Episode 903 Scott Adams: Good News is Breaking Out Everywhere. Swaddle Up and Get Some.
Date: 2020-04-10 | Duration: 36:15
Topics
Whiteboard1: Classic View of Reality Whiteboard2: Expert Consensus and Models The modeling process LINK to whiteboards: https://tinyurl.com/w2kberk
Rough Transcript
This is an auto-generated transcript and may contain errors.
Transcript
[0:12]
um hey everybody come on in here it’s time for the simultaneous swaddle it’s that bonus entertainment you get commercial free if you’re seeing it live and it’s coming at you with a ferocity matched only by a tiger with coronavirus yeah you need to add a tiger to the coronavirus to get this ferocity so today I’m going to show you a little bit about reality that some of you may have not been exposed to you know how I’m always talking about if you have a good talent stack you can see situations a little more clearly you can see in more different windows if you have experience in different fields so I’m going to give you a little glimpse of what I see when I look into this this whole life coronavirus situation here and see if it matches what do you think now for those
[1:13]
matches what do you think now for those of you who have similar backgrounds you’re gonna say to yourself I knew that but you might get a new way to explain it to other people so stay around for that and for some of you you’re just gonna be sort of surprised and you’re just gonna say I’m not even sure that’s real but we’ll find out but first some things in the news happening just recently did hear that Japan is going to reportedly I think I would wait for confirmation on this but reportedly Japan is going to pay its own firms to leave China they’re gonna pay them to leave ouch I think China is gonna have some serious problems have you noticed that everybody is talking about the camel Harris for Vice President president candidate with Joe Biden look
[2:13]
president candidate with Joe Biden look how close we are we’re so close because even if you don’t think she would be the best choice would you at least agree that the consensus has formed around her would you at least agree that other people have sorted decided because she’s leaving in unpredicted you know at the bedding pole she’s number one forgetting the spot then Biden has signaled it about as clearly as you possibly could so you realize that I’m on the verge of the greatest prediction of all time but it might not still still might not happen so I don’t want to get too cocky anything could happen if it doesn’t work out all I’ll be the first one to admit it but it’s so close I don’t know if you’ve noticed but there are some really hopeful signs number one the number of hospital admit II went way
[3:19]
the number of hospital admit II went way down in New York number deaths went up but I think that’s a trailing indicator meaning those are people who are already in trouble for a while but if fewer people are going into the hospital what’s that mean what’s it mean is it because of hydroxychloroquine is it because of social distancing we don’t know but it’s a good sign and everybody’s talking about the curve bending then maybe things are gonna have in the right direction but there’s another good sign one that we’ve all been waiting for my town has tolet paper I’m getting reports from other people in other places that they too do not have fully stocked two shelves but there’s at least in some stores they too have the great
[4:19]
in some stores they too have the great white the great white hope and so normality is returning approximately when I told you it would a month after the toilet paper disappeared I said to you give it one month it’ll be back there it is so let’s talk about some other stuff oh poor bored Rockler died if you don’t know who he is that won’t mean as much to you here’s a question for you when we’re done when we’re on the other side of this coronavirus there will be the heated debate of whether that the number of deaths was lower than predicted because we did such a good job or because it was never dangerous in the first place so in order to anticipate that problem that we’re not going to know why we didn’t have a lot of deaths
[5:20]
know why we didn’t have a lot of deaths and people will disagree could we agree in advance who are comparable is in other words could we find another country who is sufficiently like us or maybe a few of them who are who are doing something different let’s say they’re not doing social distancing and they’re let’s say not testing because we didn’t have that option and they’re not wearing masks at least all of them because you know we didn’t have that option either so if you tried to find somebody who is doing something smarter than us you’d have to find some of these not doing much testing and they’re also not doing much social distancing and they’ve already got you know good amount of infection and then that would be our comparable I am I am i right so I don’t know which which country that is but let me put that out there that if if anybody is claiming that they’re gonna wait until
[6:21]
claiming that they’re gonna wait until we’re on the other side of this and say see it’s self-evident that it’s because of the social distancing or it’s self-evident that that made no difference it was no problem at all it’s not going to be obvious just by looking at it you’re gonna have to have some standard to compare it to to even get a you know a hope of understanding what happened alright here’s my little whiteboard lesson for today this is what I call the classic view of reality this is the view that we think should be the view or there should be the way reality works if you are young you may have learned that this is what reality looks like you’ve got your experts they consult with your leaders your leaders make some policies and then those policies rain down on the happy citizens who are happy that experts were part of
[7:24]
