The story is here. It’s kind of like the Excel error in that famous economics paper, except that (a) it seems to have been on purpose, not an accident (according to the alleged perp, “The hacker has hacked into my administrative account and is using my accounts to make changes”), and (b) it did not do any damage to the world economy, it just enabled a fictionalized story to be presented as fact in a Hollywood movie.
So, no biggie except for people involved in long-distance swimming—for them it’s a big deal.
Not much going on here on a cloudy day, so I clicked on the link and came across this surprising English-language phrase: “The suit accuses Steven Munatones, a swimming influencer…” The next thing you know there will be a movement to swap Greenland for our Panama Canal. The year 2025 promises to be interesting.
The Hollywood movie’s subject/namesake has a passion for Columbia.
In her 2015 memoir, Find a Way, she writes that her grandfather, George Warrington Curtis, “was captain of the Columbia swimming and track teams and he was the first person ever to swim across Long Island Sound” (pp. 38-39).
In one version of her marathon swimming origin story, she says a fellow grad student told her about the sport after watching her swim in Columbia’s pool.
Nyad started grad school in 1973. She completed her first marathon swim in 1970. No evidence supports her claims about her grandfather.
She tells the marathon-swimming origin story in “How She Does It: The Power of Perseverance with Diana Nyad” (tinyurl.com/nyad-power – 4:06) / clip: tinyurl.com/nyad-columbia.
Another data-related post regarding the alleged perp’s Nyad-related activities:
“Free Solo Confidential: How The Openwaterpedia Vandal Gave Himself Away”
Munatones shows how quickly he can work under pressure.
tinyurl.com/owp-vandal
As I recently replied to Paul Alper, the big differences between R&R and others were deliberate choices rather than the accidental Excel error. But like you say about dead horses, you need to keep beating them because they’re never really dead.
Inspiring post in surprising ways!
First, the controversy within the WOWSA makes me think of this episode of This American Life
https://www.thisamericanlife.org/371/scenes-from-a-mall/act-three-0
about the putsch that went down in the Amalgamated Order of Real Bearded Santas.
Second, its interesting that we keep seeing posts about silly errors in academic work that could be easily avoided with only a modest amount of additional care mixed in with Bob’s comments on STAN. The people building coding languages like R, Python, STAN etc are adept at testing their work until the few flaws that might be remaining are almost always inconsequential. What would the academic literature look like if research scientists spent as much time testing their work as Bob spends testing the STAN code he writes?
Third – tying the first and second together – of course we don’t see silly “missing column” errors in usefull and serious science. When the science actually matters – that is, it produces a clear and useful result – the work is thoroughly tested. This is why social science is so riddled with garbage: nothing it produces matters – it’s so bad you can’t even tell if it does anything at all: so it doesn’t do anything at all. It’s a bunch of bullshit noise mining and quack methodology underlain by thousands of spreadsheet errors, coding errors, intentional bias and the king of all garbage science, millions of erroneous assumptions. The social sciences as a whole are a lot like the WOWSA and the AORBS: just a bunch of people grabbing at power in a completely irrelevant organization. The work they produce is their little more than their campaign propaganda.
Fourth, it doesn’t follow that R&R’s spreadsheet error and their ridiculous claim that it didn’t change their result went on to “damage to the world economy”. Now *that* claim – that their erroneous result went on to “damage the world economy” – is even more ridiculous than R&Rs claim that their error doesn’t matter to their result. :)
The reality is that, like most social science research, R&R’s numerical results are just more garbage statistical noise, one way or the other. Even if their result was wrong, suggesting that it went on to “damage the world economy” is like suggesting that a coach putting Curry into the game seven final for the last five minutes because he has a “hot hand” damages a teams chances of victory: it doesn’t matter whether the “hot hand” is real or not. It doesn’t matter what the results of any “hot hand” study turn out to be. We already know from experience that putting Curry in the game for the last five minutes of the championship final is just plain common sense. The same is true of reducing the nation debt: we know from experience that its common sense. That’s why we don’t leave these decisions to economists, statisticians, or social scientists.
Anon:
1. You might not think that the “Excel error” paper is “useful and serious science,” but Reinhart and Rogoff surely do, along with major chunks of the U.S. academic, government, and media establishment.
2. Let’s just split the difference and say that the Reinhart and Rogoff paper may have damaged the world economy. Any estimate of the effect of this paper is speculative, but I do think it’s possible that it had a large effect. You may think that government and corporate economic policies are decided just from “common sense,” but “common sense” can be just about anything, ranging from anarchy to communism and just about everything in between. Policymakers have to choose which common sense to follow. Recall that quote from Keynes: “Practical men who believe themselves to be quite exempt from any intellectual influence, are usually the slaves of some defunct economist. Madmen in authority, who hear voices in the air, are distilling their frenzy from some academic scribbler of a few years back.”
3. Continuing on this point, you write of reducing the national debt that “we know from experience that it’s common sense. That’s why we don’t leave these decisions to economists, statisticians, or social scientists.” Ummm . . . the national debt keeps going up! So whoever these people are to whom we’re leaving these decisions, they’re not following what you consider to be common sense.
4. I don’t see the relevance of your Curry argument. Sometimes Curry is hot, sometimes he’s not. But even a cold Curry is gonna be one of your best shooting options available. He’s far above replacement level. Keeping Curry in is a no-brainer, however the hot-hand thing works. The more relevant example for the hot hand would be whether to keep in a marginal player. And somewhere on the margin, yes, statistical studies can be relevant. In football they go for it on fourth down much more than they used to. And in baseball, they’ve had to change the rules because statistically-motivated strategies were making the game too boring. I think that it’s naive to say that research doesn’t matter. Even complete joke research can be taken seriously by powerful institutions.
P.S. For Bob’s sake, let me remind everyone that the programming language is Stan (named after Stanislaw Ulam), not STAN.
But any “large effect” was not from the large effect of an Excel error, but instead deliberate decisions they made.
> The same is true of reducing the nation debt: we know from experience that its common sense.
This belief is so common that it has been given a name: “The household fallacy”.
Thanks, Andrew. Still reading through the links and links from links.
I am a very new/mediocre open water distance swimmer. But enough in the world to have hung with people that swam around Manhattan, etc. Nobody takes Nyad seriously because if she really wanted to document a FL Straits crossing, she would have made ultra sure to have independent observers. There’s basically only one reason to do that serious of a swim and deliberately eschew strict verification. Bayesian BS detector off the chart.
I was also into cycling (not great either, but trained with people along the path). And I felt Lance (and the whole Peleton) was full of dope WAY before the general public did. And there were PLENTY of articles discussing his A/B samples and the like and proving doping well well before he finally fessed up. But the narrative lovers (e.g. Sally Yates, Nike, etc.) didn’t even want to hear it. Avert your eyes!
Groupthink is powerful. Lot of people prefer to be wrong on trend, than the converse.