Board 8 > Politics Containment Topic 268: Hope & Change

Topic List
Page List: 1 ... 3, 4, 5, 6, 7, 8, 9, 10
red sox 777
02/14/20 7:59:35 PM
#301:


If you lacked the information about the previous cards, your 1/13 prediction is better than someone else's 100% prediction. But you are still wrong, and it's because you lacked information.

---
September 1, 2003; November 4, 2007; September 2, 2013
Congratulations to DP Oblivion in the Guru Contest!
... Copied to Clipboard!
Corrik7
02/14/20 8:00:10 PM
#302:


red sox 777 posted...
People are misunderstanding this. The Clinton 99% people were obviously wrong. They could have had a great model and a correct prediction based on the information they had, but then their information was lacking.

It's like the question, what are the odds of the next card in a pack of cards being a K? You might say 1/13. But what if the previous cards were A 2 3 4 5 6 7 8 9 10 J Q? Now it's overwhelmingly likely you have an unshuffled pack - and the probability of the next card being K is nearly 100%.
That's what I am trying to say. If an insignificant result happens. Then you have to assume your model was just wrong and wonder how it went wrong. Not just be like, that result happens one time in 100. It happens.

---
Xbox Live User Name - Corrik
Currently playing: Spider-Man (PS4), Quantum Break (X1)
... Copied to Clipboard!
HeroDelTiempo17
02/14/20 8:02:30 PM
#303:


Corrik7 posted...


You can't be more or less accurate in this scenario so it becomes pointless.

This is maybe being pedantic but you can because accuracy isn't binary which is what you personally are looking for.

I guess for a practical example, do you ever use weather forecasts? If you do, you're relying on statistics like this. They're not always right but if you check them reliably, obviously there is SOME accuracy that's useful even if it's not perfect.

---
DPOblivion was far more determined than me.
... Copied to Clipboard!
charmander6000
02/14/20 8:06:57 PM
#304:


That's the issue with models predicting an n=1 results, anyone can say anything. That's why we need replicates in statistics.

---
Congratulations to Advokaiser for winning the guru contest.
... Copied to Clipboard!
red sox 777
02/14/20 8:07:31 PM
#305:


Another thought experiment for LOTM:

Suppose I predict 100 coin flips. I predict half the coin flips as having a 51% chance of coming out heads and half the flips as having a 51% chance of coming out tails. 100 coins are flipped and the result I said was more likely is right 100 times out of 100.

Do these results suggest my model is worse than if it only predicted 51 coins out of 100 correctly? I don't think so. I think my model was wrong, but the error was that it was actually more powerfully predictive than I gave it credit for.

---
September 1, 2003; November 4, 2007; September 2, 2013
Congratulations to DP Oblivion in the Guru Contest!
... Copied to Clipboard!
HeroDelTiempo17
02/14/20 8:14:03 PM
#306:


charmander6000 posted...
That's the issue with models predicting an n=1 results, anyone can say anything. That's why we need replicates in statistics.

alternate timeline where the government is run by math nerds and we hold the same election every day for a month to get a reliable sample size

---
DPOblivion was far more determined than me.
... Copied to Clipboard!
DoomTheGyarados
02/14/20 8:21:29 PM
#307:


At least we know Steyer won't be screwing bernie lol

---
Sir Chris
Doom The Kanto Saga - Animated Series - https://www.youtube.com/watch?v=6hH4wNFCrLM
... Copied to Clipboard!
charmander6000
02/14/20 8:26:07 PM
#308:


HeroDelTiempo17 posted...
alternate timeline where the government is run by math nerds and we hold the same election every day for a month to get a reliable sample size

I mean usually these models contain state results and sometimes senate/house predictions...

---
Congratulations to Advokaiser for winning the guru contest.
... Copied to Clipboard!
HeroDelTiempo17
02/14/20 8:31:04 PM
#309:


charmander6000 posted...
I mean usually these models contain state results and sometimes senate/house predictions...

I'm aware but we need to be able to prove the statistical reliability for the one election one way or the other for corrik

---
DPOblivion was far more determined than me.
... Copied to Clipboard!
xp1337
02/14/20 8:33:08 PM
#310:


I would like to issue a correction! I wrongly said SC was 3 days after Nevada and I'm wrong, it's a week. The 3 day gap is SC -> Super Tuesday. i apologize for the error.