who are happy that experts were part of the process and that leaders took their advice and what a good world we live in so that would be the classic view of the world on the other side of my magic whiteboard that is so incredible you probably all want one but you can’t have it it’s one of the kind here’s a more accurate view of the world goes like this you’ve got your experts and let’s say that there’s a situation that comes up that they have not seen before let’s take their coronavirus situation now they’ve seen other pandemics and other infections but they haven’t seen this one so I think you’d agree that this was a brand new thing and there was a lot of fog of war and they didn’t have good information at first but they’re the experts even though they don’t have good information they still have to
[8:26]
good information they still have to inform the leaders what to do because doing nothing is also a decision so the leaders have to make a decision to do nothing or to do something very expensive and aggressive and they’re not going to make the decision on their own and the leaders aren’t going to be able to get away with saying well we really don’t know there’s not enough information if you would just let us wait a year no we can’t wait a year might be too late so the experts are forced to make a decision well before they have anything that would satisfy themselves of what to do so what do they do well here’s what I believe happens in the real world you got lots of experts but they’re not all equal some experts are more influential let’s say dr. Fauci for example so they’re probably used several experts in the field who are recognized as the most important ones so probably it only takes half a dozen or maybe fewer than ten of the top experts
[9:26]
maybe fewer than ten of the top experts to agree that they have an instinct or something or they feel there’s a problem or this could be really bad you know that all of their all their experience their pattern recognition their biases you know good and bad I’m not even saying any of this is bad I’m saying that you know an expert is not just the data and if they don’t have data at least not reliable data they’re going to take advantage of their their pattern recognition and bias but also very importantly they don’t want to kill anyone that’s going to be job one right and that’s a really a very important bias because they’re gonna make sure that nobody can blame them for killing anybody in particular that would be anybody’s natural impulse now of course they’re also professionals so they’re gonna try to be as independent minded as possible for the greater good but put yourself in that position would you ever recommend something that has a pretty good chance of killing a million people
[10:29]
good chance of killing a million people if you get it wrong you’re gonna want to do whatever is the safest in terms of not killing a million people and not taking any chance that you’re the one who made some dumbass recommendation and killed a million people so professional as they may be they’re still humans and nobody wants to kill a million people so you’ve always got that force so what do you do if you’ve got a really strong feeling that that something is bad but you don’t know and let me give you a little sort of math example of why they would be confused in the early days of the pandemic let’s say that the thing that makes a virus powerful is just I’m oversimplifying a little bit is how viral it is and how deadly it is if you get it so two different things how quickly is spreads and then if you get it what are your odds of dying and so in
[11:32]
it what are your odds of dying and so in very rough terms you could sort of multiply them multiply them to see how bad it is so if you had let’s say a very low spreading but it was very deadly well that wouldn’t be the worst thing because you might be able to stop it because even though it’s deadly it’s not going to spread that far so multiplying the low viral ‘ti times the you know high death rate gives you you know some power number but you can get to the same power number by having something that’s super viral and doesn’t kill many percentage of people but it gets so many people they even though it’s a low percentage a lot of them die so both of them could be equal power one because it’s very viral and the other because it’s very deadly but what happens if you get something that’s super viral and by the worst turn of fate super deadly I don’t believe we’ve
[12:33]
fate super deadly I don’t believe we’ve had one of those maybe the Spanish flu qualifies but I’m not even sure it was that one was you know super deadly super viral I’m not sure how that gets classified but in the earliest days of the pandemic was there enough information that the experts could know if they had something that was very viral but only a little bit deadly or something that was a nine nine and a ten in virality and deadly they didn’t know they were looking at China and things look bad right so what do you do if you’re the experts and you’ve go to advise the policy makers what to do either to do nothing or do something and you don’t have any information well you’re going to take your best instincts and judgment your risk management your not wanting to kill millions of people and you can have to convince leaders who are going to ask you for your data what
[13:35]
are going to ask you for your data what do you do so the president says you okay you know this is important I get it now show me your numbers and what do they have well not much of anything right because they just don’t have good numbers they have guesses they have assumptions or lots of variables it’s really all very complicated then what does the president do how do you convince the president even though you’re the experts and all you have is punches and guesses and patterns and things from the past that were similar you’ve done some math in the back of an envelope huh can you convince the president and could you convince the public with something like that who do you go directly to the public and say look we can’t prove it but a whole bunch of us who are experts have a really bad feeling about this thing and we don’t want a million people to die so could they convince the public no because the public would say the same thing based on what show us your proof that
[14:38]
