Tuesday assorted links

1. “We find a small hot hand effect for free throws, concentrated in second and third shots in a free throw sequence, in players shooting at least 100 free throws in a season, and in games where players shoot four to five free throws. We find the opposite results for field goal attempts. If a player makes a field goal, he is less likely to make his next field goal attempt. These results are robust to controlling for the characteristics of the previous shot. Interestingly, both offenses and defenses respond to made field goals as if the hot hand effect exists.”  Link here.

2. “[German labor in the boardroom] …lowers outsourcing, while moderately shifting employment to skilled labor.

3. Do good teachers make you taller?

4. Claudia Sahm podcast with David Beckworth.

5. Alex Bell is now awesome and not just apparently awesome (NYT).


If defenses are playing *as if* the hot ha d exists, wouldn’t that account for the lower success of second shots?

And the offense also is assuming the hot hand exists, so they feed the guy the ball even if his next shot is not so open and well chosen as the one he just made.

I'd like to see the numbers run on '70s Cavalier bomber Bobby "Bingo" Smith -- one seeriously streaky guy I recall who really did seem to flash a hot hand.

My worthless theory: the "hot hand" phenomenon exists in some sports but not others, and it's affected by the time between repetitive motions, which is why you see it in free throws but not so much during game action. there's probably at least thirty seconds between 3-point shot attempts, for example, vs usually 10-15 at the charity strip, and the distances and angles usually vary with each 3 pointer, wheareas FT's are the exact same shot, every time. As such, the hot hand is relatively easily detectable., but too many other factors are at play in regular game action, though. IE, too much running and jumping and dribbling between each shot, to allow anyone to get into much of a rhythm.

I don't discount the three point shot but the linear progression of a shot throws off defenders more than time the shooter disenables. It comes off as a stance issue when the idea is to develop a hierarchy of autonomous thinkers.

Players warm up before games, which suggests that there is a certain amount of truth to a hot hand theory at least to the extent that shooting cold isn't a good thing.

Is there an ungated version of the new hot hand paper?

I would like one too. If I pay for it, am I permitted to share with others from this thread?

The statement in link 1 (I haven't read the link) *screams* "garden of forking paths" to me.

I suspect the difference is that the 'hot hand' may exist for situations where the player can stay 'in the zone', or in a state of 'flow'. Anyone who has played a sport for a ling enough time knows the feeling of being 'in the zone'. What's really happening is that System 1 (unconscious trained behaviour) is taking over from system two, or the conscious attempt to achieve something. That's why you can totally put someone 'off their game' and getting 'inside their head' by calling attention to something they are doing, forcing their conscious mind to take note.

Field goals are separated by a lot of time and activity, so there is probably no 'flow' being maintained. And if they miss the first shot, it might even cause them to be unable to get into 'flow' again because they are thinking about the last kick. Free throws are not like that, They are rapid-fire and rhythmic, so it would be a lot easier to maintain being 'in the zone' between shots.

I'm surprised by Tyler's reaction to #5. The efforts (possibly too many pure bike lanes and stricter enforcement thereof) seem sub-optimal. You've got city blocks. You don't have universal underground delivery (a la Disney World), so what do do? I think there have to be some minutes of delivery allowed at every address.

At least until Amazon Drone Delivery Nirvana. Or Mole Man Nirvana. Either way.

I think it's more of an issue of having data to make the case that enforcement is weak. TBH the best solution is probably something like parking on the sidewalk or something -- permitting the blocking of bike lanes is not a good solution because that causes sometimes deadly conflict between bike traffic and vehicle traffic. The second order effect would be to diminish bike lane utilization, and then the third order effect is worse vehicle traffic.

The whole benefit of bike lanes is that they have significantly higher throughput than do car lanes, but you can't see a benefit from that if people don't opt to use them because they're usually blocked.

Are you talking about theoretical throughput, or actual numbers of people that move through bike lanes vs car lanes?

My city is festooned with bike lanes, on many of which I have never seen a single bicycle. The downtown bike lanes have a few, but I'd hazard a guess that the busiest bike lane in my city doesn't come close to the number of traffic movements of even average car traffic lanes.

Yep. Is there any evidence that bike lanes do anything?

I'm willing to be wrong but they strike me as the virtue signalling rather than a meaningful solution.

If you're a politician trying to get a elected/re-elected in a progressive (LOL) city, bike lanes are very useful.