More Bloomberg oppo: https://twitter.com/MattBinder/status/1228380714669871109

Bloomberg on Fox News in 2011 saying he's a friend of Trump and calls him a "New York icon." But the more damning part is he downplays Trump's role in the birther stuff and says it was "more than one person."

---
xp1337: Don't you wish there was a spell-checker that told you when you a word out?
... Copied to Clipboard!
DoomTheGyarados
02/14/20 8:41:48 PM
#311:


I would be delighted lifting this came down to bloomberg vs Bernie. Because in that debate we would witness a murder on live tv

---
Sir Chris
Doom The Kanto Saga - Animated Series - https://www.youtube.com/watch?v=6hH4wNFCrLM
... Copied to Clipboard!
Corrik7
02/14/20 8:42:36 PM
#312:


I'm pretty sure like 90% of these elites are Donald's friends who turned on him when he got the nomination cuz of politics. I wonder if they will be friends again whenever he is out of the white house.

---
Xbox Live User Name - Corrik
Currently playing: Spider-Man (PS4), Quantum Break (X1)
... Copied to Clipboard!
FFDragon
02/14/20 8:43:35 PM
#313:


lmao have we talked about Trump trying some QPQ with Cuomo and the NY AG shutting him down?

https://mobile.twitter.com/newyorkstateag/status/1228048757641678853

---
If you wake up at a different time, in a different place, could you wake up as a different person?
#theresafreakingghostafterus
... Copied to Clipboard!
LordoftheMorons
02/14/20 8:50:47 PM
#314:


red sox 777 posted...
Another thought experiment for LOTM:

Suppose I predict 100 coin flips. I predict half the coin flips as having a 51% chance of coming out heads and half the flips as having a 51% chance of coming out tails. 100 coins are flipped and the result I said was more likely is right 100 times out of 100.

Do these results suggest my model is worse than if it only predicted 51 coins out of 100 correctly? I don't think so. I think my model was wrong, but the error was that it was actually more powerfully predictive than I gave it credit for.
Assuming I'm understanding your setup correctly and exactly half are the slightly heads biased coins and exactly half are the slightly tails biased coins, we can distinguish those two models by using higher moments if we have enough data. In particular, you've slightly reduced the variance of the outcomes (this is obvious if you push the deviation all the way up so half of the coins are guaranteed to be heads and half are guaranteed to be tails; in that case we're guaranteed to get exactly 50 of each). In principle you could continue to look at higher and higher moments to recreate the entire probability distribution.

---
Congrats to Advokaiser for winning the CBX Guru Challenge!
... Copied to Clipboard!
LordoftheMorons
02/14/20 8:51:49 PM
#315:


DoomTheGyarados posted...
I would be delighted lifting this came down to bloomberg vs Bernie. Because in that debate we would witness a murder on live tv
I invite you to remember the Dems that hoped that Trump would win the 2016 GOP primary under the assumption that he would be guaranteed to lose the general...!

---
Congrats to Advokaiser for winning the CBX Guru Challenge!
... Copied to Clipboard!
CaptainOfCrush
02/14/20 8:52:42 PM
#316:


Corrik7 posted...
I'm pretty sure like 90% of these elites are Donald's friends who turned on him when he got the nomination cuz of politics. I wonder if they will be friends again whenever he is out of the white house.
If I was an acquaintance with someone and then noticed them changing their behavior for the worse (example: founding the Birther movement), I'd probably criticize them and disassociate myself. Friendship and loyalty aren't unconditional, not if your friend became an asshole.

Like, if Trump somehow ran and won with Kasich's temperament and platform, his Hollywood friends would probably still be his friends.

---
... Copied to Clipboard!
Corrik7
02/14/20 8:57:21 PM
#317:


CaptainOfCrush posted...
If I was an acquaintance with someone and then noticed them changing their behavior for the worse (example: founding the Birther movement), I'd probably criticize them and disassociate myself. Friendship and loyalty aren't unconditional, not if your friend became an asshole.

Like, if Trump somehow ran and won with Kasich's temperament and platform, his Hollywood friends would probably still be his friends.
You have a literal example of someone who "hates" Donald Trump that has the same ideas on stop and frisk and is relatively downplaying the birther movement right above. I am not convinced the Clinton's or the Epstein's or the other Hollywood elites or so on would have "disassociated" with him at all if he was exactly the same yet just not republican nominee.

---
Xbox Live User Name - Corrik
Currently playing: Spider-Man (PS4), Quantum Break (X1)
... Copied to Clipboard!
DoomTheGyarados
02/14/20 9:03:30 PM
#318:


LordoftheMorons posted...
I invite you to remember the Dems that hoped that Trump would win the 2016 GOP primary under the assumption that he would be guaranteed to lose the general...!