based on what show us your proof that this is so bad and they would say things like well there was this anecdote and there was a story and there’s this unreliable information there’s this pattern of something that was different that was there was this time we thought we were right that we were wrong and when they’re done you just say to yourself I’m not sure I’m convinced any this is real yet yet we still have to make decisions so how do you solve this situation in the real world where it’s not that classic view that’s on the other button side of the whiteboard where the experts know just what to do they look at their data they look at their spreadsheets it all adds up they tell the leaders the leaders say hey that looks good to me thank you for looking into that and they make their policies in the real world that just doesn’t work that way even the experts don’t know what to do if they don’t have data and it’s the fog of war and it’s a brand new thing now someday we might understand enough about this that if we could have traveled back
[15:39]
this that if we could have traveled back in time with our future knowledge well we maybe we would handle this differently but even the experts didn’t know what was going on thank you China right so here’s what you do in this situation where there’s something you feel really strongly about you’re pretty sure there’s a lot of evidence but you can’t communicate it you just can’t sell it and so you go to the model makers here’s where everybody in social media and on the news is making a fundamental misunderstanding and it’s this you don’t use models to give you the answer models do not produce information there are sales tool there for persuasion how do I know that I know that’s the opposite of what you think right you think that the experts are uncertain so they use a model to
[16:40]
model to become certain and to learn something that’s probably what you thought right mm-hmm ask anybody who’s ever done this for a living that’s not what happens here’s what really happens the experts say go because some models because I gotta sell this thing I need a picture give me a graph where that scary looking curve give me a few big numbers give me a range and make it a big range because I don’t want to be wrong all right whatever you do make it a big range so the bottle makers do their magic and they come up with some really scary looking graphs that are really simple that they can sell to scare the public and they can it’s compatible with the experts now what would happen hypothetically if the people making the models came up with a prediction that would kill people if it were wrong what are the experts going to do so for example let’s say the experts are looking at the carnage over in Wuhan and
[17:42]
looking at the carnage over in Wuhan and there’s they’re saying to yourselves I know there’s something bad out there there’s definitely something bad out there this is not normal as far as we can tell we haven’t confirmed it but we’re seeing a lot of stuff that just doesn’t look normal you know spray into the streets over there and everything so what happens if the model makers said well we’ve put in all the variables we’ve checked all of our assumptions and the very best model we’ve come up with says it’s not much of a problem so yeah it doesn’t look like it’ll be a problem over here so what do you do if you’re the expert because that model would disagree with what you plainly can see it disagrees with your sense of risk management it disagrees you know with observation instinct unch it just disagrees with every fiber of your being are you gonna say to the model maker oh thanks I was worried before but now that you ran the model feeling better no they discard that model because you
[18:44]
no they discard that model because you can’t you can’t keep the model that says it might not be a problem that’s the one that could kill people I remember job one job one is don’t kill anybody so the experts can’t accept a model that is - let’s - not scary because it would put the experts in the position of maybe not warning the public about whatever they’re seeing in China that doesn’t look good right so they would have to reject anything that made it look like it’s not a problem all right so let’s say the experts come back and they say it could be this big range 2 million or a hundred thousand but let’s say the first time they come back they say you know it’s 2 million or honestly it could be ten thousand what if that was the first model that came back and the experts say we’ve got a problem here
[19:45]
the experts say we’ve got a problem here if it’s only ten thousand or even if ten thousand is even in the range Americans are going to think it’s not a problem because we’re used to having things that are a huge risk but most of the time you find if you tell the Americans that even within that predicted range is maybe just ten thousand it’s like the common cold basically how are they going to act they’re gonna act in a way that they’re not taking you seriously and then what and then I’ve killed all these people because they didn’t you know potentially it could be worse than we think so here’s what happens models are simply ways that experts can convince leaders and the public that there’s something that needs to be taken seriously now you’re saying to me I’m not buying this Scott I am not buying this everybody on TV all the scientists all the experts they really sort of act like this is
[20:45]
they really sort of act like this is telling you something like it’s actually giving you a window into the future that’s the whole point of this we’re spending millions of dollars billions of dollars making them there are all these models it’s the best it’s the smartest people in the world Scott did you forget you’re a cartoonist who has a degree in economics and an MBA did you forget that because the smartest people in the world they’re not talking like you’re talking no they’re not but here’s my simple argument for why they do not glimpse the future and it goes like this if you have a situation that is really fundamentally like other situations then sometimes yes let’s say you’re into construction and you’ve built several homes that are comparable you you probably could put together a budget it says uh based on these last three homes I built in the same town that are about the same kind this square footage yeah I’ll just estimate this
[21:46]
