Board 8 > Politics Containment Topic 203: We Met at Borders

Topic List	Page List: 1 ... 2, 3, 4, 5, 6, 7, 8, 9, 10
Corrik 11/05/18 1:10:18 PM #251:	Crossfiyah posted... Corrik posted... Crossfiyah posted... Clearly because you fundamentally misunderstand probability. A solid seat is a 95% chance of winning or higher. A party will lose let's say 1 in 20 of these. A likely seat is a 75% chance. A party will lose about 1 in 4 or 5 of these. A leaning seat will be lost about 2 in 5 times. Republicans by 538s count has 135 solid seats, 49 likely seats, and 13 leaning seats. They will win an average of 128, 37, and 8 if these if I'm being generous. You are treating these 197 seats as 197 seats won. In reality they are closer to 173 seats won. Take the toss ups and split them in half and you get 182 seats for Republicans not 200 something. Do the same for Democrats now. 193 (!!!) solid seats, 17 likely seats, 10 leaning seats. Translates to 10+4+4=18 pick ups for a total of 200 seats AFTER toss ups are calculated. Not before. You're fundamentally misunderstanding how many safe or safe leaning seats will be lost. You are assuming probabilities are clear cut answers. Go back to the 2014 midterms. How many solids and likelies and such actually flipped. Okay bud. Did 25% of likelies 5% of solids, 40% of leans flip? Was it even close to those numbers? --- LoL ID = imajericho XBL GT = Corrik <cite>Corrik posted [p:911855304] in Politics C... [t:77157954]...</cite> <quote><cite>Crossfiyah posted...</cite> <quote><cite>Corrik posted...</cite> <quote><cite>Crossfiyah posted...</cite> <quote>Clearly because you fundamentally misunderstand probability. A solid seat is a 95% chance of winning or higher. A party will lose let's say 1 in 20 of these. A likely seat is a 75% chance. A party will lose about 1 in 4 or 5 of these. A leaning seat will be lost about 2 in 5 times. Republicans by 538s count has 135 solid seats, 49 likely seats, and 13 leaning seats. They will win an average of 128, 37, and 8 if these if I'm being generous. You are treating these 197 seats as 197 seats won. In reality they are closer to 173 seats won. Take the toss ups and split them in half and you get 182 seats for Republicans not 200 something. Do the same for Democrats now. 193 (!!!) solid seats, 17 likely seats, 10 leaning seats. Translates to 10+4+4=18 pick ups for a total of 200 seats AFTER toss ups are calculated. Not before. You're fundamentally misunderstanding how many safe or safe leaning seats will be lost.</quote> You are assuming probabilities are clear cut answers. Go back to the 2014 midterms. How many solids and likelies and such actually flipped.</quote> Okay bud.</quote> Did 25% of likelies 5% of solids, 40% of leans flip? Was it even close to those numbers?</quote> ... Copied to Clipboard!
Corrik 11/05/18 1:11:30 PM #252:	Suprak the Stud posted... https://apnews.com/b4a752ff50c9404c977e38bc6fe79b42 The court on Monday rejected appeals from the telecommunications industry seeking to throw out a lower court ruling in favor of the net neutrality rules. The Federal Communications Commission under President Donald Trump has rolled back the rules, but the industry also wanted to wipe the court ruling off the books. Conservative Justices Samuel Alito, Neil Gorsuch and Clarence Thomas would have granted the industrys request. Chief Justice John Roberts and Justice Brett Kavanaugh took no part in the case without offering a reason. https://www.cnet.com/news/supreme-court-declines-to-hear-appeal-of-obama-era-net-neutrality-rules/ the Trump administration pushed the Supreme Court to hear the appeal that would have wiped the ruling from the books so that the parts of the decision upholding the FCC's authority could not be used as precedent in subsequent cases. Fairly big ruling as this wouldve gutted the FCC if they won. The new swing justices will likely be Kavanaugh and Roberts with Roberts being the most important. This is the second semi-important ruling that those two have broke from the conservative wing in recent weeks. Abortion is still the big issue were waiting on a ruling from in some fashion. I dont know how Kavanaugh will side there. I have hope Roberts wont want to alter precedent too much. Nah Kavanaugh is the most conservative justice ever and so long gay marriage, abortion, and everything else. --- LoL ID = imajericho XBL GT = Corrik <cite>Corrik posted [p:911855372] in Politics C... [t:77157954]...</cite> <quote><cite>Suprak the Stud posted...</cite> <quote>https://apnews.com/b4a752ff50c9404c977e38bc6fe79b42 <quote>The court on Monday rejected appeals from the telecommunications industry seeking to throw out a lower court ruling in favor of the net neutrality rules. The Federal Communications Commission under President Donald Trump has rolled back the rules, but the industry also wanted to wipe the court ruling off the books. Conservative Justices Samuel Alito, Neil Gorsuch and Clarence Thomas would have granted the industrys request. Chief Justice John Roberts and Justice Brett Kavanaugh took no part in the case without offering a reason.</quote> https://www.cnet.com/news/supreme-court-declines-to-hear-appeal-of-obama-era-net-neutrality-rules/ <quote>the Trump administration pushed the Supreme Court to hear the appeal that would have wiped the ruling from the books so that the parts of the decision upholding the FCC's authority could not be used as precedent in subsequent cases.</quote> Fairly big ruling as this wouldve gutted the FCC if they won. The new swing justices will likely be Kavanaugh and Roberts with Roberts being the most important. This is the second semi-important ruling that those two have broke from the conservative wing in recent weeks. Abortion is still the big issue were waiting on a ruling from in some fashion. I dont know how Kavanaugh will side there. I have hope Roberts wont want to alter precedent too much.</quote> Nah Kavanaugh is the most conservative justice ever and so long gay marriage, abortion, and everything else.</quote> ... Copied to Clipboard!
xp1337 11/05/18 1:13:36 PM #253:	I think Corrik is getting his "205-200" seat thing from somewhere else and then judging 538's model off it. You just straight up can't get 538's numbers to get there. Which I mean... okay? But applying different standards and rules to 538's model when it provides its own forecasts on those numbers is pointless IMO. Just say you disagree with their model and leave it at that. By 538's numbers the "Solid" seat breakdown is 193-135. If you throw in Likely races (and I'd expect most to stay with the party they're rated for but a few will definitely be picked off) you get to 210-184. I'd balk at throwing the Lean ratings in there and saying they're somehow not competitive but if you did that then by 538's ratings you've just given the Democrats a "safe" House majority before tossups at 220 and the Republicans still haven't reached 200. tl;dr: You should ask Corrik where he's getting that seat breakdown from so we can least analyze and compare it. --- xp1337: Don't you wish there was a spell-checker that told you when you a word out? <cite>xp1337 posted [p:911855488] in Politics C... [t:77157954]...</cite> <quote>I think Corrik is getting his "205-200" seat thing from somewhere else and then judging 538's model off it. You just straight up can't get 538's numbers to get there. Which I mean... okay? But applying different standards and rules to 538's model when it provides its own forecasts on those numbers is pointless IMO. Just say you disagree with their model and leave it at that. By 538's numbers the "Solid" seat breakdown is 193-135. If you throw in Likely races (and I'd expect most to stay with the party they're rated for but a few will definitely be picked off) you get to 210-184. I'd balk at throwing the Lean ratings in there and saying they're somehow not competitive but if you did that then by 538's ratings you've just given the Democrats a "safe" House majority before tossups at 220 and the Republicans still haven't reached 200. tl;dr: You should ask Corrik where he's getting that seat breakdown from so we can least analyze and compare it.</quote> ... Copied to Clipboard!