I think I am smarter than the dems.

Also polling shows Bernie slaughtering Bloomberg 1-1!

---
Sir Chris
Doom The Kanto Saga - Animated Series - https://www.youtube.com/watch?v=6hH4wNFCrLM
... Copied to Clipboard!
ChaosTonyV4
02/14/20 9:06:00 PM
#319:


LordoftheMorons posted...
I invite you to remember the Dems that hoped that Trump would win the 2016 GOP primary under the assumption that he would be guaranteed to lose the general...!

Bernie is arguably the Trump in this scenario, and trust me when I say we havent forgotten the lessons of 2016.

---
Phantom Dust.
"I'll just wait for time to prove me right again." - Vlado
... Copied to Clipboard!
red sox 777
02/14/20 9:06:14 PM
#320:


LordoftheMorons posted...
Assuming I'm understanding your setup correctly and exactly half are the slightly heads biased coins and exactly half are the slightly tails biased coins, we can distinguish those two models by using higher moments if we have enough data. In particular, you've slightly reduced the variance of the outcomes (this is obvious if you push the deviation all the way up so half of the coins are guaranteed to be heads and half are guaranteed to be tails; in that case we're guaranteed to get exactly 50 of each). In principle you could continue to look at higher and higher moments to recreate the entire probability

In the experiment we don't know that the coins are slightly biased. That is just our model.

I think with the results, it is overwhelmingly likely that the coins are extremely biased. But in the same direction as we had predicted a small bias. So our model is wrong and yet more powerfully predictive than it claimed to be.

---
September 1, 2003; November 4, 2007; September 2, 2013
Congratulations to DP Oblivion in the Guru Contest!
... Copied to Clipboard!
LordoftheMorons
02/14/20 9:07:21 PM
#321:


DoomTheGyarados posted...
I think I am smarter than the dems.

Also polling shows Bernie slaughtering Bloomberg 1-1!
This is before Michael Bloomberg bribes literally every American with $200 personally...!

---
Congrats to Advokaiser for winning the CBX Guru Challenge!
... Copied to Clipboard!
DoomTheGyarados
02/14/20 9:07:56 PM
#322:


LordoftheMorons posted...
This is before Michael Bloomberg bribes literally every American with $200 personally...!

Yang offered 1000. Those are rookie numbers.

---
Sir Chris
Doom The Kanto Saga - Animated Series - https://www.youtube.com/watch?v=6hH4wNFCrLM
... Copied to Clipboard!
red13n
02/14/20 9:09:59 PM
#323:


LordoftheMorons posted...
This is before Michael Bloomberg bribes literally every American with $200 personally...!
Bloomberg wont spend big on anything that isnt an ad glorifying himself. The alternative is literal prison labor.

---
"First thing that crosses my mind: I didn't get any GameFAQs Karma yesterday." Math Murderer after getting his appendix removed.
... Copied to Clipboard!
LordoftheMorons
02/14/20 9:49:55 PM
#324:


red sox 777 posted...
In the experiment we don't know that the coins are slightly biased. That is just our model.

I think with the results, it is overwhelmingly likely that the coins are extremely biased. But in the same direction as we had predicted a small bias. So our model is wrong and yet more powerfully predictive than it claimed to be.
The model can only estimate a probability distribution; it cant say anything to distinguish between different processes that give the same probability distribution function (e.g. flipping 100 fair coins or flipping 100 double faced coins that each have a 50% chance of being double heads or double tails). But if you have two models with even slightly different distributions, they can be distinguished in the limit of infinite data.

---
Congrats to Advokaiser for winning the CBX Guru Challenge!
... Copied to Clipboard!
LordoftheMorons
02/14/20 10:10:27 PM
#325:


Anyway maybe people should have saved these plutocrat allegations for Mike Bloomberg instead of deploying them on freaking Mayor Pete...!

---
Congrats to Advokaiser for winning the CBX Guru Challenge!
... Copied to Clipboard!
DoomTheGyarados
02/14/20 10:24:28 PM
#326:


LordoftheMorons posted...
Anyway maybe people should have saved these plutocrat allegations for Mike Bloomberg instead of deploying them on freaking Mayor Pete...!