Pfft. As an ex-Republican independent this is one of the things that gets me. The bizarre binding that conservatives love to sit fat asses in cars, and it's progressives who get out there to run or bike.

Yep. Progressives see defenseless, bicyclists needing state protection. Conservatives hate anything Progressives like and neo-liberals see an externality in need of a Pigou tax

Its fascinating to me how cyclists really seem to consider themselves morally superior to others.

Can you really not think of any reason other than laziness, or politics, why someone might drive instead of bike?

There is a bit of a fall-off after the top twenty cities, but I think bike boosters want to climb the list authentically, rather than just signal.

For those who don't follow links, Davis CA is the outlier, with 23% bicycle commuters. Most of the top 10 are in the 5-10% range.

For comparison, the top bicycle city in the world, Copenhagen is at 35%.

For some urban planners, it doesn't actually matter if the bike lanes are used by bicycles, because the point is to make it harder to drive and park in congested areas. Look up 'traffic calming'.

It used to be the job of civic planners to design road systems that could efficiently handle traffic demands. New roads, road lane expansions and other infrastructure were normal to any growing city. In the past few years this seems to have flipped, and now urban planners spend time thinking up ways to make driving harder. The progressive view of cities seems to harken back to Beijing ca 1970. Brutalist architecture and everyone on bicycles.

Don't know what urban planners "do" but they ought to be trying to reduce the externalities of congestion. Taking "demand" for an unpriced resource as given is clearly sub-optimal s would be maximizing the cost of driving.

Also, many are done simply for safety affects for drivers. Witness lawyers road from Vienna to Reston (if you’re local). The bike lane was added as a means of making traffic single file and reducing accidents, and was lauded as a safety measure first and a victory for cyclists as an after thought. This is the thinking with some of the stranger bike lane arrangements one sees around here.

Yep. How well would NYC function if no-parking in bike and bus lanes were strictly enforced against delivery drivers?

Tax bicycles to pay for bike lanes

Tax children to pay for playgrounds while you're at it.

Tax people who work to support themselves to pay for people who choose not to support themselves.

#3: For some reason this reminded me of an article (linked below) where the author fails to realize that all the options she proposed are as "discriminating" as the one she is ranting against. https://www.theatlantic.com/magazine/archive/2016/07/the-war-on-stupid-people/485618/?fbclid=IwAR02191ZudlU1NHuPoSyunbnTkSTNQkONwR79XXxq4iBeRsDfuW1KVUe7-I

Is that the correct URL? The article was interesting, but AFAICT David Freedman offered three options that don't seem particularly discriminatory.

More vocational education. Though they are perhaps a minority, there are a lot of people across the political spectrum, from Donald Trump to raging leftists, who think the US should be doing more of this.

Reduce incentives to automate, so there are more jobs for low-skilled humans. On its face this is perhaps a less attractive option, OTOH there are plenty of people who say that US states should stop increasing their minimum wage -- because high minimum wages encourage more automation, at the expense of the unskilled workers. A more left-wing policy would be some tax incentives or regulations to discourage automation, but Freedman doesn't say that he's in favor of those policies.

Employers should put less emphasis on intelligence and on other characteristics or behaviors that contribute to workplace productivity: interpersonal skills, self-awareness, emotional intelligence. This prescription is hard to evaluate, obviously it's possible to overvalue intelligence and to undervalue it. But if most employers are overvaluing it, then yeah they should stop doing that.

I'm not saying that the article's prescriptions are correct or wise, but they don't seem discriminatory towards low-IQ people or against high-IQ people.

Obvs good teachers say things like "stand up straight" and "don't slouch!"

I've never considered the possibility that teachers can have an effect on student height. But other than your example, what about the 5th grader teacher who keeps the door and window blinds closed, not allowing any natural light enter the dark classroom, who doesn't let the kids play outside anytime after lunch break is over (not even on Fridays), and is really strict and boring. Sounds silly, but who knows.

I couldn't be bothered to put the effort into actually finding out what's going on, but maybe students who mature faster do better and that outweighs the effect of teacher induced rickets

#3 Maybe not directly, but good nutrition does. If good nutrition and self-discipline are things that can be taught, good teachers can in fact make you taller.

#5 I would be pissed off too. If you're going to have dedicated lanes for a certain type of vehicle then it's inefficient when they're blocked. That's a two-way street though bikers. That applies to you on sidewalks and car lanes.

Just to be clear though, the only "car lanes" in most of the US are freeways. All others are vehicle lanes, and bicycles are vehicles.