Crossfiyah 11/05/18 1:14:35 PM #254:	Corrik posted... Crossfiyah posted... Corrik posted... Crossfiyah posted... Clearly because you fundamentally misunderstand probability. A solid seat is a 95% chance of winning or higher. A party will lose let's say 1 in 20 of these. A likely seat is a 75% chance. A party will lose about 1 in 4 or 5 of these. A leaning seat will be lost about 2 in 5 times. Republicans by 538s count has 135 solid seats, 49 likely seats, and 13 leaning seats. They will win an average of 128, 37, and 8 if these if I'm being generous. You are treating these 197 seats as 197 seats won. In reality they are closer to 173 seats won. Take the toss ups and split them in half and you get 182 seats for Republicans not 200 something. Do the same for Democrats now. 193 (!!!) solid seats, 17 likely seats, 10 leaning seats. Translates to 10+4+4=18 pick ups for a total of 200 seats AFTER toss ups are calculated. Not before. You're fundamentally misunderstanding how many safe or safe leaning seats will be lost. You are assuming probabilities are clear cut answers. Go back to the 2014 midterms. How many solids and likelies and such actually flipped. Okay bud. Did 25% of likelies 5% of solids, 40% of leans flip? Was it even close to those numbers? Odds of winning are not pulled out of a hat they are based on historic outcomes given the lead in polling. --- Jagr_68: "hittes Blake Kesseguin a super saiyan NHL star who gives earth a chance to win the Cell gaemz" <cite>Crossfiyah posted [p:911855542] in Politics C... [t:77157954]...</cite> <quote><cite>Corrik posted...</cite> <quote><cite>Crossfiyah posted...</cite> <quote><cite>Corrik posted...</cite> <quote><cite>Crossfiyah posted...</cite> <quote>Clearly because you fundamentally misunderstand probability. A solid seat is a 95% chance of winning or higher. A party will lose let's say 1 in 20 of these. A likely seat is a 75% chance. A party will lose about 1 in 4 or 5 of these. A leaning seat will be lost about 2 in 5 times. Republicans by 538s count has 135 solid seats, 49 likely seats, and 13 leaning seats. They will win an average of 128, 37, and 8 if these if I'm being generous. You are treating these 197 seats as 197 seats won. In reality they are closer to 173 seats won. Take the toss ups and split them in half and you get 182 seats for Republicans not 200 something. Do the same for Democrats now. 193 (!!!) solid seats, 17 likely seats, 10 leaning seats. Translates to 10+4+4=18 pick ups for a total of 200 seats AFTER toss ups are calculated. Not before. You're fundamentally misunderstanding how many safe or safe leaning seats will be lost.</quote> You are assuming probabilities are clear cut answers. Go back to the 2014 midterms. How many solids and likelies and such actually flipped.</quote> Okay bud.</quote> Did 25% of likelies 5% of solids, 40% of leans flip? Was it even close to those numbers?</quote> Odds of winning are not pulled out of a hat they are based on historic outcomes given the lead in polling.</quote> ... Copied to Clipboard!
Corrik 11/05/18 1:17:28 PM #255:	xp1337 posted... I think Corrik is getting his "205-200" seat thing from somewhere else and then judging 538's model off it. You just straight up can't get 538's numbers to get there. Which I mean... okay? But applying different standards and rules to 538's model when it provides its own forecasts on those numbers is pointless IMO. Just say you disagree with their model and leave it at that. By 538's numbers the "Solid" seat breakdown is 193-135. If you throw in Likely races (and I'd expect most to stay with the party they're rated for but a few will definitely be picked off) you get to 210-184. I'd balk at throwing the Lean ratings in there and saying they're somehow not competitive but if you did that then by 538's ratings you've just given the Democrats a "safe" House majority before tossups at 220 and the Republicans still haven't reached 200. tl;dr: You should ask Corrik where he's getting that seat breakdown from so we can least analyze and compare it. RCP. I don't trust 538. --- LoL ID = imajericho XBL GT = Corrik <cite>Corrik posted [p:911855695] in Politics C... [t:77157954]...</cite> <quote><cite>xp1337 posted...</cite> <quote>I think Corrik is getting his "205-200" seat thing from somewhere else and then judging 538's model off it. You just straight up can't get 538's numbers to get there. Which I mean... okay? But applying different standards and rules to 538's model when it provides its own forecasts on those numbers is pointless IMO. Just say you disagree with their model and leave it at that. By 538's numbers the "Solid" seat breakdown is 193-135. If you throw in Likely races (and I'd expect most to stay with the party they're rated for but a few will definitely be picked off) you get to 210-184. I'd balk at throwing the Lean ratings in there and saying they're somehow not competitive but if you did that then by 538's ratings you've just given the Democrats a "safe" House majority before tossups at 220 and the Republicans still haven't reached 200. tl;dr: You should ask Corrik where he's getting that seat breakdown from so we can least analyze and compare it.</quote> RCP. I don't trust 538.</quote> ... Copied to Clipboard!
Corrik 11/05/18 1:22:50 PM #256:	Crossfiyah posted... Corrik posted... Crossfiyah posted... Corrik posted... Crossfiyah posted... Clearly because you fundamentally misunderstand probability. A solid seat is a 95% chance of winning or higher. A party will lose let's say 1 in 20 of these. A likely seat is a 75% chance. A party will lose about 1 in 4 or 5 of these. A leaning seat will be lost about 2 in 5 times. Republicans by 538s count has 135 solid seats, 49 likely seats, and 13 leaning seats. They will win an average of 128, 37, and 8 if these if I'm being generous. You are treating these 197 seats as 197 seats won. In reality they are closer to 173 seats won. Take the toss ups and split them in half and you get 182 seats for Republicans not 200 something. Do the same for Democrats now. 193 (!!!) solid seats, 17 likely seats, 10 leaning seats. Translates to 10+4+4=18 pick ups for a total of 200 seats AFTER toss ups are calculated. Not before. You're fundamentally misunderstanding how many safe or safe leaning seats will be lost. You are assuming probabilities are clear cut answers. Go back to the 2014 midterms. How many solids and likelies and such actually flipped. Okay bud. Did 25% of likelies 5% of solids, 40% of leans flip? Was it even close to those numbers? Odds of winning are not pulled out of a hat they are based on historic outcomes given the lead in polling. Leans are last second breaks can swing the result. Likelies you are now looking at polling errors. Solids is basically major and drastic polling errors. Tossups is within margin of error and anyone could win. --- LoL ID = imajericho XBL GT = Corrik <cite>Corrik posted [p:911855977] in Politics C... [t:77157954]...</cite> <quote><cite>Crossfiyah posted...</cite> <quote><cite>Corrik posted...</cite> <quote><cite>Crossfiyah posted...</cite> <quote><cite>Corrik posted...</cite> <quote><cite>Crossfiyah posted...</cite> <quote>Clearly because you fundamentally misunderstand probability. A solid seat is a 95% chance of winning or higher. A party will lose let's say 1 in 20 of these. A likely seat is a 75% chance. A party will lose about 1 in 4 or 5 of these. A leaning seat will be lost about 2 in 5 times. Republicans by 538s count has 135 solid seats, 49 likely seats, and 13 leaning seats. They will win an average of 128, 37, and 8 if these if I'm being generous. You are treating these 197 seats as 197 seats won. In reality they are closer to 173 seats won. Take the toss ups and split them in half and you get 182 seats for Republicans not 200 something. Do the same for Democrats now. 193 (!!!) solid seats, 17 likely seats, 10 leaning seats. Translates to 10+4+4=18 pick ups for a total of 200 seats AFTER toss ups are calculated. Not before. You're fundamentally misunderstanding how many safe or safe leaning seats will be lost.</quote> You are assuming probabilities are clear cut answers. Go back to the 2014 midterms. How many solids and likelies and such actually flipped.</quote> Okay bud.</quote> Did 25% of likelies 5% of solids, 40% of leans flip? Was it even close to those numbers?</quote> Odds of winning are not pulled out of a hat they are based on historic outcomes given the lead in polling.</quote> Leans are last second breaks can swing the result. Likelies you are now looking at polling errors. Solids is basically major and drastic polling errors. Tossups is within margin of error and anyone could win.</quote> ... Copied to Clipboard!
xp1337 11/05/18 1:30:27 PM #257:	Fine, no one is saying you have to trust 538, but applying RCP's ratings to 538 when 538 has their own is pointless IMO. Looking at other expert predictors: Cook Political Report has Solid Seats at 182-137. Sabato's Crystal Ball has it at 183-138. Inside Elections has it at 187-159. --- xp1337: Don't you wish there was a spell-checker that told you when you a word out? <cite>xp1337 posted [p:911856369] in Politics C... [t:77157954]...</cite> <quote>Fine, no one is saying you have to trust 538, but applying RCP's ratings to 538 when 538 has their own is pointless IMO. Looking at other expert predictors: Cook Political Report has Solid Seats at 182-137. Sabato's Crystal Ball has it at 183-138. Inside Elections has it at 187-159.</quote> ... Copied to Clipboard!