Why not both

---
Sir Chris
Doom The Kanto Saga - Animated Series - https://www.youtube.com/watch?v=6hH4wNFCrLM
... Copied to Clipboard!
MoogleKupo141
02/14/20 10:25:25 PM
#327:


petes not a plutocrat, hes just a tool of them*

*i dont actually know the legal definition of plutocrat
---
For your BK_Sheikah00.
At least Kupo has class and doesn't MESSAGE the people -Dr Pizza
... Copied to Clipboard!
MoogleKupo141
02/14/20 10:27:09 PM
#329:


this is a very funny tweet about Pete

https://twitter.com/diedoktor/status/1228397860393189380?s=21
---
For your BK_Sheikah00.
At least Kupo has class and doesn't MESSAGE the people -Dr Pizza
... Copied to Clipboard!
LordoftheMorons
02/14/20 10:27:42 PM
#330:


Because the allegations against Pete are very flimsy and if you paint every single opponent as a terrible person nobodys going to take you seriously when youre right.


---
Congrats to Advokaiser for winning the CBX Guru Challenge!
... Copied to Clipboard!
red sox 777
02/14/20 10:43:02 PM
#331:


LordoftheMorons posted...
The model can only estimate a probability distribution; it cant say anything to distinguish between different processes that give the same probability distribution function (e.g. flipping 100 fair coins or flipping 100 double faced coins that each have a 50% chance of being double heads or double tails). But if you have two models with even slightly different distributions, they can be distinguished in the limit of infinite data.

We don't have infinite data. Nor do we need it for our purposes.

I guess the question is, after running this experiment you get to predict one more coin flip. The model says 51% heads. What do you think the probability is? Do you think it's closer to 51% or 99%?

---
September 1, 2003; November 4, 2007; September 2, 2013
Congratulations to DP Oblivion in the Guru Contest!
... Copied to Clipboard!
red sox 777
02/14/20 10:51:28 PM
#332:


If you want to talk in terms of the underlying probability distribution, for this thought experiment it is 99% whatever the model predicts as having a 51% chance.

So now, should your model be considered less accurate because it predicts the correct outcome almost 99% of the time rather than 51% of the time as it claims it should?

---
September 1, 2003; November 4, 2007; September 2, 2013
Congratulations to DP Oblivion in the Guru Contest!
... Copied to Clipboard!
DoomTheGyarados
02/14/20 10:53:46 PM
#333:


LordoftheMorons posted...
Because the allegations against Pete are very flimsy and if you paint every single opponent as a terrible person nobodys going to take you seriously when youre right.

And yet people think bernie being called a socialist is a problem. Strange.

---
Sir Chris
Doom The Kanto Saga - Animated Series - https://www.youtube.com/watch?v=6hH4wNFCrLM
... Copied to Clipboard!
red sox 777
02/14/20 10:57:06 PM
#334:


And this is really relevant because this is analogous to 538 in 2012 - should their model be considered worse because it called more states correctly than the model itself would have expected itself to call? And I think the answer is no.

There is something being changed between the statements "Hillary has a 70% chance of winning" and "hold 100 elections, and Hillary wins 70 of them on average." I think that is where the error slips in that lead to the absurd results we're seeing in the thought experiments. Namely, the 2nd statement here subtly implies that seeing the outcome of election 1 doesn't change the probability (in the eyes of the observer) of the other 99 elections. It does change, because the 100 elections are highly correlated events. 538 gets this correct, btw.

---
September 1, 2003; November 4, 2007; September 2, 2013
Congratulations to DP Oblivion in the Guru Contest!
... Copied to Clipboard!
LordoftheMorons
02/14/20 10:57:37 PM
#335:


red sox 777 posted...
We don't have infinite data. Nor do we need it for our purposes.

I guess the question is, after running this experiment you get to predict one more coin flip. The model says 51% heads. What do you think the probability is? Do you think it's closer to 51% or 99%?
There's a definitely possibility that I don't understand your setup correctly, but the best estimate is 51% if I use no extra information (like in reality I would use the fact that I know a normal coin gives a 50% chance and for N flips of a fair coin the standard deviation is sqrt(N/4), so unless there were much more than 2500 flips 1% is a reasonable discrepancy).

---
Congrats to Advokaiser for winning the CBX Guru Challenge!
... Copied to Clipboard!
red sox 777
02/14/20 11:09:35 PM
#336:


LordoftheMorons posted...
There's a definitely possibility that I don't understand your setup correctly, but the best estimate is 51% if I use no extra information (like in reality I would use the fact that I know a normal coin gives a 50% chance and for N flips of a fair coin the standard deviation is sqrt(N/4), so unless there were much more than 2500 flips 1% is a reasonable discrepancy).

