Climate bet; Africa Cup of Nations

Summary

---

More or Less (BBC Radio 4)

Episode Title: Climate Bet; Africa Cup of Nations
Date: January 20, 2012
Host: Tim Harford

---

Episode Overview

In this episode, Tim Harford delves into two main topics:

• The outcome of a climate change bet between astrophysicist David Whitehouse and climate scientist James Annan, focusing on recent global temperature trends and how these are measured.
• Predictions for the Africa Cup of Nations football tournament using statistical models, featuring expert opinions and a bit of friendly disagreement.

The show maintains its signature analytical and witty tone, breaking down complex statistics for a broad audience.

---

Key Discussion Points and Insights

1. The Global Temperature Record Bet

Background

• Scientists and commentators have debated the interpretation of global temperature records, particularly after a seeming "pause" in new temperature highs since 1998.
• Four years prior, David Whitehouse and James Annan made a bet, aired on “More or Less,” about whether a new global temperature record would be set by 2011, based on UK Met Office Hadley Centre data.

How Are Global Temperatures Measured?

• Tim Harford visits the BBC rooftop weather station with forecaster Peter Cockcroft.
  • “This funny black funnel with a sort of white accordion thing at the bottom, that's our weather station.” — Peter Cockcroft (02:07)
• Dr. Peter Stott (Met Office) explains the global measurement system:
  • About 2,200 land stations, 4,000 ships recording engine intake temperatures, and 1,300 buoys monitor surface ocean temperature.
• Importance: Global temperatures aren’t measured in one place—they are aggregated across thousands of global sites.

The Bet – Results and Interpretation

• The UK Met Office data for 2011 showed it was not a record year.

Key Quotes & Discussion:

• David Whitehouse (Winner):
  • “I think it shows that we did enter a warm spell 30 years ago, and I think it shows that the rate of warming of that spell is not constant... for the past 10 years the temperatures remain constant.” (03:38)
  • He argues there is significant structure and variability in the warming trend.
• James Annan (Loser):
  • “I'd say there's absolutely no sign there's any change in the warming rate. The trend is still upwards. There's obviously some variability..." (04:33)
  • Blames the loss on “just bad luck, really.” (04:50)
• On Statistical Models:
  • “If we’d chosen either of the other two data sets, then I’d have won it.” — James Annan (04:53)
• Double or Quits?
  • The two agree to consider a new bet: Will there be a new record in the next four years? Whitehouse wants tighter criteria — not just a one-off record, but evidence of “sustained warming."
    • “I would like to see evidence of sustained warming which means more than one year, more than one standard deviation above the straight line of the past 10 years.” — Whitehouse (05:58)
  • Annan agrees, but notes that failing to break the record would “certainly start to point towards a slightly lower warming trend.” (06:32)

Summary Insight

• Both agree the decade just passed is the warmest on record, but disagree on whether short-term fluctuations matter.
• Discussion highlights the difficulty in interpreting climate trends based on short timescales, and how statistical choices can dramatically affect perceived outcomes.

---

2. Predicting the Africa Cup of Nations

Statistical Modelling for Football Predictions

• Rob Mastradomenico, a sports statistician from Swindon, discusses using complex statistical models to estimate team strengths using past international results.

Predictions (07:36):

• Top teams according to the model: Ivory Coast, Ghana, Tunisia, Morocco, Senegal.
• Surprise outlier: Burkina Faso is also flagged by the model.

Expert Rebuttal from the Field

• Farai Mangazi, reporting from Gabon, reacts with wit and skepticism:
  • “No, I think that computer is drunk. I mean, there's more chance of Burkina Faso sending a man to the moon than winning this tournament.” (08:34)
  • Not convinced by Ivory Coast, tips Senegal as the likely winner: “I think they've got the best all around team in this competition.” (08:57)

Memorable Moments

• Light-hearted banter about the challenges of basing predictions solely on statistics, and the unpredictability of sports tournaments.
• “Well, not at all. I'm sorry to be a wet blanket on Rob's mathematics. No, I don't think the Ivory Coast [are] good enough to win this.” — Farai Mangazi (08:20)

---

Notable Quotes & Memorable Moments

• “This funny black funnel with a sort of white accordion thing at the bottom, that's our weather station.” — Peter Cockcroft (02:07)
• “I think it shows that we did enter a warm spell 30 years ago... for the past 10 years the temperatures remain constant.” — David Whitehouse (03:38)
• “I'd say there's absolutely no sign there's any change in the warming rate. The trend is still upwards.” — James Annan (04:33)
• “No, I think that computer is drunk. I mean, there's more chance of Burkina Faso sending a man to the moon than winning this tournament.” — Farai Mangazi (08:34)