Crossfiyah 11/05/18 1:30:37 PM #258:	That is not how any of this works. Probability accounts for all of that in a well designed model with good data. Again your entire problem is you are overestimating how sure of a thing a generally red district will be. You are treating all of then as equally likely while some are red by 3 points and some are red by 40 points. --- Jagr_68: "hittes Blake Kesseguin a super saiyan NHL star who gives earth a chance to win the Cell gaemz" <cite>Crossfiyah posted [p:911856381] in Politics C... [t:77157954]...</cite> <quote>That is not how any of this works. Probability accounts for all of that in a well designed model with good data. Again your entire problem is you are overestimating how sure of a thing a generally red district will be. You are treating all of then as equally likely while some are red by 3 points and some are red by 40 points.</quote> ... Copied to Clipboard!
xp1337 11/05/18 1:36:48 PM #259:	Crossfiyah posted... That is not how any of this works. Probability accounts for all of that in a well designed model with good data. The argument you're making is futile. Corrik rejects 538's model, so arguing the probability from 538's standpoint will never go anywhere with him. You don't have to simplify it as "Likely flips x% of the time, Lean y% of the time" because 538's model will do that for you and without those simplifications, using the actual %'s they have calculated. If Corrik was receptive to that argument he'd accept 538's model even if he rejected the underlying seat ratings. --- xp1337: Don't you wish there was a spell-checker that told you when you a word out? <cite>xp1337 posted [p:911856715] in Politics C... [t:77157954]...</cite> <quote><cite>Crossfiyah posted...</cite> <quote>That is not how any of this works. Probability accounts for all of that in a well designed model with good data.</quote> The argument you're making is futile. Corrik rejects 538's model, so arguing the probability from 538's standpoint will never go anywhere with him. You don't have to simplify it as "Likely flips x% of the time, Lean y% of the time" because 538's model will do that for you and without those simplifications, using the actual %'s they have calculated. If Corrik was receptive to that argument he'd accept 538's model even if he rejected the underlying seat ratings.</quote> ... Copied to Clipboard!
Reg 11/05/18 1:37:41 PM #260:	Crossfiyah posted... That is not how any of this works. Somebody who has repeatedly shown a willful ignorance of basic math, probability and statistics has no idea how basic math, probability and statistics work? Coulda fooled me. <cite>Reg posted [p:911856760] in Politics C... [t:77157954]...</cite> <quote><cite>Crossfiyah posted...</cite> <quote>That is not how any of this works.</quote> Somebody who has repeatedly shown a willful ignorance of basic math, probability and statistics has no idea how basic math, probability and statistics work? Coulda fooled me.</quote> ... Copied to Clipboard!
Corrik 11/05/18 1:41:56 PM #261:	Crossfiyah posted... That is not how any of this works. Probability accounts for all of that in a well designed model with good data. Again your entire problem is you are overestimating how sure of a thing a generally red district will be. You are treating all of then as equally likely while some are red by 3 points and some are red by 40 points. I do not think it is a well-designed model and think it shows on how it has had the highest average seat gain of any model I have seen. I think it skews towards democrats too drastically and it screws up the entire model. But, we will see in a day! --- LoL ID = imajericho XBL GT = Corrik <cite>Corrik posted [p:911856976] in Politics C... [t:77157954]...</cite> <quote><cite>Crossfiyah posted...</cite> <quote>That is not how any of this works. Probability accounts for all of that in a well designed model with good data. Again your entire problem is you are overestimating how sure of a thing a generally red district will be. You are treating all of then as equally likely while some are red by 3 points and some are red by 40 points.</quote> I do not think it is a well-designed model and think it shows on how it has had the highest average seat gain of any model I have seen. I think it skews towards democrats too drastically and it screws up the entire model. But, we will see in a day!</quote> ... Copied to Clipboard!
Jakyl25 11/05/18 1:43:14 PM #262:	https://pbs.twimg.com/media/DrP6nTAU0AA7CS1?format=jpg&name=large Some Christians reallllly Love Trump --- Thank you, Eddie Guerrero. https://pbs.twimg.com/media/DRlKR5nU8AA_v_C?format=jpg&name=large <cite>Jakyl25 posted [p:911857056] in Politics C... [t:77157954]...</cite> <quote>https://pbs.twimg.com/media/DrP6nTAU0AA7CS1?format=jpg&name=large Some Christians reallllly Love Trump</quote> ... Copied to Clipboard!
Jakyl25 11/05/18 1:49:08 PM #263:	https://twitter.com/realdonaldtrump/status/1059470847751131138?s=21 Law Enforcement has been strongly notified to watch closely for any ILLEGAL VOTING which may take place in Tuesdays Election (or Early Voting). Anyone caught will be subject to the Maximum Criminal Penalties allowed by law. Thank you! basically pay extra close attention to any Hispanics in line to vote --- Thank you, Eddie Guerrero. https://pbs.twimg.com/media/DRlKR5nU8AA_v_C?format=jpg&name=large <cite>Jakyl25 posted [p:911857377] in Politics C... [t:77157954]...</cite> <quote>https://twitter.com/realdonaldtrump/status/1059470847751131138?s=21 <quote>Law Enforcement has been strongly notified to watch closely for any ILLEGAL VOTING which may take place in Tuesdays Election (or Early Voting). Anyone caught will be subject to the Maximum Criminal Penalties allowed by law. Thank you!</quote> basically pay extra close attention to any Hispanics in line to vote</quote> ... Copied to Clipboard!
xp1337 11/05/18 1:49:40 PM #264:	remember the good ol' days when 538 and Silver were mocked for giving Trump a much better chance than everyone else (something I was guilty of too) --- xp1337: Don't you wish there was a spell-checker that told you when you a word out? <cite>xp1337 posted [p:911857419] in Politics C... [t:77157954]...</cite> <quote>remember the good ol' days when 538 and Silver were mocked for giving Trump a much better chance than everyone else (something I was guilty of too)</quote> ... Copied to Clipboard!
LordoftheMorons 11/05/18 1:52:02 PM #265:	xp1337 posted... remember the good ol' days when 538 and Silver were mocked for giving Trump a much better chance than everyone else And then after the election were totally wrong for saying he had no chance...! (Because people have a two second memory and either round 30% down to 0 or conflate pundit opinion with 538s predictions) --- Congrats to BKSheikah for winning the BYIG Guru Challenge! <cite>LordoftheMorons posted [p:911857532] in Politics C... [t:77157954]...</cite> <quote><cite>xp1337 posted...</cite> <quote>remember the good ol' days when 538 and Silver were mocked for giving Trump a much better chance than everyone else</quote> And then after the election were totally wrong for saying he had no chance...! (Because people have a two second memory and either round 30% down to 0 or conflate pundit opinion with 538s predictions)</quote> ... Copied to Clipboard!
Corrik 11/05/18 1:56:32 PM #266:	xp1337 posted... remember the good ol' days when 538 and Silver were mocked for giving Trump a much better chance than everyone else (something I was guilty of too) Being the closest didn't make it less wrong and significantly in many areas. --- LoL ID = imajericho XBL GT = Corrik <cite>Corrik posted [p:911857749] in Politics C... [t:77157954]...</cite> <quote><cite>xp1337 posted...</cite> <quote>remember the good ol' days when 538 and Silver were mocked for giving Trump a much better chance than everyone else (something I was guilty of too)</quote> Being the closest didn't make it less wrong and significantly in many areas.</quote> ... Copied to Clipboard!
Jakyl25 11/05/18 2:00:04 PM #267:	https://twitter.com/elizlanders/status/1059498040833335296?s=21 Wow- @brianstelter reports: racist Trump campaign ad pulled from Fox News "Upon further review, FOX News pulled the ad yesterday and it will not appear on either FOX News Channel or FOX Business Network," ad sales president Marianne Gambelli told CNN. That Trump ad was too racist for FOX NEWS --- Thank you, Eddie Guerrero. https://pbs.twimg.com/media/DRlKR5nU8AA_v_C?format=jpg&name=large <cite>Jakyl25 posted [p:911857943] in Politics C... [t:77157954]...</cite> <quote>https://twitter.com/elizlanders/status/1059498040833335296?s=21 <quote>Wow- @brianstelter reports: racist Trump campaign ad pulled from Fox News "Upon further review, FOX News pulled the ad yesterday and it will not appear on either FOX News Channel or FOX Business Network," ad sales president Marianne Gambelli told CNN.</quote> That Trump ad was too racist for FOX NEWS</quote> ... Copied to Clipboard!
Jakyl25 11/05/18 2:06:37 PM #268:	https://twitter.com/bubbaprog/status/1059513365570174976?s=21 Trump fires back with his typical mature parry --- Thank you, Eddie Guerrero. https://pbs.twimg.com/media/DRlKR5nU8AA_v_C?format=jpg&name=large <cite>Jakyl25 posted [p:911858327] in Politics C... [t:77157954]...</cite> <quote>https://twitter.com/bubbaprog/status/1059513365570174976?s=21 Trump fires back with his typical mature parry</quote> ... Copied to Clipboard!
