by Tyler Cowen
on February 14, 2013 at 12:34 pm
in Uncategorized |
1. Should you foul when you are up by three points?
2. Izabella Kaminska on cash hoarding.
3. Sequestration? Defense stocks are rising in line with the S&P 500.
4. Ear bounties in Azerbaijan?
5. Does fertility stand a chance? If so, when? And why give someone else this kind of foothold?
6. Some pre-K studies from Latin America, and here is more on Oklahoma.
What a useful study on pre-K program studies – ‘Large positive effects on child development were found. Because of the limited evidence, no conclusions can be drawn regarding the impact on child health and nutrition.’
So, since Finland isn’t like America, and Latin America isn’t like America, why bother to do anything in America? Sounds like a study waiting to be written – at least as a blog post.
But wait, there is still the second link – ‘A wealth of research has shown the effort has improved these students’ academic, cognitive and emotional abilities, which researchers attribute to a strongly supported program with higher standards. But it’s unclear whether those gains persist further down the road and how a program like Oklahoma’s could be replicated nationwide.’
America – the country where evidence is simply not relevant because it is ‘unclear’ or no conclusions can be drawn.
As if the idea of proper nutrition in young children is somehow still needing further study, for example.
The ratio of snark to useful information in your comments is astoundingly high. Is this an internet act or are you this annoying in real life?
Well done sir.
And that’s despite the fact that there _is_ quite a lot of useful information in PA’s comments.
Whereas you comment has only snark, with no useful information.
So let’s take advantage of the natural difficulty of instantly creating such a plan to do a full randomized controlled trial which smoothly increases the coverage rate. Each year, choose 7k of the 70k elementary schools in the US at random and introduce a pre-K program. This creates two cohorts each year (7k vs 63k the first year, 14k vs. 56k the second year, and so on). Each year,the data will cover a longer period out from the introduction of the pre-K education. If it works, then the roll-out is continued or accelerated and we have great evidence of effectiveness over longer and longer times scales. If it doesn’t, the program can be scrapped after spending only a fraction of the total cost. If it’s ambiguous, the roll-out can be slowed, *but the data collection continues*.
The problem is that if it doesn’t work, it won’t be scrapped. The cries will be that it wasn’t funded enough.
My comment was addressed toward people who support the proposal but are open to the idea of looking for evidence. If you already believe the government is incapable of responding to evidence, then there’s nothing to discuss.
then there’s nothing to discuss
The ironic thing is that as thin as this is, it’s probably more supported than almost everything else the government does.
“Since I predict irrationality in the future, I will not be rational now.” Brilliant. That is just the kind of thing that has led to our high performance, national, legislature.
I don’t understand PA’s complaint about the first quote. Aren’t “large positive effects on child development” the whole goal here? How is this not a supporting argument for the pro-preschool side?
I realize that their comments are not really to be taken seriously as some others, but this one was quite eyebrow-raising.
“The Heritage Foundation points out, for instance, that fourth-grade reading scores in Oklahoma have actually declined since universal pre-K has been implemented. “Oklahoma was the only state to see a significant score decrease on the NAEP fourth-grade reading assessment and is the only state to see its reading scores decline over the 15 years from 1992 though 2007 out of all of the states that participated in the fourth-grade reading test in 1992”
What is your underlying assumption? That if Tyler posted this link he must think it is anti-preschool evidence, because he earlier expressed doubt about Obama’s plan?
I’m not sure you have read this blog very closely.
On #1, I wonder if 12 seconds should be the proper upper limit to use. I don’t often see intentional fouls in this situation with more than 5-6 seconds left on the clock since, with more time than that on the clock, it often doesn’t represent a “last possession” situation. For example, if you foul with 8 seconds left, the optimal strategy for the opponent is to try to hit both free throws, cut the lead to 1, and then commit their own foul trying to guarantee another possession of the ball (assuming they rebound a miss) with the time needed to bring it down the court and run an actual play. I would like to see the analysis redone studying fouls committed with under 6 seconds on the clock.