---

Important Segment Timestamps

• [00:12] — Introduction and program overview
• [01:34] — Playback of the original climate bet
• [02:07] — Weather station tour and temperature measurement explanation
• [03:35] — Results of the climate bet and ensuing discussion
• [07:19] — Introduction to Africa Cup of Nations predictions
• [08:07] — Farai Mangazi’s reactions and sports debate

---

Final Thoughts

The episode provides an engaging example of the complexity and nuance in using statistics to interpret both climate data and sports outcomes. It also showcases the balance between statistical rigor and real-world unpredictability, with experts offering differing perspectives—all delivered with the program's trademark blend of clarity, humor, and skepticism.

Transcript

[A]

Thank you for downloading from the BBC. For details of our complete range of podcasts and our terms of use, go to bbcworldservice.com podcasts.

[B]

You're listening to a download from the BBC. This is more or Less. We make an edition of the program for BBC Radio 4, but this is a special edition broadcast on the World Service. Hello and welcome to More or Less on the BBC World Service. We're your weekly guide to the numbers in the news and in life, and I'm Tim Harford. In this program, we'll be using a sophisticated statistical computer model to try to predict the results of the Africa cup of Nations. But first, climate change. Almost everyone agrees that in the past hundred years, the globe has warmed by about three quarters of a degree. That's not really what has provoked concern about global warming. Rather, it was a more dramatic increase in the past few decades, especially the 1980s and 1990s. According to data from the UK's Meteorological Office, the record for average global temperature was broken in 1983-1988-1990-1995, and most recently in 1998. But then over 10 years went by without a record. Some people think this is telling us something about the planet. It warmed and then it stopped. Others strongly disagree. David Whitehouse is an astrophysicist and former BBC science editor. And James Annan is a climate scientist based in Japan. And they made this bet on more or less four years ago.

[C]

I think probably by about 2011 we should see a new record, if not before.

[D]

So would you put money on that?

[C]

Yes, certainly. Yes.

[D]

Are you willing to take the other side of that bet?

[E]

Certainly am.

[D]

Should we say £100?

[E]

That sounds fine to me.

[B]

Now, the latest figures giving the average global temperature are out. In a moment, we'll reveal who won. But first, how do we calculate global average temperature? To find out, I went to a weather station perched on the roof of the BBC in London, tended by weather forecaster Peter Cockcroft.

[F]

This funny black funnel with a sort of white accordion thing at the bottom, that's our weather station.

[D]

Not very big. It's about the size of a modestly sized bucket.

[F]

Yeah. I mean, modern automatic weather stations are quite compact.

[D]

Is this the same thing as we have in the weather stations all around the world?

[F]

Very similar devices. Some meteorological services still have the traditional wooden screen with slats in, and the observer would go out every day, every hour, in fact, and open up the box, have a look inside, look at the thermometer and write down the temperature

[B]

Peter's weather station gave us. The temperature about six storeys above the streets of central London. But how do we move from that sort of reading to a global average? Here's Dr. Peter Stott of the Meteorological Office.

[G]

Well, to calculate a global temperature, you need to take many, many temperature records from around the world, both on the land and over the oceans.

[B]

How many?

[G]

Well, There are about 2,200 stations over the land and then over the oceans. It's a combination of about 4,000 ships which are reporting the temperature coming into the engine intakes. And then there's about 1300 buoys that measure the surface temperature of the ocean.

[B]

Now, for the results of the bet. It was agreed that it would be based on data from the UK's Meteorological Office's Hadley Centre. The figure for 2011 has just come out and we can reveal it was not a record year. David Whitehouse and James Annan are back.

[D]

David, congratulations, you've won the bet. What does this tell you?

[E]

Thank you very much. I think it shows that we did enter a warm spell 30 years ago, and I think it shows that the rate of warming of that spell is not constant. The 80s and the 90s, it was warming, but I think that it is structure in this data that shows for the past 10 years the temperatures remain constant.

[D]

And you think that goes beyond just looking at record years? Because I think we could agree that's not the only way to look at temperature data, right?

[E]

Oh, yes. That's by far not the only way to look at the temperature data. We live in the warmest decade of the instrumental era, the noughties were warmer than the 90s, which were warmer than the 80s. But that's not to say that you should only look at this just over a long time period, say 30 years.

[D]

So, James, you both agree that the most recent decade is the warmest decade on measure, but do you agree with David's analysis that over the last decade there's no sign of warming?