The setup is that you don't know the actual probability distribution. You know the outcome of 100 trials and you know that someone had a model before this set of trials that predicted all of them correctly while telling you that they assigned a 51% probability to their being right in each trial. You are being asked to create your model of the situation in light of these facts. Since you have the 51% model you can use it to generate future predictions.

The reasonable conclusion is that the coin has a much higher probability than 51% of turning out the way the 51% model says is 51% likely. Notice how the number 51% is not based in any actual data. It was pulled out of thin air. The person who created that model could have replaced it with 62% or 35% and it wouldn't change the analysis one jot. By the outcome of the 100 trials (you can make it 1 million if you think 100 isn't a big enough sample to you intuitively), you know that these coin flips can be predicted to a high degree of accuracy. Because the 51% model has already done it. Even if it doesn't claim that it should be able to.

---
September 1, 2003; November 4, 2007; September 2, 2013
Congratulations to DP Oblivion in the Guru Contest!
... Copied to Clipboard!
red sox 777
02/14/20 11:13:15 PM
#337:


Again, suppose the people with the Hillary 99% prediction again give a 1% chance to the next winner. And the next. And the next 100. Clearly they have a very powerful model - you just have to flip their predicted winner and loser to get the actual results.

A model that lacks predictive power cannot generate accurate predictions by mechanically applying some consistent algorithm to it.

---
September 1, 2003; November 4, 2007; September 2, 2013
Congratulations to DP Oblivion in the Guru Contest!
... Copied to Clipboard!
red sox 777
02/14/20 11:15:55 PM
#338:


Thanks. This discussion has been interesting to me; I think that statement that 538's model would be more successful if it got fewer elections right always struck me as suspicious and I think I've worked out the problems with it.

---
September 1, 2003; November 4, 2007; September 2, 2013
Congratulations to DP Oblivion in the Guru Contest!
... Copied to Clipboard!
LordoftheMorons
02/14/20 11:26:29 PM
#339:


red sox 777 posted...
The setup is that you don't know the actual probability distribution. You know the outcome of 100 trials and you know that someone had a model before this set of trials that predicted all of them correctly while telling you that they assigned a 51% probability to their being right in each trial. You are being asked to create your model of the situation in light of these facts. Since you have the 51% model you can use it to generate future predictions.

The reasonable conclusion is that the coin has a much higher probability than 51% of turning out the way the 51% model says is 51% likely. Notice how the number 51% is not based in any actual data. It was pulled out of thin air. The person who created that model could have replaced it with 62% or 35% and it wouldn't change the analysis one jot. By the outcome of the 100 trials (you can make it 1 million if you think 100 isn't a big enough sample to you intuitively), you know that these coin flips can be predicted to a high degree of accuracy. Because the 51% model has already done it. Even if it doesn't claim that it should be able to.
Ah okay yeah your setup is different than what I was thinking. Anyway, I'm not sure that your example is in the same category of question. When I specify a model, the outputs are not binary predictions but probability distributions, whereas you seem to be both. Here you're saying that your model gives both a probability and a binary outcome. Is there any relationship between the two (e.g. is the binary outcome determined by saying that the outcome above 50% is the one that it "called")?

Another point (which you kind of brought up earlier) is that 538 is not making 50 independent predictions, though the way they display their data may lead one to believe that. The probability distribution is over sets of 50 outcomes of each state (and those results are correlated). This makes it much more likely that they'll call 49 or 50 states right than we'd naively expect viewing the races as independent.

---
Congrats to Advokaiser for winning the CBX Guru Challenge!
... Copied to Clipboard!
StealThisSheen
02/14/20 11:33:39 PM
#340:


Guys, I know you've been on the edge of your seats

But being in Nevada and running out of time, and with Warren seeming dead in the water

I am committing to Bernie.

Please take a moment to react accordingly.

---
Seplito Nash, Smelling Like the Vault since 1996
Step FOUR! Get Paid!
... Copied to Clipboard!
LordoftheMorons
02/14/20 11:34:25 PM
#341:


StealThisSheen posted...
Guys, I know you've been on the edge of your seats

But being in Nevada and running out of time, and with Warren seeming dead in the water

I am committing to Bernie.

Please take a moment to react accordingly.
Wait I thought you lived in KY

---
Congrats to Advokaiser for winning the CBX Guru Challenge!
... Copied to Clipboard!
StealThisSheen
02/14/20 11:36:14 PM
#342:


I moved last year

Fuck Kentucky

Not Louisville, though. My job is still in Louisville.

