Stats Chat

The Australian postal survey on marriage equality was a terrible idea.

It was a terrible idea because that sort of thing shouldn’t be a simple majority decision.

It was a terrible idea because it wasn’t even a vote, just a survey.

It was a terrible idea because it wasn’t even a good survey, just a bogus poll.

As I repeatedly say, bogus polls don’t tell you anything much about people who didn’t vote, and so they aren’t useful unless the number voting one particular way is a notable proportion of the whole eligible population. In the end, it was.

A hair under 50% of eligible voters said ‘Yes’, just over 30% said ‘No’, and about 20% didn’t respond.

And, in what was not at all a pre-specified hypothesis, Tony Abbott’s electoral division of Warringah had an 84% participation rate and 75% ‘Yes’, giving 63% of all eligible voters indicating ‘yes’.

PS: Yay!

Each week, we would like to invite readers of Stats Chat to submit nominations for our Stat of the Week competition and be in with the chance to win an iTunes voucher.

Here’s how it works:

• Anyone may add a comment on this post to nominate their Stat of the Week candidate before midday Friday November 17 2017.
• Statistics can be bad, exemplary or fascinating.
• The statistic must be in the NZ media during the period of November 11 – 17 2017 inclusive.
• Quote the statistic, when and where it was published and tell us why it should be our Stat of the Week.

Next Monday at midday we’ll announce the winner of this week’s Stat of the Week competition, and start a new one.

(more…)

Each week, we would like to invite readers of Stats Chat to submit nominations for our Stat of the Week competition and be in with the chance to win an iTunes voucher.

Here’s how it works:

• Anyone may add a comment on this post to nominate their Stat of the Week candidate before midday Friday November 10 2017.
• Statistics can be bad, exemplary or fascinating.
• The statistic must be in the NZ media during the period of November 4 – 10 2017 inclusive.
• Quote the statistic, when and where it was published and tell us why it should be our Stat of the Week.

Next Monday at midday we’ll announce the winner of this week’s Stat of the Week competition, and start a new one.

(more…)

Seeing a headline like

might cause an unwary person to think that eating purple kumara would reduce their risk of colon cancer by seventy-five per cent.

You, of course, would be suspicious and would want to read the story.

He found that when fed to three generations of mice bred with colon cancer, using the same gene which caused the disease in humans, purple kumara reduced the number of polyps by two-thirds or more.

So, the study is in mice. Mutant mice.  And it didn’t reduce the risk of colon cancer in these mice — which was basically 100% — it reduced the number of developing tumours.

It’s true that the mutation is one that occurs in people, too. About one in ten thousand people is born with the mutation that the mice had — these people have the mutation in every cell in their bodies, and they all get colon cancer if they don’t have major surgery.  And in the majority of ordinary people who get colon cancer, part of the  process is a mutation in this same gene in one cell.  So, the mutant mice are relevant.  There isn’t any problem with the research being in mice, just with the headline. Especially as further down in the story we hear about the equivalent dose of kumara in humans

“To eat 1kg of sweet potato every day is too hard.”

and that the kumara seems to have most potential as a way to produce a concentrated extract.

So far, there’s not much evidence either way on whether anthocyanins (basically, purple food other than beets or dragonfruit) really prevent cancer in humans.  Animal studies such as this one give good reasons to be hopeful; the history of other micronutrient-based prevention trials give good reasons to be skeptical.

From the MetService rain radar

If the band of rain were moving north-east, small uncertainties in its motion and orientation would mean that you’d know there would be half an hour of rain in Auckland, but not exactly when.

If it were moving south-east (as it is), small uncertainties in the motion and orientation mean that you know it will rain for a long time somewhere, but not exactly where.

One way to communicate the difference between these two predictions would be to show a set of possible realisations of rainfall.  For NW movement, you’d get a set of curves each with a single hump but at different times. For SW movement you’d get a much wider range of curves, where some showed no rain and others showed half a day or all day. I don’t know enough about ensemble forecasting to be sure, but I think this would be feasible

In principle, the common ‘patchy torrential downpours’ Spring rain pattern would show as rain curves each with different short periods of rain. I don’t think the technology is up to that using genuine predictions, but it might be possible to predict that we’re going to get that sort of weather and simulate the ensemble curves.

Current forecast summaries are mostly (except for hurricane paths) about averages: the probability of rain,  the expected amount, the worst-case amount. As technology progresses we will increasingly be able to do better than averages.

The StatsNZ twitter account has been tweeting creatively decorated graphs.  The first one I noticed was

which prompted a discussion on Twitter about why the graph actually still worked well even though you’d normally want to avoid this sort of thing.

