Blackpool

Posted on November 3, 2010 by bdunham

I still wake up some mornings thinking it was all a dream. After nearly three decades supporting a team that looked destined to spend a lifetime yo-yoing around the lower divisions of the English soccer leagues – and yes,  I have finally got used to saying “soccer” instead of “football” – the unthinkable happened.  My beloved Blackpool football club reached the dizzying heights of the EPL (that’s the English Premier League for those of you reading this in England for whom the acronym will be unfamiliar).  That’s a whole two divisions higher than I’ve ever seen my club play in the flesh, as  before I came to Canada in 2005 for the previous twenty plus years Blackpool fans had only the wildest of dreams of supporting a top flight club.

There was a feeling that the club might be on an upward trajectory once the mysterious Latvian Valery Belekon became involved with Blackpool in 2006. He came trumpeting that we’d be in the Premier League in five years, and we all smiled knowing that was about as likely as the town being awarded the Winter Olympics.  Yet there was real progress on the pitch, and under the canny managership of Simon Grayson the club stormed to an impressive play-off run in 2007, ending with ten straight wins to gain promotion to the Championship, the second tier of English football.

Now a friend of mine asked me about decade ago how high I thought Blackpool would rise during my lifetime. I confidently predicted that the Championship would be the upper bound. After all we’d never been as high as that since 1978, and we’d spent more time oscillating between the third and fourth divisions in the intervening years than challenging for promotion to the second tier. The one season the club had aspired to the promised land, the fateful 1995/95 campaign, we were cruelly twice denied: we floundered at the end of the season to miss an automatic promotion, then we somehow threw away a 2-0 lead in the play-offs by losing at home 3-0. It seemed the soccer gods didn’t like tangerine.

In 2007 Blackpool fans were sure we faced a tough life struggling to stay in the Championship. Singling out the handful of teams we could finish above would be an annual event at the start of each season. The top teams would thrash us, being vastly richer due to larger crowds, more generous benefactors, parachute payments or all the above. Yet we did survive the first season quite comfortably and the second too,  more impressive since Grayson left half-way through that campaign. His successor was shown the door at the end of that season to be replaced by an individual destined to be much vaunted in the history of the club: Ian Holloway.

Now Holloway’s arrival at Blackpool came without a fanfare from the fans; far from it in fact.  He’d been out of the game for a year after a disastrous stint at big-spending Leicester, had enjoyed some modest success in the past with Plymouth and was seen as something of loose cannon. His bizarre quotes were a source of mirth in the game, and he was hardly held in esteem as a great thinker or tactician. Yet he’d used his time out productively, carefully planning what he’d do differently if he got another chance. That his chance came at Blackpool was a sign the soccer gods were finally smiling on the club.

Holloway’s devotion to relentless attacking football came in stark contrast to much of the dour defensive drudgery I’d seen during the previous quarter-century of supporting Blackpool. He blended a skillful, well-organized team out of players who were mostly cast-offs from slightly bigger clubs, players who were hungry to prove their worth. Some of the football Blackpool played in Holloway’s first season was the best seen at the club since the sixties. Surely the club couldn’t achieve the unthinkable. After all, Blackpool had started as bookies and pundits’ favourites for relegation. Yet somehow luck, nerve and strength prevailed as Blackpool hit form just at the right time, and as so often happens, the team sneaking into the play-offs at the death rose triumphant. The play-off final at Wembley had so many memorable moments for Blackpool fans, the highlight being “Super” Charlie Adam’s free-kick swirling into the Cardiff goal:

YouTube Preview Image

The Premiership beckoned. Come on in, tiny minnows doomed to instant relegation, a pitiful points tally and eight months of drubbings. Come join the top table for a season, “little” Blackpool. We like your colourful fans, your quaint seaside tower, and making cliched references to the rollercoasters at your tacky “Pleasure Beach”. We like that you’ll be a lesson to all the other little clubs who dream of competing with the giants of the English game.

Blackpool commenced their first Premier League season as everyone’s favourites to finish bottom. Not one pundit or bookmaker I saw predicted Blackpool would end the season above a single other team. Luckily, neither Holloway nor his players had read the script. First game up, a 4-0 win away at Wigan. Yes, a 4-0 win. Away from home. That wasn’t in the script. Neither were subsequent wins at Newcastle and Liverpool. Blackpool have defied expectations, are playing well and continue to attack relentlessly.

My confidence that Blackpool will end the season relegated has taken a few knocks of late. But Blackpool fans are pragmatic bunch – we’re going to enjoy the ride wherever it takes us. Never have I followed a season more avidly (thank you, Setanta!), never have I celebrated Blackpool goals with more gusto. Blackpool football club have climbed their own tower, and they’re not looking to jump off just yet.

Posted in Uncategorized | Leave a comment

Multiple tests

Posted on November 3, 2010 by bdunham

I often receive questions about situations where multiple hypothesis tests have been performed by a researcher. That is, someone has presented an analysis in which lots of hypothesis tests have been performed. Some of these tests have turned out to be significant, yet there is a question mark over the reliability of the analysis.

By default I’m suspicious of any report that contains results from more than handful of different hypothesis tests. One issue is the problem of the so-called type I error rate. As taught in any introductory Statistics course, in a classical hypothesis test we can control the probability of incorrectly rejecting a null hypothesis when it is true. This risk is usually set at 5% when a single test is performed. That is, we would expect on average to incorrectly reject about one time in twenty a null hypothesis that is true.

