I know it’s a long weekend, and you want fluffy stuff to read at the beach. I had a whole list, honest, but the cat ate it, so I have this one instead:
Tim Harford — Article — Cigarettes, damn cigarettes and statistics
Cigarettes, damn cigarettes and statistics
This is a short, accessible take on how a consumer of scientific information should approach the idea of correlation and causation. If all you know is the pithy phrase that the former is not the latter, then you don’t know enough to judge when there’s enough of the former to infer the latter.
These People Are Trying To Fix A Huge Problem In Science
The results of too many scientific studies aren't standing up to scrutiny. Here's how a group of scientists think they can (partly) change that.
Also short and accessible, although a little more technical, about the problem of replicating scientific findings.
Marcus Munafo, a professor of experimental psychology at the University of Bristol and one of the authors of the paper, told BuzzFeed News: "We're past the point where we can just highlight the problem and say how terrible it is. We need to think about ways in which we can improve the quality of what we do." Their suggestion is to make it much harder to declare that you've found a "statistically significant" result in the first place.
The proposed solution is about P-values, and the article explains in detail using an easy to understand dice example.
"Dealing with the size of the p-value fixes some things," [David Spiegelhalter] says. "But it’s not dealing with the most important issues." What would really help, he says, is if researchers distinguished between simply exploring, looking for interesting stuff – when a p=0.05 result is definitely worth noting – and confirming a discovery, which should need a much more stringent test.
This is causing a huge controversy amongst scientists. Consumers of the information science produces can be left wondering if we can or should trust any reported scientific findings.
It would be foolish to simply sweep aside all results of science (starting with the things we don’t like, of course) and refuse to “believe in it”. Science isn’t about faith or dogma. But it is difficult to know how to interpret what we are told in light of this.
Critical thinking skills are more important than IQ for making good decisions in life – Research Digest
Why some smart people make foolish decisions. By Alex Fradera
The study this piece is based on is not freely available, so judging just from this report, the basic idea is contained in the headline.
The researchers were especially interested in how these measures correlated with scores on an inventory of real-world outcomes, on which participants indicated whether they had experienced events ranging from the mildly bad (e.g. fined late fees for a video rental) to the more severe (e.g. acquiring a sexually transmitted disease). The avoidance of these kinds of experiences gives an indirect measure of wise, effective decision making, and the data showed higher IQ individuals did do better. However, high critical thinking was even more strongly associated with these real-world outcomes (even after factoring out IQ). So it’s possible to have a modest IQ and navigate life wisely, or to have a high IQ and make clangers that leave your peers shaking their heads. It’s a question of critical thinking.
Unfortunately, we don’t know how they measured that correlation. So if we read the first article and the second, what then should we make of this one?
The intrinsic value of choice: The propensity to under-delegate in the face of potential gains and losses | SpringerLink
Human beings are often faced with a pervasive problem: whether to make their own decision or to delegate the decision task to someone else.
This study is fully available, so you can read through it, know the methodology, see the way they arrived at their findings and read about similar studies using different methods.
Our results demonstrate that participants are willing to forgo rewards for the opportunity to make their own choices and hence to control their own payoffs. This preference was observed not only when faced with potential gains (in accord with Owens et al. 2014) but also when faced with potential losses. Moreover, our findings indicate that participants are aware of the (sub)optimality of their delegation choices, suggesting that they are also aware of the premium they are paying to maintain control.
This idea is fascinating to think about in a hockey context. GMs make choices all the time. And some of those are bad choices. Rather than the GM being stupid, as so many people are quick to say, perhaps they are either lacking in critical thinking skills, or they simply are willing to take a loss to maintain control.
One of the things that people often forget when analyzing hockey is that people are, in fact, people. There is a great deal of hesitation to include the social sciences in hockey analysis, and yet a player himself may be making this same kind of choice. She may choose to shoot the puck rather than pass it. He may choose to rush forward rather than let the play come to him on defence. To act rather than to not act, is to choose rather than to let someone else choose for you.
That behaviour in a person is contained within their measurable results on the ice. But when you move beyond what a person did (what the stats show) to who a person is (why they performed that way), you get into the messy world of subjective judgement, and I think this is where consumers of hockey analysis butt heads with the producers of the objective information. Consumers want the who a person is answers, even if they have to make them up themselves.
And that brings me to the meat of this meal. Those other things were just the appetizer. This is your main course, and it’s very filling, one of the best articles about sports analysis and statistics I’ve ever read.
On soccer analytics: Numbers aren't the problem (people are)
The value of analytics to professional soccer clubs should be obvious, but what is the value of analytics to everybody else?
Yes, it’s about soccer. Read it anyway, because the turning point in the world of soccer statistics as Jake Walerius asserts, is the advent of expected goals. This is right where hockey is at now.
