Posts from June 2025 (8)

• Ask Stuff’s BudgetBot anything you need to know about Budget 2025. Or, perhaps, don’t.  The budget is an occasion where there’s actually quite a lot of expert analysis both in mainstream media and social media, as well as all the unreliable vibes-based commentary anyone could want
• I’m not saying that AI is useless.  Mike Caulfield has some very impressive demonstrations of forcing Claude to help with fact checking.
• Various sources report that smartphones cause haemorrhoids.  This is based on unpublished research presented at a conference, of a sort that can’t show more than a correlation, and that just barely provides evidence of a correlation (this is the closest we have to details).  The problem is that the study has no ability to tell whether using a smartphone on the toilet increases the risk of hemorrhoids, or having hemorrhoids increases the chance that you’ll use a smartphone on the toilet, or that something else affects both things.  Also, it’s about time for this story to reappear, since we had it in 2023 and 2020  and 2018, though those times it was at least presented as expert opinion rather than research.
• RadioNZ have launched a political poll with Reid Research (hat tip to @danxduran on Bluesky).  Notably, the article describes not just the maximum margin of error, for proportions near 50%, but margins of error for smaller proportions such as 10% or 20% (not for 5%, unfortunately). These are uncertainty estimates for an idealised mathematical model of polling, and underestimate the true uncertainty a bit, but they are a big step forward.  I’ve written about the uncertainty for smaller probabilities on StatsChat before

There’s a new study of cancer and exercise that’s just been reported at a cancer conference in the USA and published in a major scientific journal, and which has made it to the media.  It’s good news; and actually real good news.

The study finds that exercise actually improves survival in colon cancer.  More precisely, it finds that providing an exercise coach improves survival compared to just providing the usual “exercise good; junk food bad” information.  Obviously this wasn’t a double-blind trial — you can’t make people exercise without them knowing — but it measured objective health outcomes. In particular, the medical profession can measure death very reliably.

There have been lots of papers in the past showing that exercise is correlated with better health in people with colon cancer. The correlation is robustly unsurprising: the less well you are, the less you are able to exercise, and there was no way to be confident anything more than this was going on.  This study was different, because people were randomly assigned to higher or lower pressure to exercise.  It’s pretty unusual to have a study that actually changes people’s level of exercise over a long period, and even more unusual to show that it actually improves their health. We don’t know if the effect translates to other cancers — previous studies have had hypotheses about mechanisms that are specific to colorectal cancer and others that aren’t.

Since this is StatsChat, I do want to compare what the research paper and the news said about the size of the effect. Here’s the graph from the research paper

At the planned 8-year follow-up point, the difference in survival was 7 percentage points. Basically the same was true at the planned 5-year point for survival cancer-free.  The overall survival difference narrowed a bit if you took the data out to ten years, and the cancer-free survival widened a bit.   You could also quote the average ratio of the death rates (or cancer recurrence rates) in the two groups, which is common in the statistical analysis of cancer but is a bit harder to translate into real-world impact (and which gives much bigger numbers 28% or 36% reduction)

The Guardian just reported the relative rates. The BBC reported both, very clearly.  Ars Technica reported both, but didn’t link the absolute and relative numbers as clearly as the BBC.

The Guardian also made a lot of the “better than a drug” comment by the chief medical officer of the American Society for Clinical Oncology,

“It’s the same magnitude of benefit of many drugs that get approved for this kind of magnitude of benefit – 28% decreased risk of occurrence, 37% decreased risk of death. Drugs get approved for less than that, and they’re expensive and they’re toxic.”

I think it’s worth noting that this is not saying exercise is better than chemotherapy. It’s saying exercise plus chemotherapy is better than chemotherapy along, and the margin is large enough that if it were new drug+chemo vs chemo alone you’d easily get approval.

Team Ratings for the Semi-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 148 matches played, 105 of which were correctly predicted, a success rate of 70.9%.
Here are the predictions for last week’s games.

Predictions for the Semi-Finals

Here are the predictions for the Semi-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.

Team Ratings for Round 26

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 175 matches played, 133 of which were correctly predicted, a success rate of 76%.
Here are the predictions for last week’s games.

Predictions for Round 26

Here are the predictions for Round 26. 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.

Team Ratings for the Quarter-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 77 matches played, 55 of which were correctly predicted, a success rate of 71.4%.
Here are the predictions for last week’s games.

Predictions for the Quarter-finals

Here are the predictions for the Quarter-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.

Team Ratings for the Semi-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 90 matches played, 62 of which were correctly predicted, a success rate of 68.9%.
Here are the predictions for last week’s games.

Predictions for the Semi-finals

Here are the predictions for the Semi-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.

Team Ratings for Round 14

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 100 matches played, 54 of which were correctly predicted, a success rate of 54%.
Here are the predictions for last week’s games.

Predictions for Round 14

Here are the predictions for Round 14. 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.

Team Ratings for Week 14

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 105 matches played, 66 of which were correctly predicted, a success rate of 62.9%.
Here are the predictions for last week’s games.

Predictions for Week 14

Here are the predictions for Week 14. 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.