Problems arise with the type I error rate when multiple tests are performed. By the above logic, if twenty independent tests were performed all on null hypotheses that were true, we would expect one null hypothesis to be rejected if all tests adopt the same type I error risk of 5%. Hence we would report one false positive response, incorrectly rejecting a null hypothesis.

One approach to addressing the problem described above is to reduce the significance level for each test, from say 5% to some lower value. This has intuitive appeal, and there are several ways to implement the idea, including the well-known Bonferroni correction. Yet problems persist: in situations where multiple tests have or could have been performed, on which set should one apply the correction? It is not legitimate to find a set of tests that give significant results, then apply Bonferroni’s correction post hoc. Moreover, only in rare cases are multiple tests independent of one another, so a key assumption is usually violated.

In some circumstances there would only be modest damage done by multiple testing, with or without the application of any corrective approach. Yet in certain situations the number of tests performed is so large that the reliability of any single p-value vanishes. In the mid-nineties I became involved in analyzing fMRI data. The studies looked at pixelated images of the brain over time as the subject responded to some kind of stimulus (such as wiggling fingers at the sound of a buzzer). The aim would be detect which areas of the brain had been stimulated, the equipment monitoring blood oxygenation levels over space and time.

The standard procedure for analyzing fMRI data had involved performing multiple t-tests, comparing levels pixel-by-pixel and applying Bonferroni’s correction. Of the several problems with this approach was the issue that in no sense could the tests be considered independent – obviously the blood level at one pixel depended in some way on the levels in neighbouring pixels, making it difficult to adjust the type I error rates for all the tests being performed. A flawed analysis can make futile the task of interpreting fMRI data; for example, Laura Sanders (in Science News, October 2009)  reported how an fMRI study on the brain of a dead salmon appeared to show the same level of brain activity in response to emotional images shown to the dead salmon as would be expected from a live human.

So what is the solution to the multiple-testing problem? Sad to say, there isn’t one. Even what is currently considered the most effective corrective approach, false discovery rate (Benjamini and Hochberg, 1995) fails to handle the type of multiple comparison problems thrown up in fields such as fMRI and searches for genetic markers. So while there may be multiple problems associated with multiple testing, there is no single fix.

Posted in Statistics | Tagged , | Leave a comment

The wonders of modern technology

Posted on November 3, 2010 by bdunham

Back in the early days of the internet, I made a couple of decisions about how I would use this new-fangled technology. Over the years I’ve steadfastly stuck by these decisions, despite the choices being contrary to conventions that are commonly accepted today. From time to time I’ve felt obliged to re-consider my choices.

Back in 1990, email seemed a neat and convenient alternative to writing letters. Although transatlantic email correspondence was plagued by the necessity of passage through some mysterious “gateway” or other, it seemed to work at least most of the time. As someone who enjoyed letter writing, I decided that for me an email would simply be an electronic form of what I would put in a letter to the person concerned, commencing with “Dear whoever” and subsequently following the style and form of a letter. At the time, there were no accepted conventions about how emails should look or read, and my approach seemed about as valid as any other. Of course these days my rather formal style has come to raise eyebrows – even my girlfriend recently politely queried me on matter.

When email was in its infancy I can well recall the frisson of excitement I felt when something popped into my inbox – perhaps a weekly event back in 1990. That someone had gone to the trouble of writing to me prompted me to put a decent amount of effort into composing my response.  These days, there’s little respite from the constant influx of emails, and just reading the darn things feels like a part-time job. Plus of course for me just to respond to the plethora of emails I receive each week – never mind composing a “letter” where a reply is appropriate – eats up great chunks of my time.

When websites were becoming ubiquitous back in the early nineties, I recall a senior colleague of mine confidently predicting they would be a “ten year wonder”. Who, he questioned, would have the time to maintain all these web pages? Over ten years on and the demise of the web page seems about as likely now as the end of the cellphone, and maintaining websites has become a routine part of many people’s jobs. Like it or not, www is here and going to be a part of lives for the foreseeable future.

With the web in its infancy I looked rather disdainfully at those who appeared to use their websites for what seem to me like shameless self-publicity. I refer here to academics not commercial ventures, as the latter of course would use the web to attract potential buyers and peddle their wares.  Somehow it struck me as unseemly that an academic would stoop to using a medium like the web to draw people’s attention to their work. Our community publicized valuable work in journals, books and at conferences and not by bill-poster style advertising on web pages, I thought.

So how do I view these decisions made back in the early nineties as we sit on the brink of 2011? Well I’m happy that my emails look and read rather differently from everyone else’s, so please don’t be put out if I respond to your three lines of twitter-ese with a formal response that was carefully proof-read, contains complete sentences (even paragraphs!) and starts “Dear …”.  Twenty years on from the web being a novelty, everyone and everything needs to fight for the attention of those who may be interested. Using a website for promoting research and other achievements does I’d admit appear today less distasteful to me today than it did two decades ago. So I’m going to relent, and one day when I get round to it I’ll stick some self-promotional stuff on my home page located via www.stat.ubc.ca. One day, that is, when I can squeeze in an hour or so amidst my email correspondence …

Posted in Uncategorized | Tagged , | Leave a comment