Crossfiyah 11/05/18 2:10:09 PM #269:	Corrik posted... Crossfiyah posted... That is not how any of this works. Probability accounts for all of that in a well designed model with good data. Again your entire problem is you are overestimating how sure of a thing a generally red district will be. You are treating all of then as equally likely while some are red by 3 points and some are red by 40 points. I do not think it is a well-designed model and think it shows on how it has had the highest average seat gain of any model I have seen. I think it skews towards democrats too drastically and it screws up the entire model. But, we will see in a day! I mean it doesn't really matter. You are still making the same error regardless. You think all districts have the same chance of going the color they probably will go. --- Jagr_68: "hittes Blake Kesseguin a super saiyan NHL star who gives earth a chance to win the Cell gaemz" <cite>Crossfiyah posted [p:911858519] in Politics C... [t:77157954]...</cite> <quote><cite>Corrik posted...</cite> <quote><cite>Crossfiyah posted...</cite> <quote>That is not how any of this works. Probability accounts for all of that in a well designed model with good data. Again your entire problem is you are overestimating how sure of a thing a generally red district will be. You are treating all of then as equally likely while some are red by 3 points and some are red by 40 points.</quote> I do not think it is a well-designed model and think it shows on how it has had the highest average seat gain of any model I have seen. I think it skews towards democrats too drastically and it screws up the entire model. But, we will see in a day!</quote> I mean it doesn't really matter. You are still making the same error regardless. You think all districts have the same chance of going the color they probably will go.</quote> ... Copied to Clipboard!
Not_an_Owl 11/05/18 2:17:55 PM #270:	Crossfiyah posted... Corrik posted... Crossfiyah posted... That is not how any of this works. Probability accounts for all of that in a well designed model with good data. Again your entire problem is you are overestimating how sure of a thing a generally red district will be. You are treating all of then as equally likely while some are red by 3 points and some are red by 40 points. I do not think it is a well-designed model and think it shows on how it has had the highest average seat gain of any model I have seen. I think it skews towards democrats too drastically and it screws up the entire model. But, we will see in a day! I mean it doesn't really matter. You are still making the same error regardless. You think all districts have the same chance of going the color they probably will go. the probability of any district going democrat is 50/50 either it does or it doesn't checkmate, atheists --- Besides, marijuana is far more harmful than steroids. - BlitzBomb I headbang to Bruckner. <cite>Not_an_Owl posted [p:911858968] in Politics C... [t:77157954]...</cite> <quote><cite>Crossfiyah posted...</cite> <quote><cite>Corrik posted...</cite> <quote><cite>Crossfiyah posted...</cite> <quote>That is not how any of this works. Probability accounts for all of that in a well designed model with good data. Again your entire problem is you are overestimating how sure of a thing a generally red district will be. You are treating all of then as equally likely while some are red by 3 points and some are red by 40 points.</quote> I do not think it is a well-designed model and think it shows on how it has had the highest average seat gain of any model I have seen. I think it skews towards democrats too drastically and it screws up the entire model. But, we will see in a day!</quote> I mean it doesn't really matter. You are still making the same error regardless. You think all districts have the same chance of going the color they probably will go.</quote> the probability of any district going democrat is 50/50 either it does or it doesn't checkmate, atheists</quote> ... Copied to Clipboard!
xp1337 11/05/18 2:19:15 PM #271:	Not_an_Owl posted... Crossfiyah posted... Corrik posted... Crossfiyah posted... That is not how any of this works. Probability accounts for all of that in a well designed model with good data. Again your entire problem is you are overestimating how sure of a thing a generally red district will be. You are treating all of then as equally likely while some are red by 3 points and some are red by 40 points. I do not think it is a well-designed model and think it shows on how it has had the highest average seat gain of any model I have seen. I think it skews towards democrats too drastically and it screws up the entire model. But, we will see in a day! I mean it doesn't really matter. You are still making the same error regardless. You think all districts have the same chance of going the color they probably will go. the probability of any district going democrat is 50/50 either it does or it doesn't checkmate, atheists the probability of any seat going third party is also 50/50 boom no % left for republicans 435-0 house let's go --- xp1337: Don't you wish there was a spell-checker that told you when you a word out? <cite>xp1337 posted [p:911859038] in Politics C... [t:77157954]...</cite> <quote><cite>Not_an_Owl posted...</cite> <quote><cite>Crossfiyah posted...</cite> <quote><cite>Corrik posted...</cite> <quote><cite>Crossfiyah posted...</cite> <quote>That is not how any of this works. Probability accounts for all of that in a well designed model with good data. Again your entire problem is you are overestimating how sure of a thing a generally red district will be. You are treating all of then as equally likely while some are red by 3 points and some are red by 40 points.</quote> I do not think it is a well-designed model and think it shows on how it has had the highest average seat gain of any model I have seen. I think it skews towards democrats too drastically and it screws up the entire model. But, we will see in a day!</quote> I mean it doesn't really matter. You are still making the same error regardless. You think all districts have the same chance of going the color they probably will go.</quote> the probability of any district going democrat is 50/50 either it does or it doesn't checkmate, atheists</quote> the probability of any seat going third party is also 50/50 boom no % left for republicans 435-0 house let's go</quote> ... Copied to Clipboard!
LordoftheMorons 11/05/18 2:25:09 PM #272:	Not_an_Owl posted... Crossfiyah posted... Corrik posted... Crossfiyah posted... That is not how any of this works. Probability accounts for all of that in a well designed model with good data. Again your entire problem is you are overestimating how sure of a thing a generally red district will be. You are treating all of then as equally likely while some are red by 3 points and some are red by 40 points. I do not think it is a well-designed model and think it shows on how it has had the highest average seat gain of any model I have seen. I think it skews towards democrats too drastically and it screws up the entire model. But, we will see in a day! I mean it doesn't really matter. You are still making the same error regardless. You think all districts have the same chance of going the color they probably will go. the probability of any district going democrat is 50/50 either it does or it doesn't checkmate, atheists This reminds me of one of my absolute favorite Daily Show field pieces: http://www.cc.com/video-clips/hzqmb9/the-daily-show-with-jon-stewart-large-hadron-collider --- Congrats to BKSheikah for winning the BYIG Guru Challenge! <cite>LordoftheMorons posted [p:911859378] in Politics C... [t:77157954]...</cite> <quote><cite>Not_an_Owl posted...</cite> <quote><cite>Crossfiyah posted...</cite> <quote><cite>Corrik posted...</cite> <quote><cite>Crossfiyah posted...</cite> <quote>That is not how any of this works. Probability accounts for all of that in a well designed model with good data. Again your entire problem is you are overestimating how sure of a thing a generally red district will be. You are treating all of then as equally likely while some are red by 3 points and some are red by 40 points.</quote> I do not think it is a well-designed model and think it shows on how it has had the highest average seat gain of any model I have seen. I think it skews towards democrats too drastically and it screws up the entire model. But, we will see in a day!</quote> I mean it doesn't really matter. You are still making the same error regardless. You think all districts have the same chance of going the color they probably will go.</quote> the probability of any district going democrat is 50/50 either it does or it doesn't checkmate, atheists</quote> This reminds me of one of my absolute favorite Daily Show field pieces: http://www.cc.com/video-clips/hzqmb9/the-daily-show-with-jon-stewart-large-hadron-collider</quote> ... Copied to Clipboard!
Eddv 11/05/18 2:35:59 PM #273:	This has been a stunning display of Corrik trying desperately to proves he's Just That Dumb vs Always Trolling. It brings a tear to your eye to see such a competitor in action. --- Board 8's Voice of Reason https://imgur.com/chXIw06 <cite>Eddv posted [p:911859939] in Politics C... [t:77157954]...</cite> <quote>This has been a stunning display of Corrik trying desperately to proves he's Just That Dumb vs Always Trolling. It brings a tear to your eye to see such a competitor in action.</quote> ... Copied to Clipboard!
Not_an_Owl 11/05/18 2:38:47 PM #274:	Eddv posted... This has been a stunning display of Corrik trying desperately to proves he's Just That Dumb vs Always Trolling. It brings a tear to your eye to see such a competitor in action. He's gritty, a real gamer. --- Besides, marijuana is far more harmful than steroids. - BlitzBomb I headbang to Bruckner. <cite>Not_an_Owl posted [p:911860070] in Politics C... [t:77157954]...</cite> <quote><cite>Eddv posted...</cite> <quote>This has been a stunning display of Corrik trying desperately to proves he's Just That Dumb vs Always Trolling. It brings a tear to your eye to see such a competitor in action.</quote> He's gritty, a real gamer.</quote> ... Copied to Clipboard!
xp1337 11/05/18 2:42:05 PM #275:	Well, I mean this discussion led me to finding out a bit more about Silver's model and correcting some mistaken assumptions I had about it so I consider it a net win for myself! I don't intend to get too into the weeds on it with Corrik though. Like I said, he should just have said he disagrees with Silver's model instead of trying to take other models and apply their standards to it. That's just dumb. And I also think it's silly for anyone here to try and convince him of Silver's model. If he wants to disagree with it, let him. He doesn't have to believe in its validity and if you do think it's valid then just take his attacks on it into that context. --- xp1337: Don't you wish there was a spell-checker that told you when you a word out? <cite>xp1337 posted [p:911860247] in Politics C... [t:77157954]...</cite> <quote>Well, I mean this discussion led me to finding out a bit more about Silver's model and correcting some mistaken assumptions I had about it so I consider it a net win for myself! I don't intend to get too into the weeds on it with Corrik though. Like I said, he should just have said he disagrees with Silver's model instead of trying to take other models and apply their standards to it. That's just dumb. And I also think it's silly for anyone here to try and convince him of Silver's model. If he wants to disagree with it, let him. He doesn't have to believe in its validity and if you do think it's valid then just take his attacks on it into that context.</quote> ... Copied to Clipboard!
