Monday, November 5, 2012

Predictions

I love to see people offer predictions.  It's like science.  You know your stuff if your predictions are right.  So let's just note the predictions being made here now prior to the election.

Dick Morris says it's a Romney landslide.  Newt Gingrich also expects a Romney landslide.  Fred Barnes thinks it's Romney.  Karl Rove seems to think Romney will win comfortably.  Peggy Noonan tells us Romney will win.  Even Michael Barone is telling us Romney wins big.

All the betting markets I know of expect Obama to win.  Nate Silver has Obama pretty comfortably ahead.  The more right wing Real Clear Politics has a slight lead for Obama.

For my part, I think it's safe to say that Michigan is out of play, so I'm thinking I'll just vote for Jill Stein.  If Michigan were in play I'd have a tough decision to make.  Obama is functionally a dictator, killing whoever he wants without oversight.  Muslims are the new blacks.  They get killed and oppressed by this administration and so many people just couldn't care less.  Even people I know that regard themselves as liberal are happy with Obama's due process free assassination program.  Obama is imprisoning whistle blowers and all the while those that commit crimes that actually harm others go free, like the people that wrecked our economy in 2008 or super wealthy tax dodgers.  He's a nightmare.  But Romney is all that and more.  More being he wants to implemented the failed economic policies of Spain, Ireland, Greece, Haiti, and all the African countries, which are quite good at making the super rich even more rich and they are an extreme hardship for the poor.

The bright side of a Romney win might be that liberals would care about civil liberties again.  If Romney enacts his policies he'll probably slow our economy down and that will be hard for a lot of people.  But on the plus side it would mean less greenhouse gases.  Africans, which are subject to Republican economic policies, don't produce a lot of greenhouse gases.  They also are starving though and the destruction done to their environment by corporations free of government regulation is a horror.

Another bright side of a Romney win is you can make the case that Obama really wants to pursue a right wing austerity route just as he's pursued right wing violence overseas.  When Obama advocates cutting social services he doesn't have to deal with liberal voices objecting, whereas Romney would.  You can make the case that Romney is less likely to succeed implementing a cruel austerity package and so Romney is actually better for the poor.

And they call this a choice.  Like a lot of people I'll just be glad when it's over.  I wasn't planning to watch election coverage tomorrow just because it's frustrating to watch this whole sham play out like voters are given a real choice.  But I learned that Glenn Greenwald will be on all day with Al Jazeera.  That should provide some interesting perspective.

Update: Krugman's "informed guess" is that Obama will win. George Will has a big win for Romney.  Check their track records here.

12 comments:

Jonathan said...

Personally, I was actually pretty close to voting Independent this morning, was really dissuaded against Romney due to Bain and Global warming, but the bird in hand vs two in the bush logic got me, and I ultimately went Romney b/c I think there's a decent chance Roe V. Wade might eventually get overturned with a Republican in office the next 4 years with a possible judge retiring.

At least after tonight I can get away from rhetoric and focus back on shipping all those jobs oversees with my VA outsourcing business as some see my entrepreneurial endeavors ;-)...

I like the NYT Monte Carlo analysis, and I've also been tracking intrade, which was particularly interesting during the debates.

http://www.intrade.com/v4/misc/scoreboard/

http://fivethirtyeight.blogs.nytimes.com/

My prediction: VA will fall early to Obama, all but sealing the deal especially after PA is quickly called, FL will drag on, but if it gets called early, it's lights out for Romney. By 8:35 CST I'm saying it will be called.

HispanicPundit said...

You really have to factor in bias when you judge these predictions.

Both sides have a vested interest in saying their candidate is going to win. Think about it this way: if you are sitting at home thinking about voting, you are less likely to vote if you hear that even people on your side are predicting the other side is going to win. This is so especially in close elections, where polling lines are long, weather factors are an issue, etc make for a more uncomfortable voting process.

So of course Krugman is going to say he thinks Obama will win. Karl Rove and GOP operatives are going to say Romney is going to win. etc.

What's impressive is when you have no political bias or predict against your political bias. In which case Nate Silver, Real Clear Politics, and even one of my favorite columnists, Ross Douthat deserve kudos.

