frequently mentioned. I can’t see that trace amounts of precipitation have increased in their own frequency but noticing them increased in frequency. You wonder what other observing trends lie undetected in the records. But the main conclusion I made from this massive study was that the recent warming is rather faint compared to a much greater warming trend around 1890 to 1920 that set up the modern climate. It has continued to warm but probably not much faster than expected from the growing size of the city.
But people see what they want to see. I am sure others could look at my data sets and find “unassailable evidence” of “accelerating climate change.” There are enough data points in there to support almost any hypothesis, testing them becomes the key question.
By the way, I have also published some arctic Canada studies in that same climate forum on net-weather, you can easily find them under one heading a little further into the index of topics.
Thanks Peter. It’s not the rate of climate change that bothers me (because rapid changes from one normal to another is the base, when one looks at periods of 100,000 years or more). It’s the absurd “we’re all gonna die, all of life will end!” by a change that is on the order of 10% of what happened during the Younger Dryas.
Those who have this as their religion won’t even think of asking what has come before, or become belligerent and ask why I want to promote change (as though I think that there’s anything that humanity can do to start or stop it).
Re Hayes: Talk about your tunnel vision. Not everything is about America, not everything is about Trump. It’s like Hayes never heard of Imperial College London and their projections or the rest of the planet, for that matter.