As is the case with many advanced stats, xG is a way of assigning an objective numerical value to concepts already employed in even the most Neanderthalic argument in the pub. Liverpool deserved to win. Bollocks, United were better. If Rashford hadn’t missed that chance, we would’ve won. xG gives us, among other things, a way to objectively evaluate such claims.
As James Yorke, managing editor of the soccer analytics website StatsBomb, wrote to me in an email, “xG is an interesting line in the sand, because its adoption represented a before and after between which anyone could get interested, pull some numbers and get really actively involved [in analytics], and the not so trivial exercise of sourcing data and model building that came with xG.”
xG, and the more complicated models that are following it in hockey like Something Above Replacement and Game Score, are the thing that divides with finality the consumers from the producers. It turns out that how we consume the information produced now by analysts with better skills at math and statistics than us often upsets the producers.
Step away from the numbers for a minute. If I said to you, “Oh yeah, that Game of Thrones show, what a laugh. I love Joffrey,” you might actually get a little angry that my interpretation of the show was so far off yours. If you are the producers of the show, you might be outraged. Anyone who has ever even dipped a toe in media criticism knows this tale of internet wankery.
For hockey analysis the issue is even more difficult. We consumers are left wading through the concepts that are growing ever more difficult without a very sophisticated level of training ourselves, and a concern about where we should place our trust.
Soccer is ahead of hockey in terms of media adoption of statistics. And it’s not universally successful.
The problem with stats is their interpretation requires expertise, and the media outlets best exploiting the stats boom are often not the ones with that expertise. It’s not uncommon, for example, to see Sky Sports tweet a graphic listing, say, the five “best” passers in the Premier League in 2016-17, who as it turns out are the same five players who completed the most passes in the Premier League in 2016-17. But of course you don’t have to be Pep Guardiola to figure out the number of passes a player completes does not a good passer make.
I don’t actually know who Pep Guardiola is, but I take his point.
It isn’t just fans who consume the information, either directly or through the media, it’s also teams. And this article should be a cautionary tale to every producer of statistical models or fan who has lamented the lack of teams that “believe in” stats.
“And I [ESPN senior writer and Times columnist Gabriele Marcotti] think in some ways there’s a reason for that. I think when … data first became available there was a lot of what I consider bad data or meaningless decontextualized data, like, you know, distance covered or passing percentage or possession percentage and I think a lot of the managers looked at this and quite clearly, quite soon realized that this is kind of nonsense on its own. And so I think a lot of the more inquisitive ones, you know, they became naturally more skeptical towards some of the, what I would consider, more meaningful analytics that came in later.”
At this point, the article wanders down a path I’m not interested in: the search for a moral value to fan education about statistics in sports. This is the big fight over if it’s okay to not care about stats. Is it okay to “not believe”. I don’t think morality can be found here. I don’t think there’s a genuine should you can attach to the phrase “learn about Corsi”.
As Marcotti said, “I think everybody could benefit, at least from understanding it [analytics]. I think it’s also quite grotesque that, at the high end, there are football clubs who employ whole teams of analysts, but they’re kind of there as kind of like window dressing, you know, managers don’t listen to them. And again I don’t want to generalize, but there are a lot of situations like that, and I speak anecdotally.
I am a cynical person, and my expectation of the use of statistics by teams will fall prey to human impulses like the one that makes us willing pay to maintain agency, and we will see stats bent to say what the boss wants to hear.
We do this ourselves already, often without any venal motive. No matter how hard you try to avoid bringing your preconceptions and desires into the mix, you do it, and it affects your interpretation of the character of Joffrey as well as the meaning of an xGF%.
My conclusion is different from Walerius’s. He wants to tell fans how to integrate their consumption of stats into their consumption of the game. I was left thinking that what is needed is less searching for the moral imperative to educating the masses and more getting on with the job of educating those masses!
That might well be a job, like the one of distributing the stats to the masses in a form they can consume, best left to someone not a producer but rather a professional with the right expertise.
Summer is here, and while we are busy running up the T25U25 to the mysterious number one prospect, we are leaving the weekends open for long links-essays on hockey statistics or any other topic you want to talk about. If we like it, and we have the time to edit it and the space to feature it, we might put your FanPost on the front page.
Here’s some hints on how to make that likely: proofread your writing, and don’t use any copyrighted material or images you don’t have the rights to. Be entertaining, be yourself, and be about the Leafs, Toronto hockey or what’s it like to be a fan.
Most of the writers here at PPP started out writing FanPosts, me included. So give it a whirl.
If you are interested in writing for us, we’re going to say go write a couple of FanPosts first, anyway, so might as well.
A video of new guy Andreas Borgman working out and playing. The guy with him at the start is the goalie, so that’s why he’s a little bit skinny looking.