Let's not forget, that even Paul Krugman has shown to be a horrible predictor when his political interests are involved, see here.

Jon said...

HP, I followed your links and I actually can't find any prediction that was made by Krugman which turned out to be wrong. What exactly did he predict would happen?

The way I see it being wrong is more telling than being right. Dick Morris was wrong. We know that he failed badly. But what would have happened if Romney had won? We wouldn't have known that Morris had failed, but we wouldn't have known that he used the data correctly either. In this case being right doesn't prove your method right. A person could have just guessed and been right. Being wrong however does show your method to be wrong.

So take the so called case of Mankiw v Krugman. Let's say Mankiw made a prediction about growth rate and it turned out to be true. That doesn't mean his method is right. Maybe he's just a conservative that took a stab and said growth rates would be poor because he wanted Obama do do badly. We don't know that his methods are bad, but we don't know that they are good either. He could select a growth rate and said it will be below this value, but anybody flipping a coin could do the same. If he's wrong, then clearly his methods are bad, but if he's right this doesn't prove his methods are good.

As far as I can see Krugman made no claim about what GDP growth rates would be. Maybe he didn't take the bet because he thought that this is not something that can be known. This doesn't make him a bad predictor. Had he taken the bet and lost, then you'd have evidence. If you make a mistaken prediction this shows you have bad methods, but where is his prediction?

Think of it this way. Suppose you say there will be very little rain this year and I say your methods are bad. So you say "I'll bet we get less than 5 inches." I decline to bet because while I think your methods are bad I also recognize that there are many other factors that can effect the amount of rain. You just go forward with your bet and I don't participate, and it turns out you are right. So now I'm a "horrible predictor"? No. I'm a guy that thinks your methods are bad, but I also don't pretend that I know exactly how much rain we can get since there are many factors involved. I'm not proved to be wrong. I may be wrong, but this wouldn't show it.

HispanicPundit said...

Start here. This is the scenario: The White House is predicting rosier than expected growth forecasts. Mankiw tramples them. Says their reasoning is bad. Krugman disagreed. Even went so far as to suggest the White House predictions have merit (read his last paragraph here). Mankiw calls him to task on this, and even offers up a bet.

What turned out to be the case? The White House predictions turned out to be FAR too optimistic. Just as Mankiw had argued.

Curious, in this case, what does Krugman's failure tell us about his prediction ability? :-)

Jon said...

I really don't see a prediction from Krugman at all, let alone a failed prediction. Something about the rate of growth after slack capacity is taken up. Has slack capacity been taken up? I don't think it has (unemployment still pretty high). So it says the conditions Krugman is suggesting for when rapid growth would occur hadn't even happened yet. I could be missing it here but I don't want to dig into it to find a possible failed prediction. If you say he made a failed prediction tell us what his prediction was and how it failed. It's not my job to go find one, starting at one website, proceeding to another. I have Dick Morris specific predictions right here, just telling you what they are. I don't just tell you that Morris made some prediction about the election in very vague terms and ask you to go figure out what it was. If you don't want to show us, fine, but I'm not going to do the work for you.

HispanicPundit said...

You have a very strong affirmation bias Jon, I probably won't convince you so not going to spend too much typing energy to try.

I see it as obvious that Krugman is on the side of the White House growth projects. We can atleast agree that the White House growth predictions were horrible.

Either way, I am fine with letting readers read it for themselves and see who, in retrospeak, looks more prophetic.

Jon said...

What I saw Krugman say was IF certain conditions applied THEN growth rates would proceed as follows. Did those conditions apply? You have to know that before you can claim he made a mistaken prediction. As I said, maybe I've misread it, but it's not my job to make your argument. You made a very bold claim. Krugman is a HORRIBLE predictor. But you won't quote him showing a mistaken prediction that he made. You have to at least quote his so called horrible prediction. That's not asking too much. I provide sources directly to links that show specific mistaken predictions from the right wing. This is what you won't do. You once again run away with a "Let the reader decide" before offering the evidence needed. This is a repeating pattern.

HispanicPundit said...