ChaosTonyV4 11/05/18 2:48:45 PM #276:	Corrik posted... xp1337 posted... remember the good ol' days when 538 and Silver were mocked for giving Trump a much better chance than everyone else (something I was guilty of too) Being the closest didn't make it less wrong and significantly in many areas. That isn't how probability works. If someone says there is a 30% chance of something happening, it doesn't make them wrong when it happens, because it will happen 3 out of 10 times. --- Phantom Dust. "I'll just wait for time to prove me right again." - Vlado <cite>ChaosTonyV4 posted [p:911860642] in Politics C... [t:77157954]...</cite> <quote><cite>Corrik posted...</cite> <quote><cite>xp1337 posted...</cite> <quote>remember the good ol' days when 538 and Silver were mocked for giving Trump a much better chance than everyone else (something I was guilty of too)</quote> Being the closest didn't make it less wrong and significantly in many areas.</quote> That isn't how probability works. If someone says there is a 30% chance of something happening, it doesn't make them wrong when it happens, because it will happen 3 out of 10 times.</quote> ... Copied to Clipboard!
Corrik 11/05/18 2:49:30 PM #277:	xp1337 posted... Well, I mean this discussion led me to finding out a bit more about Silver's model and correcting some mistaken assumptions I had about it so I consider it a net win for myself! I don't intend to get too into the weeds on it with Corrik though. Like I said, he should just have said he disagrees with Silver's model instead of trying to take other models and apply their standards to it. That's just dumb. And I also think it's silly for anyone here to try and convince him of Silver's model. If he wants to disagree with it, let him. He doesn't have to believe in its validity and if you do think it's valid then just take his attacks on it into that context. Yeah, I think silvers model has the highest projection of gains due to how it operates. I suspect we don't have previous elections to judge it by, but I do not think it is correct. Maybe it will be the most correct! Guess we will see tomorrow! --- LoL ID = imajericho XBL GT = Corrik <cite>Corrik posted [p:911860698] in Politics C... [t:77157954]...</cite> <quote><cite>xp1337 posted...</cite> <quote>Well, I mean this discussion led me to finding out a bit more about Silver's model and correcting some mistaken assumptions I had about it so I consider it a net win for myself! I don't intend to get too into the weeds on it with Corrik though. Like I said, he should just have said he disagrees with Silver's model instead of trying to take other models and apply their standards to it. That's just dumb. And I also think it's silly for anyone here to try and convince him of Silver's model. If he wants to disagree with it, let him. He doesn't have to believe in its validity and if you do think it's valid then just take his attacks on it into that context.</quote> Yeah, I think silvers model has the highest projection of gains due to how it operates. I suspect we don't have previous elections to judge it by, but I do not think it is correct. Maybe it will be the most correct! Guess we will see tomorrow!</quote> ... Copied to Clipboard!
xp1337 11/05/18 2:53:54 PM #278:	This is the first election cycle that 538 has a House forecast. They've done Senate and Presidential ones in the past. Silver was famously right in 2012, getting every state right while most pundits thought Romney/Rs would do much better/had a chance. (Or did he miss like... NC only which was the closest race?) 2016 goes without saying, though again, he was ahead of most of the pundits and other models who wrote Trump off and again 30% chance events happen well... 30% of the time. It'd be like saying I was wrong if I said you were most likely to get 1 Heads and 1 Tails if you flipped a coin twice and you got 2 Tails. --- xp1337: Don't you wish there was a spell-checker that told you when you a word out? <cite>xp1337 posted [p:911860958] in Politics C... [t:77157954]...</cite> <quote>This is the first election cycle that 538 has a House forecast. They've done Senate and Presidential ones in the past. Silver was famously right in 2012, getting every state right while most pundits thought Romney/Rs would do much better/had a chance. (Or did he miss like... NC only which was the closest race?) 2016 goes without saying, though again, he was ahead of most of the pundits and other models who wrote Trump off and again 30% chance events happen well... 30% of the time. It'd be like saying I was wrong if I said you were most likely to get 1 Heads and 1 Tails if you flipped a coin twice and you got 2 Tails.</quote> ... Copied to Clipboard!
Corrik 11/05/18 2:55:50 PM #279:	ChaosTonyV4 posted... Corrik posted... xp1337 posted... remember the good ol' days when 538 and Silver were mocked for giving Trump a much better chance than everyone else (something I was guilty of too) Being the closest didn't make it less wrong and significantly in many areas. That isn't how probability works. If someone says there is a 30% chance of something happening, it doesn't make them wrong when it happens, because it will happen 3 out of 10 times. That is like saying I ain't wrong if I have it 99% because 1 time out of 100 it happens. If you ran that election exactly the same the outcome would happen the same every time. There is no randomness involved in it. You can't honestly believe if election day ran 10 times you would get 30% of one result and 70% of another result. That "probability" is just a confidence level. He was closest. But still far off. I mean, the actual outcome electorally was probably actual close to 1% territory tbqh. He was close on the popular vote. Which, again I think the problem is that he is going to attribute the popular vote to the entire country instead of accounting for just the further deepening of already deep areas. And closing of gaps that don't flip results. --- LoL ID = imajericho XBL GT = Corrik <cite>Corrik posted [p:911861095] in Politics C... [t:77157954]...</cite> <quote><cite>ChaosTonyV4 posted...</cite> <quote><cite>Corrik posted...</cite> <quote><cite>xp1337 posted...</cite> <quote>remember the good ol' days when 538 and Silver were mocked for giving Trump a much better chance than everyone else (something I was guilty of too)</quote> Being the closest didn't make it less wrong and significantly in many areas.</quote> That isn't how probability works. If someone says there is a 30% chance of something happening, it doesn't make them wrong when it happens, because it will happen 3 out of 10 times.</quote> That is like saying I ain't wrong if I have it 99% because 1 time out of 100 it happens. If you ran that election exactly the same the outcome would happen the same every time. There is no randomness involved in it. You can't honestly believe if election day ran 10 times you would get 30% of one result and 70% of another result. That "probability" is just a confidence level. He was closest. But still far off. I mean, the actual outcome electorally was probably actual close to 1% territory tbqh. He was close on the popular vote. Which, again I think the problem is that he is going to attribute the popular vote to the entire country instead of accounting for just the further deepening of already deep areas. And closing of gaps that don't flip results.</quote> ... Copied to Clipboard!
NFUN 11/05/18 2:59:07 PM #280:	not even the field of psychology says that a 30% chance means that it can't happen your standards are lower than psychology's, corrik. let that sink in --- Life before death. Strength before weakness. Journey before destination. <cite>NFUN posted [p:911861294] in Politics C... [t:77157954]...</cite> <quote>not even the field of psychology says that a 30% chance means that it can't happen your standards are lower than psychology's, corrik. let that sink in</quote> ... Copied to Clipboard!
LordoftheMorons 11/05/18 3:12:57 PM #281:	Theres a ton of uncertainty (enough to give Trump a 30% chance of winning) primarily because no one has perfect information. Of course if you duplicate the world at 6am on Election Day and run things 1000 times youre not gonna see Clinton win it 700 (though maybe she wins it in one or two universes due to, like, the Russia story breaking that morning or something). The sense in which we can see if the probabilities are well callibrated is to average over many elections. Notably, if Nate calls 1000 matches as being 70/30 matches and the favorite wins in 900 of them, hes just as wrong as if the favorite had won in 500 of them. You cant do this across just one election though because the errors are, to a large extent, correlated (i.e. if the polls underestimated Trump in Iowa they probably also underestimated him in Ohio). --- Congrats to BKSheikah for winning the BYIG Guru Challenge! <cite>LordoftheMorons posted [p:911862117] in Politics C... [t:77157954]...</cite> <quote>Theres a ton of uncertainty (enough to give Trump a 30% chance of winning) primarily because no one has perfect information. Of course if you duplicate the world at 6am on Election Day and run things 1000 times youre not gonna see Clinton win it 700 (though maybe she wins it in one or two universes due to, like, the Russia story breaking that morning or something). The sense in which we can see if the probabilities are well callibrated is to average over many elections. Notably, if Nate calls 1000 matches as being 70/30 matches and the favorite wins in 900 of them, hes just as wrong as if the favorite had won in 500 of them. You cant do this across just one election though because the errors are, to a large extent, correlated (i.e. if the polls underestimated Trump in Iowa they probably also underestimated him in Ohio).</quote> ... Copied to Clipboard!