[C]

I'd say there's absolutely no sign there's any change in the warming rate. The trend is still upwards. There's obviously some variability. That means that if you look over a very short period of time, you can't tell very precisely what the underlying trend is, but there's no sign of any reduction in trend.

[B]

Why did you lose the bet, then?

[C]

Just bad luck, really?

[E]

I don't think so.

[C]

If we'd chosen either of the other two data sets, then I'd have won it.

[B]

But after you took the bet, James,

[D]

you did write a blog post that said that according to the Models, you were pretty confident that there should be another record year.

[B]

Is the model wrong?

[C]

I'm still fairly confident that we're due another record year quite soon. There's some inter annual variability, which means that you can't guarantee that any specific year is going to be warmer than the one before.

[E]

But surely, James, we're talking about global warming here. And the fundamental parameter of global warming is warming. And if you look at the past 30 years, you have two decades in which the temperature was warming, though not at a great deal in the 1980s. You have the last decade which it hasn't warmed. And I think you made the mistake of drawing a straight line between the beginning and end of that period and projecting it forward.

[D]

I hesitate to leap in at this point, but there is of course an obvious way to resolve this disagreement. We could go double or quits. Will we have a record year in the next four years? James, are you tempted?

[C]

I think there's every possibility that will happen, yes.

[D]

David, would you be tempted?

[E]

There is a possibility. But remember, if we extend the bet to nine or 10 years, then there is a chance due to statistical fluctuations that one of the years might be high just to fluke. So I would say yes, I'm up for double or quits. But I would like to see evidence of sustained warming which means more than one year, more than one standard deviation above the straight line of the past 10 years.

[D]

We'll nail you down over email about exactly what you are or are not going to agree to. James, if you lost this bet again, would that start to make you question what you believe about climate change?

[C]

I think it would start, yes. If the record temperature in 1998 isn't beaten in the next four years, then it would certainly start to point towards a slightly lower warming trend.

[B]

James Annan and David Whitehouse Allen has written on his blog that there's little sign of the acceleration in warming that most models had predicted. We'll keep an eye on if and when that changes. Now, the whole premise of more or less is that statistics can be very revealing. So revealing they might be able to tell us the results of the Africa cup of nations football tournament before the games have been played. Rob Mastradomenico is a sports statistician in Swindon in the UK and he's been doing some high end statistical calculations on this. Welcome, Rob. What have you been doing?

[H]

Hi, Tim. Well, I've got a complex statistical model which uses historical international results as an input and allows us to output the strengths of the teams who are competing very Good.

[D]

Well, who's going to win?

[H]

Let me just make the final calculations.

[D]

The results are coming in.

[H]

The clear best team is Ivory coast. And the next team coming out is Ghana.

[D]

That's not going to surprise anybody. Many people have them as favorites.

[H]

And the next three teams, Tunisia, Morocco and Senegal. The next team that's coming out is Burkino Faso, not considered one of the favourites.

[D]

Okay, well, that's what the computer tells us. Ferrari Mangazi, the voice of African sport, is joining us live from Gabon. Farai, is that a swimming pool I hear in the background?

[I]

It is indeed. I'm sitting by the swimming pool. It's a hard job this, but somebody's got to do it.

[D]

Well, you've heard what the computer says. Are you now confident enough to declare the Ivory Post winners of the tournament before it even starts?

[I]

Well, not at all. I'm sorry to be a wet blanket on Rob's mathematics. No, I don't think the Ivory coast good enough to win this. They've got very good players, but they don't have players that gel into a good team.

[B]

What about Burkina Faso then?

[D]

The, the surprise pick from the computer?

[I]

No, I think that computer is drunk. I mean, there's more chance of Burkina Faso sending a man to the moon than winning this tournament.

[D]

You know, Rob, your computer's drunk. What do you say to that?

[H]

Oh, it's had a long day. It's been doing a lot of work. I think one thing to take into account is it is using the historical results. Tournament format doesn't always result in the best team winning.

[D]

And Ferrari, who would you tip to be champions?

[I]

Well, I've got my money on Senegal. I think they've got the best all around team in this competition.

[D]

Ferrari Mangazi reporting for the BBC World

[B]

Service from Sonny Gabon and Rob Mastradomenico

[D]

from Not so Sunny Swindon in the uk.

[B]

Thank you both for very much. And you can send your questions and comments to more or lessbc.co.uk and you can download our podcast via our website bbcworldservice.com more or less. More or Less was presented by me, Tim Harford, the undercover economist at the Financial Times. The producer was Ruth Alexander and the editor was Richard Varden.

[A]

There are dozens of different podcasts now available from the BBC, including news, documentaries, science, business, arts and sport. For details of them all go to bbcworldservice.com podcasts.