And here's a preemptive link for anyone who wants to be tedious.

You are correct; roads are for bicycles and cars (even if stop signs are only for cars based on my observations)

Most bicyclists accept a "track stand" which may actually be at some fraction of a few mph.

Not that kids are always so careful.

The "right turn on red after stop" is interpreted by drivers as a suggestion to slow down somewhat, if at all.

> All others are vehicle lanes, and bicycles are vehicles.

Right up until they have to stop at stop signs or lights, and then bicycles are special.

If you are expecting a "stop and dismount" you're never going to get it, sorry. Probably never in the history of the US bicycle or stop sign.

There's a bit of difference in the amount of energy carried by a car going 15mph and the energy carried by a cyclist going 15mph.

I love it. We deserve to be treated equally, except, of course, when it is inconvenient for us. Then we are more equal.

ooh boy let's get that bike v cager flame war ready

In my experience with usage of "hot hand" growing up, making one FG wasn't sufficient. You had to have made at least two, perhaps more. I think one of the NBA video games from way back (name?) confirms this: players weren't (ever?) granted being "on fire" based on one made FG. Does this literature address this anywhere?

Example #1,927,375 of an analysis being performed by someone with little actual knowledge and experience in a subject but who employs super cool maff methods.

I only read the abstract but I didn't see anything controlling for time. MASSIVE ERROR. The second most important variable in the "hot hand" function, after shots being made, is time. Hot-handedness does not exist if a player makes a shot in the first minute of the game and then his next made shot occurs in the final minute.

On the contrary, there are probably a dozen choices of definition of what a hot hand is. Which is part of the problem.

Sure, and if any of those 12 ignore the element of time, they're foolish.

Agreed. If there is a 'hot hand', it is synonymous with being 'in the zone', a phenomenon any experienced athlete can tell you about. Being 'in the zone' is a temporary state of mind, and time between attempts should be absolutely crucial. Staying in the zone is not easy, and fast repetition, rhythm, and unbreaking concentration are required.

#5. Why not have just bike and bus streets? Take two avenues and only allow buses, bikes, and delivery trucks. One goes north, the other goes south. Then there would be no blockage, as the buses will have plenty of room and bikes can stay to the left.

Even better: why not have only a handful of private motor-vehicle thoroughfares, and everything else could be reserved for pedestrians, bikes, delivery vehicles, and buses? The fact that I can get in my SUV, alone, and drive it into Manhattan in the middle of the day is an astoundingly terrible waste of scarce resources. Nobody who didn't grow up in North American car culture would see this as anything other than idiotic. Like letting people bring their horses with them into office buildings. "It's my horse, I need it to get to work! You're a hippy elitist!"

This is a much better idea than people will give it credit for. IMO the best bike cities aren’t the ones with a lot of bike lanes, but the ones with plenty of straight through streets, so that one can select a less trafficked one to use. Applying that and simply designating one street as bikes only (or correspondingly, trucks only) does a much better job of controlling traffic for most of the commute. There’s still a bit of a “last mile” problem once leaving the bike or truck road, but in dense cities hopefully that last mile is more like a quarter mile for most.

a critical neurosis

Traffic congestion isn't a modern problem. In Imperial Rome commercial deliveries were restricted to the night. It doesn't make any sense for trucks to be allowed to congest urban streets and freeways during morning and afternoon rush hours.

#3 Academic achievement and height are both casually dependent on physical activity in childhood. Perhaps the teachers with the longest recess are the best teachers?

#5 I think were I designing a city from scratch there would be no parking of any kind allowed on city streets. You want people to have cars at your building, it’s on you to figure out how to provide access.

2, So the German model leads to increased human capital?

Well, clearly, there must be some other explanation why Germany is such a successful industrial exporting society that does not positively reek of something like socialism, with workers having a say in company direction.

Is it still OK for capitalists to buy Porsches in this case?

1. I don't understand missed free throws. I played HS basketball and a free throw is a "free" throw: it's the same distance from the same angle to the basket every time and there's nobody playing defense. And the basket is . . . . the same height every time. My problem in basketball is that I was too short, too slow, and too white (that's redundant). At the "free" throw line, none of that mattered. It's not a matter of a hot hand, it's a matter of repeating the same motion with every shot. My primary sport was golf. What do good golfers do? They repeat the same swing every time. I bet if there were a free throw contest among athletes from different sports, the golfers would win. Repeatedly.

You don't understand missed free throws? So you shot 100% from the free throw line? Did you also shoot 100% in your driveway with zero defense?