I think playing defense is the optimal strategy when up by one, not fouling. Unless you mean, fouling on a shot that the other team is definitely going to make.
The foul would come after making 2 shots to be DOWN by 1, not up by 1. The team down by 1 wants to foul to get the ball back immediately.
It’s actually a pretty complicated problem. With 12 seconds left, it’s much easier to set up a 3 pt attempt for a variety of reasons. But that attempt will actually most likely happen with quite a bit less time on the clock. You don’t want to exclude these cases Part of the goal is to figure out whether you should look to foul when your opponent takes possession with 12 sec left down by 3. You aren’t looking to foul immediately though and you may never get a good opportunity (can’t risk fouling the shooter during the shot, for example). The balance might be different if there are only 5 seconds left when the ball is inbounded, which is a different situation. That makes it difficult to build the right set of cases and also to even tell whether a team’s strategy was to foul or not. Just because they don’t end up fouling, doesn’t mean they didn’t want to.
I think any data on how often the shooting team rebounds a missed FT needs to be specific to the desperation present in an intentional miss scenario. Before then, the offensive rebounder is risking drawing an over the back foul and a quick trip to the bench and the 2 players not on the lane are guarding against a fast break. In an intentional miss scenario, 3 or 4 guys crash the boards hard and 1 or 2 set up for a three.
That is explicitly addressed at the link
@#3- die DC, die. Beltway bandits working for Pentagon will have to hit the road. The rest of the USA outside the beltway will benefit.
1. There is an endogenous choice involved here to defend or foul. Coaches choose to defend four times as often. Maybe they know something about the game momentum, relative fatigue. The choice itself might make the difference. There is some value to the information that the team that’s down wins 55% of the overtime games.
Moneyball showed us that coaches’ intuition might be wrong but that was an operational decision, not a tactical decision.
The information is not that “the team that’s down wins 55% of the overtime games,” it’s that “the team that was down won 55% of overtime games falling into this arbitrary dataset over a three year period.” Quite different things. Most likely the odds are very close to 50-50.
And coaches make plenty of tactical mistakes every day.
It’s not an “arbitrary data set,” it’s a sample from the population, and provided that it’s a random or representative sample, the sample proportion should be relatively close to the population proportion. Based on the size of the sample, you would almost certainly reject Ho: p= .50 in favor of Ha: p > .50 where p is the proportion of games won by the trailing team within 12 seconds of the buzzer. I can’t and won’t tell you what theory suggests this hypothesis, but that’s not necessary.
If the odds of winning were pretty close to 50-50 (even in overtime) you wouldn’t see winning percentages that varied much and you wouldn’t see large point spreads from book makers.
Yes, I’m sure that coaches make lots of tactical mistakes. But here we are looking at a single tactical decision at the decisive point of the game. The decision process might not be memoryless, but it is still clear that most coaches prefer defense to fouling. That decision seems to have a slight edge in the margin of victory, and the slightest of edges is crucial to long-run success. Many sports contests are decided by fractions of a second or fractions of an inch. The Super Bowl victory was decided, at the margin, by four plays five yards from the goal. Minutia matters in modern professional sports.
There have been similar analyses of going for it on fourth and short in football versus punting. It’s possible that most coaches have been making that decision wrong given a rigorous analysis. However, once more coaches start going for the first down on fourth and short or faking a punt, the defensive tactics will change, and then punting might be better. These are competitive games, and probabilities of success are depending upon choices made by the competitors including past decisions.
When the 1980 US Olympic Hockey team won the gold medal, there is no doubt that superior conditioning was what made all of the come-from-behind victories possible. Stamina at the end of the game, down to an inch of precision on several single shots, made all the difference between first and worst.
3. Defense stocks are rising because they are slashing costs and improving projected profit margins. Defense firms have already anticipated their recurring and nonrecurring profits.
Lockheed Martin yields 5.2% at a time when 10 year treasuries and S&P both yield 2%. That generates a lot of demand from income oriented investors.