Corrik 11/05/18 3:30:48 PM #282:	https://www.foxnews.com/politics/senate-report-reveals-mans-encounter-with-ford-that-fits-her-allegations-against-kavanaugh Weird --- LoL ID = imajericho XBL GT = Corrik <cite>Corrik posted [p:911863188] in Politics C... [t:77157954]...</cite> <quote>https://www.foxnews.com/politics/senate-report-reveals-mans-encounter-with-ford-that-fits-her-allegations-against-kavanaugh Weird</quote> ... Copied to Clipboard!
TotallyNotMI 11/05/18 3:43:42 PM #283:	Peace___Frog posted... Lmao I'm not going to try to explain math to corrik for the umpteenth time This line gave me so much hope for the day. --- I'm not sure who this MI guy is but he sounds sexy. <cite>TotallyNotMI posted [p:911863975] in Politics C... [t:77157954]...</cite> <quote><cite>Peace___Frog posted...</cite> <quote>Lmao I'm not going to try to explain math to corrik for the umpteenth time</quote> This line gave me so much hope for the day.</quote> ... Copied to Clipboard!
Eddv 11/05/18 4:03:02 PM #284:	McBath may actually win in the GA-06, which would basically be hilarious. Ossoff gets all this attention for being the "perfect" dem to win that district as a moderate white male seemingly tailored to appeal to white suburban men and loses. McBath is a black woman running on a much more liberal , gun control centered platform and seems to be about to pull off the upset --- Board 8's Voice of Reason https://imgur.com/chXIw06 <cite>Eddv posted [p:911865125] in Politics C... [t:77157954]...</cite> <quote>McBath may actually win in the GA-06, which would basically be hilarious. Ossoff gets all this attention for being the "perfect" dem to win that district as a moderate white male seemingly tailored to appeal to white suburban men and loses. McBath is a black woman running on a much more liberal , gun control centered platform and seems to be about to pull off the upset</quote> ... Copied to Clipboard!
TheRock1525 11/05/18 4:10:08 PM #285:	Seriously Corrik must get super confused by baseball. "It said there was only a 30% chance he'd get a hit, and yet he did. The numbers were wroooooong!" --- TheRock ~ I had a name, my father called me Blues. <cite>TheRock1525 posted [p:911865566] in Politics C... [t:77157954]...</cite> <quote>Seriously Corrik must get super confused by baseball. "It said there was only a 30% chance he'd get a hit, and yet he did. The numbers were wroooooong!"</quote> ... Copied to Clipboard!
Crossfiyah 11/05/18 4:12:49 PM #286:	LordoftheMorons posted... Theres a ton of uncertainty (enough to give Trump a 30% chance of winning) primarily because no one has perfect information. Of course if you duplicate the world at 6am on Election Day and run things 1000 times youre not gonna see Clinton win it 700 (though maybe she wins it in one or two universes due to, like, the Russia story breaking that morning or something). The sense in which we can see if the probabilities are well callibrated is to average over many elections. Notably, if Nate calls 1000 matches as being 70/30 matches and the favorite wins in 900 of them, hes just as wrong as if the favorite had won in 500 of them. You cant do this across just one election though because the errors are, to a large extent, correlated (i.e. if the polls underestimated Trump in Iowa they probably also underestimated him in Ohio). Nah if you run that election even a week earlier Clinton wins it's handily. The Comey letter was a 3 point slide that we know about. Extrapolate it and it's probably more. Also there was a huge amount of undecided voters at the time. Way more than historically. To top it off the electoral college exists and 538 actually said it was way more likely Trump would win the presidency and lose the popular vote than win both. None of those things are true this time. --- Jagr_68: "hittes Blake Kesseguin a super saiyan NHL star who gives earth a chance to win the Cell gaemz" <cite>Crossfiyah posted [p:911865746] in Politics C... [t:77157954]...</cite> <quote><cite>LordoftheMorons posted...</cite> <quote>Theres a ton of uncertainty (enough to give Trump a 30% chance of winning) primarily because no one has perfect information. Of course if you duplicate the world at 6am on Election Day and run things 1000 times youre not gonna see Clinton win it 700 (though maybe she wins it in one or two universes due to, like, the Russia story breaking that morning or something). The sense in which we can see if the probabilities are well callibrated is to average over many elections. Notably, if Nate calls 1000 matches as being 70/30 matches and the favorite wins in 900 of them, hes just as wrong as if the favorite had won in 500 of them. You cant do this across just one election though because the errors are, to a large extent, correlated (i.e. if the polls underestimated Trump in Iowa they probably also underestimated him in Ohio).</quote> Nah if you run that election even a week earlier Clinton wins it's handily. The Comey letter was a 3 point slide that we know about. Extrapolate it and it's probably more. Also there was a huge amount of undecided voters at the time. Way more than historically. To top it off the electoral college exists and 538 actually said it was way more likely Trump would win the presidency and lose the popular vote than win both. None of those things are true this time.</quote> ... Copied to Clipboard!
LordoftheMorons 11/05/18 4:19:16 PM #287:	Crossfiyah posted... Nah if you run that election even a week earlier Clinton wins it's handily. The Comey letter was a 3 point slide that we know about. Extrapolate it and it's probably more. Also there was a huge amount of undecided voters at the time. Way more than historically. To top it off the electoral college exists and 538 actually said it was way more likely Trump would win the presidency and lose the popular vote than win both. None of those things are true this time. I'm not saying anything about this time (and I agree that the Comey letter almost certainly swung the election). I'm just making a statement about whether the source of the uncertainty in these models is due to incomplete info, statistical uncertainty, or the result just being fundamentally uncertain. --- Congrats to BKSheikah for winning the BYIG Guru Challenge! <cite>LordoftheMorons posted [p:911866131] in Politics C... [t:77157954]...</cite> <quote><cite>Crossfiyah posted...</cite> <quote>Nah if you run that election even a week earlier Clinton wins it's handily. The Comey letter was a 3 point slide that we know about. Extrapolate it and it's probably more. Also there was a huge amount of undecided voters at the time. Way more than historically. To top it off the electoral college exists and 538 actually said it was way more likely Trump would win the presidency and lose the popular vote than win both. None of those things are true this time.</quote> I'm not saying anything about this time (and I agree that the Comey letter almost certainly swung the election). I'm just making a statement about whether the source of the uncertainty in these models is due to incomplete info, statistical uncertainty, or the result just being fundamentally uncertain.</quote> ... Copied to Clipboard!
red sox 777 11/05/18 4:28:16 PM #288:	538 is really great, although I like using the polls-only models. The other stuff is speculation and I don't think it's appreciably better than yours or mine. With 2016, I took the 538 model, made one adjustment for a systemic polling error and got 48 states right (and was less than 0.5% off in the last 2 states). --- September 1, 2003; November 4, 2007; September 2, 2013 Congratulations to DP Oblivion in the Guru Contest! <cite>red sox 777 posted [p:911866715] in Politics C... [t:77157954]...</cite> <quote>538 is really great, although I like using the polls-only models. The other stuff is speculation and I don't think it's appreciably better than yours or mine. With 2016, I took the 538 model, made one adjustment for a systemic polling error and got 48 states right (and was less than 0.5% off in the last 2 states).</quote> ... Copied to Clipboard!