footage yeah I’ll just estimate this next house would be that I’d be fine but that doesn’t work with a brand-new situation there’s no way to estimated and if anybody could use any complicated model to predict the future even even statistically better than guessing they would not be making models they would be sitting on their trillion dollar yachts if anybody could do this in actuality like actually predict the future with Bora with some statistical validity whatsoever they do it for stocks right they do have heard knowing what other startups to invest in they could make bets on big events in other words they could bet that there’ll be a you know tragedy or a forest fire you know this is that if anybody could look into the fog of war which is what was happening
[22:48]
fog of war which is what was happening in the beginning of the pandemic if anybody could look into that and build a model on a spreadsheet or with any kind of software there could actually reveal the future better than chance they would be almost magic and there’s no such thing as magic so this is this is me letting you peer beyond the screen part of the reason I know this is I did financial projections for a living and I can tell you that if my if my model and my prediction did not give management what they felt was the right thing to do for personal reasons professional reasons or the good of mankind I had to go back and tweak it until it did because we didn’t know what was the right thing to do but we knew that the model wouldn’t tell us we knew that the variables were the unknowns were so extreme for example one of the things I was trying to do is calculate
[23:49]
things I was trying to do is calculate whether this technology that’s you know obsolete now but was new then called ISDN would have a big future and whether it was economical to put it into the network and sell it to people and they were just gigantic unknowns about where the technology would go and what would happen with costs and if you tweaked any one of them you just get wildly different results it was either a great idea or a terrible idea it was my job to calculate that and so it was was that in other situations where I learned that you start with the answer and the answer was yeah we’re a high tech company we have to be investing in high tech stuff don’t bring me back some financials to say we don’t do that because we’re going to do that we have to do that is you know is you know our experience or pattern-recognition our bias all of our external and internal incentives tell us we’re going to invest in the new technology and it doesn’t matter bean-counter if you come back and tell us it’s a bad idea I’m
[24:49]
back and tell us it’s a bad idea I’m gonna tell you to go back and tweak a variable until it’s a good idea because that’s the business world we’re in this business we’re not going to be not in that business because your spreadsheet said so so once you see behind the curtain you realize that the models are always just an expression of what the experts or the leaders want them to be and their sales tools they are not things that tell you what is true so for all the people who are not quite seeing behind the curtain you can see them debating in public you can see it on TV you can see it on social media and they say things like this those models were inaccurate doesn’t mean anything if you actually understand what the models are for and what is possible it is meaningless to say the models turned out not to be accurate because they were never built to be accurate you can’t be accurate in the same way that the models
[25:50]
accurate in the same way that the models were not built to crap bars of gold for all of us the reason is would be great if you could build a model that would crap gold wouldn’t you do it I mean why not you’re a nice person you’d you’d give us all little gold with your algorithm but you can’t really build a model that can predict the future it’s not a thing so you can’t be mad at the people who built the model for not doing a thing that’s not a thing which is predicting the future from unknowns again if you had a situation where there were it was very similar to the last one that’s different but in the fog of war nobody can predict there’s not a thing so the people who understand the least and I would argue people who have never been around financial predicting people who are not economists maybe maybe people are not scientists they are maybe journalists if I may say so I think the journalists and
[26:51]
I may say so I think the journalists and the artists are more likely to think they don’t understand what’s going on because the predictions were so far off what we’re actually experiencing and I think to them they’re saying well was this a conspiracy was everybody stupid and I don’t think it was a conspiracy and I don’t think anybody involved was stupid and I don’t think anybody involved was acting acting with bad intent I think everybody involved used the only tools they could to get a result they thought would protect people the most and maybe with a little bias toward making sure they were not the ones who killed a million people personally so I think everybody had good intentions and used the tools they had but don’t be fooled into thinking that the models tell you what the future is that’s not a thing but they are good for persuasion and you can see how well they work now having having observed this situation
[27:54]
having having observed this situation and I think I succeeded in at least convincing some of you that it’s the way it works is not that models tell you something it’s pretty much the experts tell the models what the models need to say in order to get to the next level if you will you can see why some people have a problem with the climate models I want to get into that I just want to yeah I just want to alert you that it’s hard to form an opinion on one of these things without extending it to well does that apply to climate change and so here’s what I would ask you the climate change models again would be accurate if they are similar to things that have worked before in other words if we were good at modeling climates and this was just another one maybe we’d be good at it but we’ve never we’ve never modeled the climate 80 years in advance using
[28:56]