Then, a few days ago, I saw this one via Harkanwal Singh

Unlike the cars, which worked as labels, the motorbikes don’t do anything except provide a distraction. They’re a bit better at a smaller size, where they reinforce the local trends in the graph, but I still don’t think they’re a net positive

Then today (again via Harkanwal)

Yes, ok, I get that it’s a joke. But pumpkins prices rise  at this time of year in New Zealand because it’s Spring. Halloween isn’t a big driver. And most of our pumpkins aren’t even orange (which is why Stuff has a story on alternative things to carve). And winter isn’t a single point in time. And the decoration distracts from the potentially-important observation that prices didn’t really drop last winter.  And the vertical axis doesn’t say what the units are (average retail price per kilogram, it turns out).

And… just no.

Figure.NZ has the version you want if you’re after information.

A rugby team that has won a lot of games this year is likely to do fairly well next year: they’re probably a good team.  Someone who has won a lot of money betting on rugby this year is much less likely to keep doing well: there was probably luck involved. Someone who won a lot of money on Lotto this year is almost certain to do worse next year: we can be pretty sure the wins were just luck. How about mutual funds and the stock market?

Morningstar publishes ratings of mutual funds, with one to five stars based on past performance. The Wall Street Journal published an article saying (a) investors believe these are predictive of future performance and (b) they’re wrong.  Morningstar then fought back, saying (a) we tell them it’s based on past performance, not a prediction and (b) it is, too, predictive. And, surprisingly, it is.

Matt Levine (of Bloomberg; annoying free registration) and his readers had an interesting explanation (scroll way down)

Several readers, though, proposed an explanation. Morningstar rates funds based on net-of-fee performance, and takes into account sales loads. And fees are predictive. Funds that were good at picking stocks in the past will, on average, be average at picking stocks in the future; funds that were bad at picking stocks in the past will, on average, be average at picking stocks in the future; that is in the nature of stock picking. But funds with low fees in the past will probably have low fees in the future, and funds with high fees in the past will probably have high fees in the future. And since net performance is made up of (1) stock picking minus (2) fees, you’d expect funds with low fees to have, on average, persistent slightly-better-than-average performance.

That’s supported by one of Morningstar’s own reports.

The expense ratio and the star rating helped investors make better decisions. The star rating and expense ratios were pretty even on the success ratio–the closest thing to a bottom line. By and large, the star ratings from 2005 and 2008 beat expense ratios while expense ratios produced the best success ratios in 2006 and 2007. Overall, expense ratios outdid stars in 23 out of 40 (58%) observations.

A better data analysis for our purposes would look at star ratings for different funds matched on fees, rather than looking at the two separately.  It’s still a neat example of how you need to focus on the right outcome measurement. Mutual fund trading performance may not be usefully predictable, but even if it isn’t, mutual fund returns to the customer are, at least a little bit.

• From Politico:  “Is Washington Bungling the Census?”

• From Wired: China’s planned ‘Social Credit’ score

• Florida man arrested for meth that was actually donut glaze.  Georgia police ‘drug recognition expert’ can recognise drug use that even blood tests can’t detect. In NZ, though, 95% of those who fail a roadside impairment test are positive on follow-up blood tests.

• A McDonalds promotion in Canada advertises some high-level prizes and a 1 in 5 chance of winning. Some guy bought 100 orders of large fries, thinking he’d get forty prizes (2 tickets per order).  It’s actually two half-tickets per order, so he won 23 prizes. Mostly cheeseburgers. The moral: (a) read the instructions, and (b) most of the prizes are always just cheeseburgers (or the moral equivalent in other lotteries).

• rawgraphs.io is a new tool for producing fairly attractive graphs quickly from spreadsheet data

• Mark Hanna wrote a guide to the New Zealand Official Information Act

• Andrew Gelman: Advice for Science Writers

• [update] I nearly forgot Chris McDowall’s graphs of electorate vs party vote in the NZ election, from the Spinoff

• Computer maps: then

Computer visualization of population density in Los Angeles County, 1971 https://t.co/IIAD7xjiGp pic.twitter.com/BkOXgbaXrI

— Eric Fischer (@enf) October 14, 2017

and now (from Wikipedia)

Each week, we would like to invite readers of Stats Chat to submit nominations for our Stat of the Week competition and be in with the chance to win an iTunes voucher.

Here’s how it works:

• Anyone may add a comment on this post to nominate their Stat of the Week candidate before midday Friday November 3 2017.
• Statistics can be bad, exemplary or fascinating.
• The statistic must be in the NZ media during the period of October 28 – November 3 2017 inclusive.
• Quote the statistic, when and where it was published and tell us why it should be our Stat of the Week.

Next Monday at midday we’ll announce the winner of this week’s Stat of the Week competition, and start a new one.

(more…)

Team Ratings for the Mitre 10 Cup Finals

The basic method is described on my Department home page.

Here are the team ratings prior to this week’s games, along with the ratings at the start of the season.

Performance So Far

So far there have been 74 matches played, 51 of which were correctly predicted, a success rate of 68.9%.
Here are the predictions for last week’s games.

Predictions for the Mitre 10 Cup Finals

Here are the predictions for the Mitre 10 Cup Finals. The prediction is my estimated expected points difference with a positive margin being a win to the home team, and a negative margin a win to the away team.