red sox 777 11/05/18 4:29:42 PM #289:	LordoftheMorons posted... Crossfiyah posted... Nah if you run that election even a week earlier Clinton wins it's handily. The Comey letter was a 3 point slide that we know about. Extrapolate it and it's probably more. Also there was a huge amount of undecided voters at the time. Way more than historically. To top it off the electoral college exists and 538 actually said it was way more likely Trump would win the presidency and lose the popular vote than win both. None of those things are true this time. I'm not saying anything about this time (and I agree that the Comey letter almost certainly swung the election). I'm just making a statement about whether the source of the uncertainty in these models is due to incomplete info, statistical uncertainty, or the result just being fundamentally uncertain. All of those things are the same thing.....the model is not a perfect representation of reality and does not claim to be. --- September 1, 2003; November 4, 2007; September 2, 2013 Congratulations to DP Oblivion in the Guru Contest! <cite>red sox 777 posted [p:911866789] in Politics C... [t:77157954]...</cite> <quote><cite>LordoftheMorons posted...</cite> <quote><cite>Crossfiyah posted...</cite> <quote>Nah if you run that election even a week earlier Clinton wins it's handily. The Comey letter was a 3 point slide that we know about. Extrapolate it and it's probably more. Also there was a huge amount of undecided voters at the time. Way more than historically. To top it off the electoral college exists and 538 actually said it was way more likely Trump would win the presidency and lose the popular vote than win both. None of those things are true this time.</quote> I'm not saying anything about this time (and I agree that the Comey letter almost certainly swung the election). I'm just making a statement about whether the source of the uncertainty in these models is due to incomplete info, statistical uncertainty, or the result just being fundamentally uncertain.</quote> All of those things are the same thing.....the model is not a perfect representation of reality and does not claim to be.</quote> ... Copied to Clipboard!
red sox 777 11/05/18 4:35:33 PM #290:	As for Comey, I doubt it mattered. What was important was the large number of undecided voters, many of whom were not really undecided. They didn't like Trump and wanted to call themselves undecided, but it was super predictable they would vote for him in the end. Many of them were evangelicals, who ended up breaking 83-17 for Trump or thereabouts. That's not something 538 can include in their polling model, but it was something an observer could foresee. --- September 1, 2003; November 4, 2007; September 2, 2013 Congratulations to DP Oblivion in the Guru Contest! <cite>red sox 777 posted [p:911867150] in Politics C... [t:77157954]...</cite> <quote>As for Comey, I doubt it mattered. What was important was the large number of undecided voters, many of whom were not really undecided. They didn't like Trump and wanted to call themselves undecided, but it was super predictable they would vote for him in the end. Many of them were evangelicals, who ended up breaking 83-17 for Trump or thereabouts. That's not something 538 can include in their polling model, but it was something an observer could foresee.</quote> ... Copied to Clipboard!
red sox 777 11/05/18 4:39:03 PM #291:	Corrik posted... ChaosTonyV4 posted... Corrik posted... xp1337 posted... remember the good ol' days when 538 and Silver were mocked for giving Trump a much better chance than everyone else (something I was guilty of too) Being the closest didn't make it less wrong and significantly in many areas. That isn't how probability works. If someone says there is a 30% chance of something happening, it doesn't make them wrong when it happens, because it will happen 3 out of 10 times. That is like saying I ain't wrong if I have it 99% because 1 time out of 100 it happens. If you ran that election exactly the same the outcome would happen the same every time. There is no randomness involved in it. You can't honestly believe if election day ran 10 times you would get 30% of one result and 70% of another result. That "probability" is just a confidence level. He was closest. But still far off. I mean, the actual outcome electorally was probably actual close to 1% territory tbqh. He was close on the popular vote. Which, again I think the problem is that he is going to attribute the popular vote to the entire country instead of accounting for just the further deepening of already deep areas. And closing of gaps that don't flip results. That's not what the probability is of. Once you have more information (like the actual results) of course the probability changes. Rerun the election in alternate universes and Hillary probably does not win 1 in 10,000. But that's because we know the result in our universe already. --- September 1, 2003; November 4, 2007; September 2, 2013 Congratulations to DP Oblivion in the Guru Contest! <cite>red sox 777 posted [p:911867377] in Politics C... [t:77157954]...</cite> <quote><cite>Corrik posted...</cite> <quote><cite>ChaosTonyV4 posted...</cite> <quote><cite>Corrik posted...</cite> <quote><cite>xp1337 posted...</cite> <quote>remember the good ol' days when 538 and Silver were mocked for giving Trump a much better chance than everyone else (something I was guilty of too)</quote> Being the closest didn't make it less wrong and significantly in many areas.</quote> That isn't how probability works. If someone says there is a 30% chance of something happening, it doesn't make them wrong when it happens, because it will happen 3 out of 10 times.</quote> That is like saying I ain't wrong if I have it 99% because 1 time out of 100 it happens. If you ran that election exactly the same the outcome would happen the same every time. There is no randomness involved in it. You can't honestly believe if election day ran 10 times you would get 30% of one result and 70% of another result. That "probability" is just a confidence level. He was closest. But still far off. I mean, the actual outcome electorally was probably actual close to 1% territory tbqh. He was close on the popular vote. Which, again I think the problem is that he is going to attribute the popular vote to the entire country instead of accounting for just the further deepening of already deep areas. And closing of gaps that don't flip results.</quote> That's not what the probability is of. Once you have more information (like the actual results) of course the probability changes. Rerun the election in alternate universes and Hillary probably does not win 1 in 10,000. But that's because we know the result in our universe already.</quote> ... Copied to Clipboard!
LordoftheMorons 11/05/18 4:49:38 PM #292:	red sox 777 posted... All of those things are the same thing.....the model is not a perfect representation of reality and does not claim to be. They're not the same thing on a fundamental level. It's true that the model has some total uncertainty, and that if I'm making a particular prediction I might not care where the uncertainty is coming from. But to understand what's going on we can distinguish between several types of uncertainty, e.g.: 1) Statistical uncertainty: maybe I really do have a perfect model for the underlying distribution of voters, but with a limited amount of data I'm fundamentally going to get some statistical noise. 2) Systematic errors: if I do something like screw up my likely voter model and think that youth turnout is going to be 20% instead of 30% when I'm weighting things I'm going to screw up (this is one level removed from 538 since it's a pollster error, but these mistakes are likely to be correlated among pollsters) 3) Time based errors: the final polls have pretty much been conducted already. If undecideds break differently than expected, that will change the results (or maybe a huge news story changes voter preferences in the last few days) 4) Fundamental uncertainty: uncertainty in the same sense as quantum uncertainty, e.g. even with perfect information you could not predict the outcome. This source shouldn't exist for elections, but we could philosophically consider it when we're asking about why the polls or models were wrong. --- Congrats to BKSheikah for winning the BYIG Guru Challenge! <cite>LordoftheMorons posted [p:911868028] in Politics C... [t:77157954]...</cite> <quote><cite>red sox 777 posted...</cite> <quote>All of those things are the same thing.....the model is not a perfect representation of reality and does not claim to be.</quote> They're not the same thing on a fundamental level. It's true that the model has some total uncertainty, and that if I'm making a particular prediction I might not care where the uncertainty is coming from. But to understand what's going on we can distinguish between several types of uncertainty, e.g.: 1) Statistical uncertainty: maybe I really do have a perfect model for the underlying distribution of voters, but with a limited amount of data I'm fundamentally going to get some statistical noise. 2) Systematic errors: if I do something like screw up my likely voter model and think that youth turnout is going to be 20% instead of 30% when I'm weighting things I'm going to screw up (this is one level removed from 538 since it's a pollster error, but these mistakes are likely to be correlated among pollsters) 3) Time based errors: the final polls have pretty much been conducted already. If undecideds break differently than expected, that will change the results (or maybe a huge news story changes voter preferences in the last few days) 4) Fundamental uncertainty: uncertainty in the same sense as quantum uncertainty, e.g. even with perfect information you could not predict the outcome. This source shouldn't exist for elections, but we could philosophically consider it when we're asking about why the polls or models were wrong.</quote> ... Copied to Clipboard!
SupremeZero 11/05/18 4:54:50 PM #293:	Crossfiyah posted... LordoftheMorons posted... Theres a ton of uncertainty (enough to give Trump a 30% chance of winning) primarily because no one has perfect information. Of course if you duplicate the world at 6am on Election Day and run things 1000 times youre not gonna see Clinton win it 700 (though maybe she wins it in one or two universes due to, like, the Russia story breaking that morning or something). The sense in which we can see if the probabilities are well callibrated is to average over many elections. Notably, if Nate calls 1000 matches as being 70/30 matches and the favorite wins in 900 of them, hes just as wrong as if the favorite had won in 500 of them. You cant do this across just one election though because the errors are, to a large extent, correlated (i.e. if the polls underestimated Trump in Iowa they probably also underestimated him in Ohio). Nah if you run that election even a week earlier Clinton wins it's handily. The Comey letter was a 3 point slide that we know about. Extrapolate it and it's probably more. Also there was a huge amount of undecided voters at the time. Way more than historically. To top it off the electoral college exists and 538 actually said it was way more likely Trump would win the presidency and lose the popular vote than win both. None of those things are true this time. To be fair there's like a billion individual things that swung the election. Take it back ANOTHER week and it would've been the reverse. --- There's always hope for better things in life. But you can't let anything, friend, lover,God himself,be your hope. You have to be your own hope <cite>SupremeZero posted [p:911868362] in Politics C... [t:77157954]...</cite> <quote><cite>Crossfiyah posted...</cite> <quote><cite>LordoftheMorons posted...</cite> <quote>Theres a ton of uncertainty (enough to give Trump a 30% chance of winning) primarily because no one has perfect information. Of course if you duplicate the world at 6am on Election Day and run things 1000 times youre not gonna see Clinton win it 700 (though maybe she wins it in one or two universes due to, like, the Russia story breaking that morning or something). The sense in which we can see if the probabilities are well callibrated is to average over many elections. Notably, if Nate calls 1000 matches as being 70/30 matches and the favorite wins in 900 of them, hes just as wrong as if the favorite had won in 500 of them. You cant do this across just one election though because the errors are, to a large extent, correlated (i.e. if the polls underestimated Trump in Iowa they probably also underestimated him in Ohio).</quote> Nah if you run that election even a week earlier Clinton wins it's handily. The Comey letter was a 3 point slide that we know about. Extrapolate it and it's probably more. Also there was a huge amount of undecided voters at the time. Way more than historically. To top it off the electoral college exists and 538 actually said it was way more likely Trump would win the presidency and lose the popular vote than win both. None of those things are true this time.</quote> To be fair there's like a billion individual things that swung the election. Take it back ANOTHER week and it would've been the reverse.</quote> ... Copied to Clipboard!