Glad you looked at the numbers because I don’t analyze individual stocks anymore.
But yeah, a steady profit and dividend stream, even at a relatively low level, almost becomes a high yield fixed income asset. It has a beta close to 1, so yeah it’s going to follow the S&P. The article makes it sound like a surprise.
the price is down almost 9% in the past month which makes it a bit of a bargain. The firm projects higher profits than the street by 70 to 80 cents a share. I think the recent dump is dumb money betting on the headlines. The firm knows what its own prospects are.
LMT has a lower PE than its peers and a much higher dividend yield. It’s payout rate is 50. The industry is still paying better than treasuries. I wouldn’t have a large stake in that industry for a portfolio, but obviously you need some exposure. LMT would be a buy rec from me.
1. It depends on who you foul. Seriously.
In other words, you can mathematically solve this problem for a perfectly average free throw shooter and a perfectly average defense. But in reality there are so many variables in any given game that it’s impossible to give a pat answer. Which is why, I think, the % outcomes of either choice have so closely converged. Coaches take several variables into account when deciding whether to foul or not, and they usually make the right choice.
I agree with you in theory, but the offense has the choice of who has the ball and, in a sense, who draws the foul if there is a foul. The ball is probably going to be in the hands of the best shooter most of the time.
And that’s probably why coaches prefer defense 5 to 1 in this data set. The percentage differences are razor thin, but professional sports are won by razor thin margins.
Consider a casino that has a 51 to 49 vig. Given enough time, it will make a lot of money. It only has to worry about liquidity from a bad run and the casino is likely to be farther above the felt than any of its customers.
The “demographic transition” fantasy is like the fantasy that we can over-use antibiotics without developing resistant strains of highly virulent bacteria.
Economic development is like an antibiotic against human reproduction. The fact that you’re successfully committing genocide against a huge portion of humanity blinds you to the fact that you’re selecting for strains of humanity that will find ways of more efficiently turning wealth into an exponential growth of their babies. That optimum appears to involve lowering the age of female puberty and increasing the rate of de facto transfer payments to support their offspring. The idea that “property rights” are the answer must take into account the political dynamics of the recent election as a warning: Liberal democracy has a _strong_ tendency to serve the most reproductive.
A female that pumps out 1 child a year from age 8 until age 38 has a 30 to 1 gain over those 38 years. That means an effective doubling time of under 4 years. Many of us may live to see this new breed of “human” become a dominant demography.
The Framingham Study is showing evidence for this type of natural selection (age of menarche, age of menopause) in humans:
3. Basically the argument is evolution will select “takers” because they will have more offspring.
If you believe in total self reliance rational action, then you will put off having children until you have lots of savings, a good income — but the costs of having kids just keeps getting higher and higher.
Locke wrote of the opportunity of the Americas because, unlike England, it offered lots of common land, so much common land that anyone who wanted the security of land could carve it out of the common. As long as sufficient common remained to support all those without land, then the land could support everyone.
But today, the common land is gone, so you either need to own land to live off the land, or you must work for others, and if working for others is insufficient to provide what you need, then you aspire to less. Or be a taker.
Yes, of course the problem is a lack of free farmland which otherwise I’m sure lots of people would avail themselves of.
Not sure why people say children are expensive. Daycare’s pretty expensive but it goes away after a few years. Otherwise, it’s just food and used clothing and a few toys. Well, I guess technically you need a bigger house, but most people have much bigger houses than they need anyway.
You’re ignoring the biggest expense of children: insulating them from “bad schools”. That means paying for private school, or paying up for a house in a neighborhood with “good schools”.
Did anyone look at the benefits of pre-k allowing parents to work more?
Comments on this entry are closed.
Previous post: Interview with James Heckman
Next post: UK fact of the day
Email Tyler Cowen
Follow Tyler on Twitter
Email Alex Tabarrok
Follow Alex on Twitter
Subscribe in a reader
Follow Us on Twitter
Marginal Revolution on Twitter Counter.com
Get smart with the Thesis WordPress Theme from DIYthemes.