the climate 80 years in advance using the tools that we’re using we’ve never done it we think we’re doing it now but there’s nothing to you know to test and against I would suggest that the best way to understand the climate change models is this way as well which is that the scientist genuinely believed based on everything they’ve seen every scientific test every paper their pattern recognition the people they’ve talked to just the whole weight of their experience I think is screaming to them we’re in trouble planets going to warm up but they don’t have the ability because it’s impossible it’s not because they’re bad at it it’s just impossible to communicate all the things they know you know the full weight of their knowledge to a leader or to the public because it just doesn’t trends transmit it which just sounds like gobbledygook and we’d say I don’t know I’m not even sure you know what you’re talking about and so the scientists do the same thing that you’re
[29:57]
scientists do the same thing that you’re seeing here which is they go to them they for them the model is really persuasion but the public thinks the model is actually information the public thinks it’s an actual prediction now it might be a prediction that the temperature’s going up and you know there’s a risk but the actual model will see and by the way the model can be right if they make the range big enough if you make the range big enough there’s pretty good chance it’s going to fall in that range so that is my lesson for today I hope that was useful was it I would like to see your feedback I’ll stay just for a minute to look at that is Bill Gates in here I don’t know what you’re talking about who wins big business well I don’t know if anybody’s winning very few people are winning I see more people asking about
[31:00]
winning I see more people asking about dr. Shiva yeah so I’m still waiting for somebody to make a statement oh good I’m getting some good feedback here I’m waiting for somebody to make a statement that they think they would like my opinion on that is something that dr. Shiva says so narrow it down to me not not just like a whole topic like is there a statement or a fact just narrow down and I’ll give you an opinion I pushed your whiteboard on Twitter yeah I’ll take a picture of it as soon as we get off here oh good actually the feedback is better than I thought I didn’t know if didn’t know if that was going to work as well as it did how delightful all right um say it doesn’t seem to me that you treat climate models like this should they be treated like this well I just I just did I don’t maybe you just signed on or something I
[32:07]
wouldn’t watch you if I didn’t like your take on things that’s true are you ruining models forever well I’ll tell you models were ruined for me when I became the person who made them you know nothing will disillusion you more than being the person who’s doing that thing yeah if you want to be unimpressed with something that you used to be impressed that try try being that person on it let me let me give you a concrete example when I became a cartoonist I immediately set my sights on winning the the top award and cartooning because I thought you know if you’re an actor you probably daydream of getting the Academy Award right that would be natural so whatever business you’re in you probably spend a little time daydreaming about winning the Super Bowl or getting the gold medal or something like that so I I wanted to win this top award and cartooning it was called the Reuben and named after Rube Goldberg and every year
[33:10]
named after Rube Goldberg and every year the cartoonists get together with sort of an Academy Awards and there’s a vote and they gives a the top of word to a cartoonist so I of course wanted that and years go by something like 10 years and you know I’m not even nominated and Dilbert just takes off and gets financially successful so suddenly Dilbert is like one of the biggest things in the country and that ain’t got nominated and I said to myself wait a minute I don’t think the comic is that much better than it was a year ago or two years ago or three years ago but what is different is that events in society sort of propelled me in front of a lot of people and then the comic took off so the thing that was different is I was making a lot of money and getting a lot of attention I thought to myself what kind of what kind of award for cartooning depends on how much PR our beginning and so I actually ended up
[34:13]
beginning and so I actually ended up winning the two top awards in cartooning one is the overall ordeal the biggest one you can get and then the the top award in my category which was cartoon strips as I won them both in the same year and of course I was you know terribly honored and everything but almost as soon as I won them it got home the entire thing seemed like a farce and didn’t beat anything to me because once I got on the inside and I was the actual winner of it I just realized it wasn’t so much because of my cartooning it’s because I had become a name brand and the organization that puts on the event likes tab name brands so it makes everybody else all want to go just makes the event better if you’ve got you know some famous people there so I had I had won the award simply as a way to get me to attend the event nothing to do with anything I spent it spent years of my career like lusting
[35:13]
spent years of my career like lusting after this award that when I wanted I realised was completely meaningless if there’s anybody out there who has won that award sorry sorry about that my college my undergraduate college did the same thing when I got famous they they made me love night of the year or something and invited me to go back and give a talk and everything I thought my god this is what another I’m alumni of the year of like little college I went to it’s like then of all those people won’t they Envy me my old classmates when they see that I’m an alumni of the year and then you go back and you rise is just a fundraising thing you know once you get famous for doing something your alumni of the year so that they can have more access to your for fundraising anyway that’s more than you wanted to know and I will talk to you in the morning you know when