Also, the paper doesn't argue that free throw success is a "matter" of the hit hand - whatever that means. It simply tests if the hot hand effect exists w.r.t. free throws.

Also, your argument that golfers would fin free throw contests versus other athletes makes no sense. Consider the inverse, "What do good shooters do? They repeat the same shot every time. I bet if there were a free throw contest among athletes from different sports, the basketball players would win. Repeatedly."

Your argument is silly at best and reeks of bias.

The only Dem POTUS candidate's bumper sticker I have seen (in western Nassau County, NY) was a "Tulsi 2020."

Good point. Mrs. Gabbard's run has got momentum because the average American identifies with her, her down-to-earth style, her patriotism and her sense of duty. I think she will inherit many of Harris' supporters.

#3: One of the commenters to Kling's post has a good idea: tall children are more likely to be offspring of successful (economically and academically) parents. And will show more learning and academic improvement during any given year in school.

Voila, more value added. But due to height (as a proxy for the amount of learning the student will be doing during the year), rather than the causality going the other way.

I haven't looked closely at the research into value added in education. As with any complex measure there are a ton of pitfalls. In particular, it's probably not enough to account for students' entering test scores, one should also account for students' entering "ability to learn during the year". I'm guessing that researchers have not used height as a measure of that "ability to learn" -- but they may have looked at parental income, parental education, parents reading to children or spending time with them on homework, etc. etc.? If so, then they've probably already covered the (apparent) effects of height.

We're into the territory of Lord's Paradox, where we may have an omitted variable problem if we simply look at gains, and fail to account for other variables such as family background (or height as a proxy).

The policy and political context makes this problem especially difficult: once we realize that we need to account for covariates such as family background, we now have the problem that bedevils every researcher who uses multivariate data, namely how to decide which covariates or right-hand side variables to use. Different researchers will disagree, and when important policy decisions ride on the results, politics will start being a factor with different factions championing different models.

Oh and #4: Speaking of paradoxes, here's another one. One of the main reasons I come here is to read the comments. Yet at the same time many of the comments are trolling or useless insults. It's too bad the crappy comments drove Claudia away from commenting on MR.

Yet despite the often worse-than-useless comments, MR still has some of the best comments on the web (maybe that's more of an indictment of the web than praise for MR). Maybe Claudia can post (or has been posting??) anonymously, as "C Mash" or something.

Do we really want to live in a world where every single comment is deeply thoughtful, insightful and respectful? I don't think so.

MR's comments seem to strike a pretty happy medium between for example Zerohedge (animal house doltishness) and Slate Star Codex (belabored earnestness).

The fact that the wealthiest nation in world history had enabled an educated minority to reflect upon the nuances of a children's game played by hyper-thyroid adults is truly mind boggling.

Tyler has misrepresented the paper, which doesn't make any claim about 'good teachers'. It shows that the proportion of variation in height explained by the teacher one has is similar to the proportion of variation in test scores explained by the teacher one has. The set of teachers with 'tall' students may not intersect with the set of teachers with 'high scoring' students. The conclusion of the authors is not that teachers explain height but that the models used to study achievement must be flawed. However, it is entirely plausible that (say) taller, more athletic students end up concentrated in certain classes, and thus more often having the same teachers than would be observed if they were distributed randomly across teachers. Same with high scoring teachers.

This is a powerful piece from Professor Harrington experiencing firsthand the dangers of anti-foreigner laws and government. I hope you can give it the broad audience it deserves.


"Mike G." comments on number three, regarding teachers, at that link:

At minimum, mostly fake news.

Read or skim the paper.

VAM is supposed to use shrinkage across multiple years. That’s the proper technique.

The authors acknowledge very late in the paper:

“When we apply the shrinkage across multiple years, the teacher effect on height goes away.”

To summarize their paper:

1. When we followed the recommended approach, we found zero height relationship. ZERO.

2. But that doesn’t make a publishable paper.

3. So we used a non-recommended approach, justifying it by saying sometimes other dumb people use it.

4. Result: publishable paper!

The same technologies that cities could use to collect taxes from vehicles in congestion or parking could also charge for blocking bus lanes. Less global solutions would give reporting citizens a cut of evidence they produce of traffic violations.

"Now awesome? The article is from March 2018.

Don't think anything has changed since then. On my Avenue, the bus lane is blocked 24/7 by NYPD officers parking their private vehicles.

Comments for this post are closed