[kenrexford]

I love this new math.

Take economics. If we do nothing, we will have 9% unemployment, per our analysis. If we spend a bazillion dollars on weird things we like, we will only have 8% unemployment, thereby saving or creating millions of jobs. Later, when unemployment hits 10%, we revise the original estimates of what we would have had without the massive spending, upwards to 11% or so, and then claim a million jobs saved or created. Can't go wrong.

What about health care? We estimate how much health care will cost if we do nothing. Then, we estimate how much it will cost if we do it our new way. We substract one estimate from another, and poof we have savings. Ten years from now, if the Health Plan costs twice as much as we thought, we simply revamp the old estimates of what it would have cost if we had not done this, and we win again.

How about Tarp? We estimate how many companies would have collapsed if we did not pay fats cats money. We then estimate what happens if we do. Poof -- we save the economy. If companies collapse anyway, we revise how many would have collapsed, and we still win.

Climate? We estimate how much the temperature would rise if we do nothing. We then estimate what happens if we spends trillions on carbon reductions, and amazingly we again save the earth. I bet that if temperatures still rise, over estimates, we will revised the estimates. As it is, if temperatures drop or stablize, we introduce a new cooling trend to explain the cooling or stabilization, so that temperatures still rise even though they don't.

What about war? We estimate what a surge will do and what a lack of a surge will do. If the surge succeeds, it made sense, because without the surge we would have lost. If we lose anyway, then the lack of a surge would have made it worse. We win again.

I wonder if this could be used in other issues? Tiger Woods would have had affairs with 400 women if it were not for how good of a husband he is? If Florida and Alabama had played 10 more games each, Alabama would have lost more of them, so Florida should be in the national championship?

[hrothgar]

Wow...

People don't have perfect foresight
Who would have thunk it...

[nigel_k]

Funny how people have these problems with mathematics when risking other people's lives and money but can work things out just fine when it's their own. Have you also noticed how people in government always seem to miscalculate in a way that justifies government being bigger and more powerful?

[hrothgar]

nigel_k, on Dec 14 2009, 10:02 PM, said:

Funny how people have these problems with mathematics when risking other people's lives and money but can work things out just fine when it's their own. Have you also noticed how people in government always seem to miscalculate in a way that justifies government being bigger and more powerful?

Funny how some people have the heads stuck so far up their own ass that the come up with tripe like this...

The company that I work for sells some of the best software for mathematical modeling in the world. I don't dispute the existence of principal - agent problems, however, I don't think that "probems with mathematics" are involved in any significant way.

[Lobowolf]

I guess "We don't really have much of an idea what unemployment will be if we do X, and we don't really have much of an idea what unemployment will be if we don't do X, but we do know that it'll be a lot worse if we don't do X" would be a tougher sell.

[jonottawa]

Your reality's still a little skewed, here's how it really went down:

Economics. These tax cuts for the rich will pay for themselves and lead to prosperity for all!

Health care. Let's do nothing! We have the best healthcare system in the world!

TARP. Let's keep it a secret until the last minute and then announce IF YOU DON'T PASS THIS RIGHT NOW JUST AS IT IS THE SKY WILL FALL!!!

Climate. Instead of getting behind a politically unpopular but principled issue like a carbon tax, let's see how fast we can wipe out species, how dependent we can become on despotic foreign governments and how dirty we can make our environment.

War. Let's torture and kill as many brown people as we can. Let's wonder why they hate us. Actually, scrap that second part, they don't hate us, they hate FREEDOM.

Other issues?

3 or 4 blow jobs are more important than destabilizing the US economy by impeaching the greatest president ever, who brought us unprecedented prosperity. Those are some expensive blow jobs.

[kenrexford]

Let's look at this as a bridge thing. Suppose I were to claim that my "Convention X" works wonders according to my model, predicting its benefit. However, in reality, the end result is that I bid games that only make 20% of the time, but when I stay out of games (on relevant hands), the game would have made 60% of the time.

This may sound like a really bad convention. However, my models may show otherwise. The practice might just not work.

