It was all bound to come to an end eventually. After two weeks of correct predictions the model got not one but two wrong in week 3. Let’s have a look at what was predicted versus what actually happened, and spend some time looking in a little more detail at the predictions beyond the point estimate of the scoreline. Yes, it’s time for statistics: it was bound to happen eventually.
This isn’t the first time that I’ve decided to have a look at the finish times for a running race: you can find previous analyses of the Glasgow Half Marathon and the Edinburgh Marathon. In this post, I wanted to have a look at an altogether different kind of race: the Tinto Hill Race. At 6.7km and 475m of climb, it’s by no means the biggest of hill races, but it was the first one I ever ran, so it makes sense to look at it first.
black-holes
I’ve recently been asked to give a short lecture about the physics of black holes as depicted in a science fiction movie — I can’t say any more about it yet, you’ll just have to bear with my being enigmatic for the moment. A twenty minute talk isn’t enough to do justice to black holes, though, and since I talk about them a lot, I figured a series of posts looking at what they actually are — as well as the fascinating history of how we discovered them, first in theory and then observationally — would be a good thing to have.
This is the latest update of a post I first write over a year ago which tried to answer the apparently simple question “how many gravitational waves have we detected?”. My last version is now a little out of date, so time for a quick update.
I've taken the scoring data from the last 26 years of Six Nations matches for a quick look at when each team tends to score their points.
Do some teams generally score early in the game?
Do some leave it last minute?
Each petal shows how many tries were scored in that 5-minute window, across the selected Six Nations seasons.
The dashed line marks half-time.
Drag the handles on the slider to change which seasons are included.
We’ve had another rollercoaster weekend in the Six Nations, so it’s time to follow-on from my predictions from last week and have a look first of all at how my statistical model faired, and also at what we might expect in the next weekend’s worth of fixtures.
Way back in 2017 I was trying to learn about a new statistical technique, hierarchical Bayesian inference, using a python package called PyMC. One of the tutorials back then illustrated how to use it to predict football scores, and I thought it would be fun to apply it to rugby. Back then I had a model which looked at historical scores and tried to predict the outcome of future matches just based on those (at some point I’ll write up a more detailed and technical post about how the model works), but this year I’ve been playing around with updating things, and taking into account things like the players actually on the teamsheet each week.
This summer I’ve written “my biggest day in the hills so far” or something of that sort more than in any previous year. To some extent this has been the choice of tough route, big days in Fisherfield and in Knoydart spring to mind. This weekend, however, undoubtedly has the right to this description. With the wind and rain of Knoydart, and the difficulty of Fisherfield and its river-crossings, this is likely to retain the record for a while as well.
This summer has been quite a big one for gravitational-wave announcements. I’ve managed to make it into the media once or twice as a result!
Imagine standing in the Albert Hall, and just before the performance begins, as the orchestra is tuning-up, you hear a perfect, virtuosic Mozart melody coming through the noise. Ten years ago, to the day, the world of astronomy (along with the rest of the world) was shaken as the first gravitational waves ever to be detected were measured with the newly-upgraded LIGO detectors. While they were still tuning-up. The gravitaitonal wave took just a fraction of a second to pass through the Earth, but in its wake it created a new field of astronomy.
Earlier in the year I wrote a post which tried to answer the apparently simple question “how many gravitational waves have we detected?”. Well, with today’s announcement of 128 new detections, it’s time to update my calculations.