Eddv 11/05/18 5:00:45 PM #294:	Baseball really is a great sport. In no other sport is the most likely outcome of each individual "play" or "at bat" to be "basically nothing". Neither a Strikeout nor a hit is the most likely result. Really Corrik must hate Fire Emblem too. "What the fuck it said they only had a 3% crit chance why did I get crit?" --- Board 8's Voice of Reason https://imgur.com/chXIw06 <cite>Eddv posted [p:911868764] in Politics C... [t:77157954]...</cite> <quote>Baseball really is a great sport. In no other sport is the most likely outcome of each individual "play" or "at bat" to be "basically nothing". Neither a Strikeout nor a hit is the most likely result. Really Corrik must hate Fire Emblem too. "What the fuck it said they only had a 3% crit chance why did I get crit?"</quote> ... Copied to Clipboard!
red sox 777 11/05/18 5:01:12 PM #295:	I'm not sure that 1-3 are distinguishable in a coherent, meaningful, way. I don't think we need worry about quantum uncertainty unless our elections are much much closer than 2016. --- September 1, 2003; November 4, 2007; September 2, 2013 Congratulations to DP Oblivion in the Guru Contest! <cite>red sox 777 posted [p:911868789] in Politics C... [t:77157954]...</cite> <quote>I'm not sure that 1-3 are distinguishable in a coherent, meaningful, way. I don't think we need worry about quantum uncertainty unless our elections are much much closer than 2016.</quote> ... Copied to Clipboard!
NFUN 11/05/18 5:14:57 PM #296:	Eddv posted... Really Corrik must hate Fire Emblem too. "What the fuck it said they only had a 3% crit chance why did I get crit?" tbf everybody yells this whenever they get hit by a crit --- Life before death. Strength before weakness. Journey before destination. <cite>NFUN posted [p:911869583] in Politics C... [t:77157954]...</cite> <quote><cite>Eddv posted...</cite> <quote>Really Corrik must hate Fire Emblem too. "What the fuck it said they only had a 3% crit chance why did I get crit?"</quote> tbf everybody yells this whenever they get hit by a crit</quote> ... Copied to Clipboard!
SupremeZero 11/05/18 5:21:44 PM #297:	http://photobucket.com/gallery/http://s49.photobucket.com/user/katietiedrich/media/comic259.png.html --- There's always hope for better things in life. But you can't let anything, friend, lover,God himself,be your hope. You have to be your own hope <cite>SupremeZero posted [p:911869983] in Politics C... [t:77157954]...</cite> <quote>http://photobucket.com/gallery/http://s49.photobucket.com/user/katietiedrich/media/comic259.png.html</quote> ... Copied to Clipboard!
hockeydude15 11/05/18 5:26:21 PM #298:	NFUN posted... Eddv posted... Really Corrik must hate Fire Emblem too. "What the fuck it said they only had a 3% crit chance why did I get crit?" tbf everybody yells this whenever they get hit by a crit To me it's mostly that the 3% crit chance is accompanied by a "12%" chance to hit which the way FE does hit chance is really like 1% so that crit is soooo unlikely to hit. --- Yawn <cite>hockeydude15 posted [p:911870274] in Politics C... [t:77157954]...</cite> <quote><cite>NFUN posted...</cite> <quote><cite>Eddv posted...</cite> <quote>Really Corrik must hate Fire Emblem too. "What the fuck it said they only had a 3% crit chance why did I get crit?"</quote> tbf everybody yells this whenever they get hit by a crit</quote> To me it's mostly that the 3% crit chance is accompanied by a "12%" chance to hit which the way FE does hit chance is really like 1% so that crit is soooo unlikely to hit.</quote> ... Copied to Clipboard!
Jakyl25 11/05/18 5:34:00 PM #299:	SupremeZero posted... To be fair there's like a billion individual things that swung the election. Each individual email --- Thank you, Eddie Guerrero. https://pbs.twimg.com/media/DRlKR5nU8AA_v_C?format=jpg&name=large <cite>Jakyl25 posted [p:911870694] in Politics C... [t:77157954]...</cite> <quote><cite>SupremeZero posted...</cite> <quote> To be fair there's like a billion individual things that swung the election.</quote> Each individual email</quote> ... Copied to Clipboard!
Corrik 11/05/18 5:40:51 PM #300:	red sox 777 posted... Corrik posted... ChaosTonyV4 posted... Corrik posted... xp1337 posted... remember the good ol' days when 538 and Silver were mocked for giving Trump a much better chance than everyone else (something I was guilty of too) Being the closest didn't make it less wrong and significantly in many areas. That isn't how probability works. If someone says there is a 30% chance of something happening, it doesn't make them wrong when it happens, because it will happen 3 out of 10 times. That is like saying I ain't wrong if I have it 99% because 1 time out of 100 it happens. If you ran that election exactly the same the outcome would happen the same every time. There is no randomness involved in it. You can't honestly believe if election day ran 10 times you would get 30% of one result and 70% of another result. That "probability" is just a confidence level. He was closest. But still far off. I mean, the actual outcome electorally was probably actual close to 1% territory tbqh. He was close on the popular vote. Which, again I think the problem is that he is going to attribute the popular vote to the entire country instead of accounting for just the further deepening of already deep areas. And closing of gaps that don't flip results. That's not what the probability is of. Once you have more information (like the actual results) of course the probability changes. Rerun the election in alternate universes and Hillary probably does not win 1 in 10,000. But that's because we know the result in our universe already. Like I said. The probability is just a confidence level in the data presented. It isn't an actual probability. --- LoL ID = imajericho XBL GT = Corrik <cite>Corrik posted [p:911871118] in Politics C... [t:77157954]...</cite> <quote><cite>red sox 777 posted...</cite> <quote><cite>Corrik posted...</cite> <quote><cite>ChaosTonyV4 posted...</cite> <quote><cite>Corrik posted...</cite> <quote><cite>xp1337 posted...</cite> <quote>remember the good ol' days when 538 and Silver were mocked for giving Trump a much better chance than everyone else (something I was guilty of too)</quote> Being the closest didn't make it less wrong and significantly in many areas.</quote> That isn't how probability works. If someone says there is a 30% chance of something happening, it doesn't make them wrong when it happens, because it will happen 3 out of 10 times.</quote> That is like saying I ain't wrong if I have it 99% because 1 time out of 100 it happens. If you ran that election exactly the same the outcome would happen the same every time. There is no randomness involved in it. You can't honestly believe if election day ran 10 times you would get 30% of one result and 70% of another result. That "probability" is just a confidence level. He was closest. But still far off. I mean, the actual outcome electorally was probably actual close to 1% territory tbqh. He was close on the popular vote. Which, again I think the problem is that he is going to attribute the popular vote to the entire country instead of accounting for just the further deepening of already deep areas. And closing of gaps that don't flip results.</quote> That's not what the probability is of. Once you have more information (like the actual results) of course the probability changes. Rerun the election in alternate universes and Hillary probably does not win 1 in 10,000. But that's because we know the result in our universe already.</quote> Like I said. The probability is just a confidence level in the data presented. It isn't an actual probability.</quote> ... Copied to Clipboard!
Topic List	Page List: 1 ... 2, 3, 4, 5, 6, 7, 8, 9, 10

Board 8 > Politics Containment Topic 203: We Met at Borders (+) Yeah! (+)

Board 8 > Politics Containment Topic 203: We Met at Borders