Now, one could say that my models did not predict accurately the end results, because I forgot a factor (maybe that the convention suggests the killing lead too much when I use it). However, if the model did not work, then something is terribly wrong. Maybe the model is BS. Maybe I entered the wrong numbers in. Maybe I forgot a factor.

If this happens a lot, though, I start looking like an idiot, especially when someone else has a different perspective.

I mean, if I say that Convention X is predicted to work wonders, and John Doe says that Convention X will result in going down in a ton of games but missing a ton of making games, and the results seem to confirm what John Doe said, then I will have a hard time proving that my model actually works.

[Lobowolf]

jonottawa, on Dec 14 2009, 02:42 PM, said:

Other issues?

3 or 4 blow jobs are more important than destabilizing the US economy by impeaching the greatest president ever, who brought us unprecedented prosperity. Those are some expensive blow jobs.

Not to nitpick or anything, but Clinton wasn't impeached for extramarital blowjobs (considered a bit sleazy by most conservatives I know, but not worth even thinking about impeachment over), but rather for perjury and obstruction of justice.

Clinton didn't exactly "bring us" unprecedented prosperity; he took and held office at the peak of the dot com boom, when people thought it was a great idea to pay $400 a share for companies with no earnings, because, hey, you can always sell 'em for $450 in 6 months. That's a bit like saying Madoff shouldn't have been arrested, because he did great for the people who cashed out in time. As Lloyd Bentsen once said, "I could create the illusion of prosperity, too, if I could write a trillion dollars' worth of hot checks."

[barmar]

The reason for this "new math" is that there's no other alternative. It's not like you can run an experiment to see what would happen to the economy with or without a tax cut, TARP, healthcare reform, etc. Sometimes you can look at other countries, but they generally don't have the same constraints, so it's an apples/oranges comparison.

So all you can do is guestimate. And when your predictions turn out wrong, you revise. And just as you can't predict definitely what will happen as a result of any particular action, you also can't say what would have happened *without* it.

But what's the alternative? Life is full of uncertainties, you do the best you can. I suppose the "but if we hadn't" statements are mostly self-serving, and don't really serve any non-political purpose.

[Jlall]

http://www.youtube.c...h?v=obIGsb-IZMo

[kfay]

Jlall, on Dec 14 2009, 04:07 PM, said:

http://www.youtube.c...h?v=obIGsb-IZMo

nice

[Elianna]

Jlall, on Dec 14 2009, 01:07 PM, said:

http://www.youtube.c...h?v=obIGsb-IZMo

Sorry, doesn't beat Tom Lehrer's New Math for me.

[nigel_k]

hrothgar, on Dec 15 2009, 07:26 AM, said:

nigel_k, on Dec 14 2009, 10:02 PM, said:

Funny how people have these problems with mathematics when risking other people's lives and money but can work things out just fine when it's their own. Have you also noticed how people in government always seem to miscalculate in a way that justifies government being bigger and more powerful?

Funny how some people have the heads stuck so far up their own ass that the come up with tripe like this...

The company that I work for sells some of the best software for mathematical modeling in the world. I don't dispute the existence of principal - agent problems, however, I don't think that "probems with mathematics" are involved in any significant way.

Point taken. I would normally have described them as failures of logic rather than failures of mathematics, but was following the description in the original post. Incidentally, I always assume an abusive response is an indication I have hit a raw nerve. If you want to be dismissive it's more effective to do it politely.

[Gerben42]

Quote

Sorry, doesn't beat Tom Lehrer's New Math for me.

Touché

[Lobowolf]

nigel_k, on Dec 14 2009, 04:46 PM, said:

hrothgar, on Dec 15 2009, 07:26 AM, said:

nigel_k, on Dec 14 2009, 10:02 PM, said:

Funny how people have these problems with mathematics when risking other people's lives and money but can work things out just fine when it's their own. Have you also noticed how people in government always seem to miscalculate in a way that justifies government being bigger and more powerful?

Funny how some people have the heads stuck so far up their own ass that the come up with tripe like this...

The company that I work for sells some of the best software for mathematical modeling in the world. I don't dispute the existence of principal - agent problems, however, I don't think that "probems with mathematics" are involved in any significant way.

If you want to be dismissive it's more effective to do it politely.