Today we’re announcing that we’ve more-than-doubled the number of black hole collisions which we’ve observed using gravitational waves. We’re also releasing an enormous amount of data collected between May 2023 and January 2024 which measures tiny changes in the Universe’s geometry measured by gravitational-wave detectors, which are how we identify these collisions. We collect the information about these collisions into publications which we call a “catalog”, which contain all of the information we’re able to extract from the signals we detect and interpret. This includes things like the mass of the objects which were colliding, how fast they were spinning, and whereabouts in the Universe they were. We’ve published earlier versions of this catalogue before now: GWTC-1.0 has events from the first two observing-runs, GWTC-2.0 adds events from the first part of the third observing-run, and GWTC-3.0 events from the second part of that run. GWTC-4.0, which we’ve released today on the arXiv, is the latest version, analysing the first part of our fourth observing run (O4a). The amount of information in these has become so large that we now split the publication describing the catalog into three parts: an introduction paper, a paper describing the methods used in the analyses, and the results of those analyses. I’ve had the pleasure of leading a team which performed one of these analyses (the “parameter estimation” analysis), which takes signals which have been identified in the data and then works out the properties of the black holes and neutron stars which produced them. It took around 60 of us to do this (and there’s more about that later in this post). I then took-over as the editor of the results part of the catalog publication in June. It’s quite nice to know that I’ve joined a bit of a Glasgow tradition of doing this: my line-manager John Veitch was responsible for GWTC-2.0, and my colleague Christopher Berry for GWTC-3.0. I can only apologise to either of them if I ever thought they seemed to have an easy job!
I often find myself wondering why I climb hills, why I’m not prepared to enjoy normal holidays which involve pottering around pretty towns, lying on the beach, or visiting theme parks. I think my experiences of almost two weeks far north of the Great Glen, in the hills vaguely close to Ullapool, helped to answer that question for me: it’s clearly a feeling of achievement, of going somewhere only a tiny fraction of the population will ever go. Why I insist on having this feeling of achievement while also walking through the driving rain, however, is still an open question. But in the hours I spent in Fisherfield, alleged to be Scotland’s only real wilderness, I asked myself this question many, many times.
Last week I got an email from a colleague, asking a question which, on the face of it, should have been easy to answer.
It started as a nice, simple, short project to test some code I was working on. Things… got out of hand. Today I’ve submitted a paper describing a project I’ve been working on for around 18 months, which was not initially planned as a paper.
Lately I’ve been going through lots of little half-finished analyses and trying to write them up in some form. Today it’s my attempt to answer the question “how long does a river stay in spate?”. The primary motivation of this being that when the water level on the Clyde is high it’s not safe to row on it. Clubs, such as Clyde ARC might want to have an idea of how long it will be before the water level is low enough that it’s safe.
I seem to be incapable of just doing something without wondering about the statistics behind it these days.
A couple of weeks ago I posted some images from a new map I’d been working on. The motivation behind this was mostly curiosity, though I’ve been dabbling with trying to do things like this in python for a while, before realising that implementing hill-shading was probably more effort than it was worth, and turning to a tool actually designed for this: QGIS.
It’s been a while since I posted anything, and I have a big backlog of things to write-up and add to this site. In the meantime, enjoy some maps I’ve made (with the ultimate intention of eventally using them on my route descriptions on the site!).
In the years I’ve lived in Glasgow I’ve had the chance to experience quite a number of the big events which the city has a tendency to attract: The Commonwealth Games, The UCI Championships, COP-22, those sorts of things. I’ve also had the chance to talk to lots of people about my research, and the science which we do in my research group at the University of Glasgow. Over the last week I had an opportunity to talk to a lot of people at one of these big international events: The World Science Fiction Convention, or WorldCon as it’s known.
I started this post with the idea of summarising all of the new science which has resulted from observations made up to the end of the third observing run of the advanced LIGO and Virgo detectors. It turns out there’s rather a lot, so this looks like it’ll be the first of several! I’ve tried to include links to as many papers as possible.
I’m pleased to announce the release of the latest version of asimov.
I recently ran the Great Scottish Run half marathon in Glasgow, and finished in what I figured was a pretty decent time. But of course, this just leads on to another question: what is a “good” time in the half marathon? Of course, one of the beauties of a sport like running which is so individual is that a good time is really what you make it, and nowhere is that more true than a mass participation event like GSR. But again, before I dive into the statistics, a good time in the half marathon is clearly one that you’re happy with, and comparing yourself to others probably shouldn’t be why you’re running.
Every so often I get the notion that I ought to be taking a look at the performance of various rugby teams in various tournaments, and then trying to perform some sort of statistical analysis on it. Well, it looks like this might spin out into a series of blog posts in the run up to the 2023 Mens’ Rugby Union World Cup.