But they let me work remote

So fuck the rest of Kentucky

---
Seplito Nash, Smelling Like the Vault since 1996
Step FOUR! Get Paid!
... Copied to Clipboard!
DoomTheGyarados
02/14/20 11:37:31 PM
#343:


StealThisSheen posted...
Guys, I know you've been on the edge of your seats

But being in Nevada and running out of time, and with Warren seeming dead in the water

I am committing to Bernie.

Please take a moment to react accordingly.

Welcome!

---
Sir Chris
Doom The Kanto Saga - Animated Series - https://www.youtube.com/watch?v=6hH4wNFCrLM
... Copied to Clipboard!
StealThisSheen
02/14/20 11:38:13 PM
#344:


Also while I have committed to Bernie I think I'm willing to visit Steyer two weekends a month.

I will wait while all zero of you celebrate that.

---
Seplito Nash, Smelling Like the Vault since 1996
Step FOUR! Get Paid!
... Copied to Clipboard!
DoomTheGyarados
02/14/20 11:39:14 PM
#345:


Steyer seems nice

---
Sir Chris
Doom The Kanto Saga - Animated Series - https://www.youtube.com/watch?v=6hH4wNFCrLM
... Copied to Clipboard!
StealThisSheen
02/14/20 11:42:39 PM
#346:


Atleast the Vegas area of Nevada is very into Steyer for some reason.

I kinda get it now whereas I was baffled earlier, but I get it more in a second or third place kind of way, not in a "This is my candidate" kind of way.

---
Seplito Nash, Smelling Like the Vault since 1996
Step FOUR! Get Paid!
... Copied to Clipboard!
LordoftheMorons
02/14/20 11:44:10 PM
#347:


Steyer does seem like a nice guy

---
Congrats to Advokaiser for winning the CBX Guru Challenge!
... Copied to Clipboard!
Suprak the Stud
02/14/20 11:48:16 PM
#348:


StealThisSheen posted...
Guys, I know you've been on the edge of your seats

But being in Nevada and running out of time, and with Warren seeming dead in the water

I am committing to Bernie.

Please take a moment to react accordingly.

I was big on Warren earlier too but wouldn't waste my vote on her now. She finished at 9% in NH. Might as well vote for someone who agrees with you on like 90-95% of things that might actually win!

---
Moops?
"I thought you were making up diseases? That's spontaneous dental hydroplosion."
... Copied to Clipboard!
red sox 777
02/14/20 11:48:38 PM
#349:


LordoftheMorons posted...
Ah okay yeah your setup is different than what I was thinking. Anyway, I'm not sure that your example is in the same category of question. When I specify a model, the outputs are not binary predictions but probability distributions, whereas you seem to be both. Here you're saying that your model gives both a probability and a binary outcome. Is there any relationship between the two (e.g. is the binary outcome determined by saying that the outcome above 50% is the one that it "called")?

Another point (which you kind of brought up earlier) is that 538 is not making 50 independent predictions, though the way they display their data may lead one to believe that. The probability distribution is over sets of 50 outcomes of each state (and those results are correlated). This makes it much more likely that they'll call 49 or 50 states right than we'd naively expect viewing the races as independent.

Yes, technically the binary prediction based on the 51% model is a second model. And since the binary model was created after seeing the outcome of the trials, we must exercise some caution. Although the numbers were designed here to make the evidence overwhelming, retrospective modeling notwithstanding.

And yeah, that's another reason 538's results based on the transformation of its model into binary predictions will usually be more accurate in some sense than the underlying model. But sometimes it will be systematically wrong.

---
September 1, 2003; November 4, 2007; September 2, 2013
Congratulations to DP Oblivion in the Guru Contest!
... Copied to Clipboard!
KamikazePotato
02/14/20 11:49:34 PM
#350:


Warren is my #2 pick out of this pool so it's a shame she imploded

---
Black Turtle did a pretty good job.
... Copied to Clipboard!
red sox 777
02/15/20 12:00:15 AM
#351:


In other news, Michael Avenatti was convicted of extorting Nike today. He's looking at some substantial prison time.

---
September 1, 2003; November 4, 2007; September 2, 2013
Congratulations to DP Oblivion in the Guru Contest!
... Copied to Clipboard!
Topic List
Page List: 1 ... 3, 4, 5, 6, 7, 8, 9, 10