Good luck with that, Nigel.

[helene_t]

I don't see the problem.

It is very difficult to predict future changes to employment, climate etc. Much easier (yet still difficult) is to predict the effect of political measures. And more relevant for decision making, too. I don't really care how much unemployment will be next year. What I do care about is how much difference it will make to employment if we, say, lower the interest rate, or reduce taxes, or spend some money on public construction works, or whatever policies are considered.

Of course, we won't be able to verify whether the policies worked. But what else is new?

[luke warm]

nigel_k, on Dec 14 2009, 04:46 PM, said:

If you want to be dismissive it's more effective to do it politely.

what made you think he was being dismissive? the 'head up your ass' comment? hell, that wasn't dismissive - it's the way people with double masters from mit normally talk

[hrothgar]

luke warm, on Dec 15 2009, 02:01 AM, said:

nigel_k, on Dec 14 2009, 04:46 PM, said:

If you want to be dismissive it's more effective to do it politely.

what made you think he was being dismissive? the 'head up your ass' comment? hell, that wasn't dismissive - it's the way people with double masters from mit normally talk

Fairly often, actually...

Nigel posted an ignorant little rant.
Its not like he deserves or has any reason to expect a polite response...

[kenrexford]

Oh, and one more. This one I think is not subject to the "it's a model" or "it's difficult to know for sure" retorts.

We pay taxes for health care for 10 years. But, we only get health care for 5 years. So, for the 10-year projection, health care doesn't cost that much and may even save us (the government) money.

That's brilliant. We could also do this one a different way. We get health care for the full 10 years, but we pay twice as much in taxes. Again, the problem (for the government budget) is solved.

But, that would be too obvious, relying on old math. With the new math, this works out much better.

[kenberg]

From an article in today's post by Robert Samuelson:


http://www.washingtonpost.com/wp-dyn/conte...9121302450.html

Quote

In 2009, national health spending will total an estimated $2.5 trillion, or 17.7 percent of gross domestic product. By 2019, it's projected to rise to $4.67 trillion under present policies, or 22.1 percent of GDP. With CAP's "savings," it rises a little less sharply to $4.49 trillion, or 21.3 percent of GDP, according to Harvard economist David Cutler, the study's co-author, who provided these figures. Similarly, family health insurance premiums rise from 19 percent of median family income in 2009 to 25 percent in 2019 under present policies and 23 percent with CAP's "savings." The point is simple: Even with highly optimistic assumptions, health spending remains out of control. It absorbs more of government, business and family budgets. Higher health spending would put pressure on future budget deficits, already projected to total about $9 trillion over the next decade. If new taxes and Medicare "savings" are real, they could be used exclusively to pay down deficits, not finance new spending.

Myself, I don't like the quotation marks around "savings". If an action reduces costs from what they would be without that action, I don't think that it is misleading to call it savings. If I am going to (as I did) undergo a painful medical procedure they may well give me medicine to reduce the pain. I would understand that they mean reduce the pain from what I would experience without the medicine, I don't think they are saying my pain will be reduced from it's current level.


So the quotes I don't like. Looking carefully at the claims sounds very right, however. With regard to medical costs, here is another story. With my recent procedure I got a prescription for anti-nausea pills. Twenty pills, the cost to me was $5, I used two although they were not strictly necessary. Speaking later with the pharmacist, I learned that the cost of the pills, paid somehow by someone, was $400. I am a pretty healthy guy so this stuff with pills is pretty new to me, although I did talk earlier of the cream I got for an infection, $250 a tube, two tubes. Among those making decisions, me, the doc, the pharmacist, no one has any motivation whatsoever to blink at the cost. If it was my $400, directly paid, the pills would still be in the store.

I am sure that there are a lot of people out there who use far more pills than I do. That's a great deal of money changing hands, and it seems likely that it was not just my $400 anti-nausea pills that will be pitched in the trash.


So I think that it is very fair to ask whether the planned programs are doing something to bring some of these costs under control. If indeed the projection is to bring the costs down from an expected 4.67 trillion to 4.49 trillion, I think it is fair to call it a savings (rather than "savings") but I would hope we could do better.