Dude, read the exchange. Mankiw is saying the WHITE HOUSE PREDICTIONS ARE GARBAGE...Krugman is saying, hold on a minute, the WHITE HOUSE PREDICTIONS have merit.

IN the end, the WHITE HOUSE PREDICTIONS turned out to be...exactly that, garbage.

The main point of the discussion was precisely THE WHITE HOUSE PREDICTIONS. Again, I have no chance to overcome your confirmation bias. I dont know why I even try...

Jon said...

Chomsky gets heat from a lot of people because prior to the war in Afghanistan he said it was gravely immoral because the food aid agencies said it put something like 2 million people at risk of starvation. What happened was the worst wasn't realized. Starvation was not as bad as the aid agencies had suggested. So people do to Chomsky what I think you're doing here to Krugman. Since he was saying that 2 million people would starve and that didn't happen he's a bad predictor.

But of course he wasn't saying that. He's saying that we must make choices based on the best available evidence we have at the time. If some conservative had said "I bet less than 2 million will starve" Chomsky would have declined the bet because as a rational person he knows that he doesn't have the kind of information that would allow him to know that. Conservatives are often different. They're ready to throw down, like they can know what's actually going to happen. They know that less than 2 million will die. They know that Romney will win in a landslide. Liberals don't pretend to know. They do what Krugman did regarding the election. We don't know who will win, but we are inclined to think Obama will.

I mean, suppose Nate Silver said Obama would win and you ask how he knows and he says he flipped a coin. You criticize the method and he says "Oh I see. You think Romney will win. Let's bet on it." And you decline. Not that you think Obama will win. Maybe you think Romney will win. But you also know that you're not in a position to know for sure. So when Obama wins Silver walks around pretending like you made a bad prediction. You are a HORRIBLE predictor. Look at the prediction you made. What prediction? You didn't make a prediction. You criticized a method. That's not the same.

HispanicPundit said...

The difference is that Krugman threw his weight behind the White House Predictions. He defended their math AND overall argued that it was a reasonable projection.

Projection turned out to be utterly wrong.

Jon said...

A projection can be both reasonable and wrong. Do you understand how?

To me you very much have a double standard. Here's another case where Krugman claimed a projection was reasonable. He said Nate Silver's projections were reasonable and Obama would probably win. But he also said that he might not because you can only have so much confidence in these kind of projections.

For me if you make a prediction you stick your neck out a bit. Did Krugman stick his neck out on Obama? Not really. So he doesn't get a lot of credit for saying he expected Obama to win. If Romney had won you COULD NOT say that Krugman made a mistaken prediction. He didn't say Obama would win.

But by your standard (he threw his weight behind Silver) he made a correct prediction. By my standard he did not. He did not stick his neck out and tell us who would win. If you were consistent you'd give Krugman credit for throwing his weight behind projections that turned out to be accurate, but you don't do that. You only notice the case when he throws his weight behind a projection that turns out to be wrong.

Notice I didn't say Obama would win. In comments I made you may have seen that I directed Chad to look at Nate Silver and also noted that Obama was out raising Romney and this would make Obama tough to beat. In my view elections are basically bought, so whoever has the most money should be expected to win. But I'm not sticking my neck out too far. For all I knew Romney could have won. Chad was different. He said Romney by 6 minimum and R's take the Senate. A specific prediction. His neck is stuck out. If he had been right we'd have to give him credit. I thought a certain projection was reasonable, but I was not willing to stick my neck out and say Obama would win, so I can't be given credit for an accurate prediction. That's all I see here with Krugman. He wouldn't take Mankiw's bet, meaning he would not stick his neck out and make a call. If it turned out we did grow like Obama's advisers had said Krugman could not be given credit for an accurate prediction, just like he can't be given credit for calling the election for Obama. He was unwilling to really stick his neck out. But on the flip side if Romney had won you couldn't say Krugman had made a mistaken prediction. He made no prediction at all.

HispanicPundit said...

Read the Krugman defense of the White House projections...his response to Mankiw. He goes into the math (wonkish) and defends the premises. It's a FAR BIGGER stake than what he said regarding the Nate Silver projections.