← More or Less
More or Less
Ash Die Back and Fergie Time
00:27:31
This is the first in the new series of the programme. Tim Harford has been busy some...
Loading summary
Transcript100 lines
- [00:00]
A
Thank you for downloading More or Less from the BBC. Here's Tim Harford. Hello and welcome to a brand new series of More or Less your weekly dose of numbers in the news. I'm Tim Harford and if listening figures were votes, we'd be police and crime commissioners by now. This week we'll be felling a few tree based statistics as we look at ash dieback. We'll be revisiting Trumpton to check up on Dibble and asking if Manchester United do get a little extra Fergie time at the end of matches compared to other teams.
- [00:34]
B
You're sitting there going, okay, there's a couple of subs, a couple of goals, that's wasted time, injury there, you know, three minutes, maybe four, you find yourself saying five. So where did that extra minute come from?
- [00:43]
A
While we've been off air, a mass killer has been found stalking the land. Killara ash diebach. 20 years after the first confirmed case was recorded in Eastern Europe, the disease has arrived in the British countryside. It was found in woodlands in East Anglia last month and since then, every time I open a newspaper or switch on the television, it seems to have spread. Ash dieback Number of affected counties doubles Ash dieback beyond control ash dieback 100,000 trees fell. To combat killer fungus, Ruth Alexander has been trying to keep up with this fast paced story. Hello, Ruth.
- [01:20]
C
Hello.
- [01:20]
D
And it's been quite easy to actually, because the disease isn't spreading like wildfire, we're finding more and more of it because we've started looking for it. The Forestry Commission and the Woodland Trust have had their SWAT teams out in the countryside. So surveying about 1,000 sites which have received saplings from nurseries where ash dieback's been found, they're also sampling two and a half thousand areas of countryside each 100 square kilometres. And when they find a diseased tree, a dot goes on the map and
- [01:48]
A
each dot is a new sighting, but it's not necessarily a new infection.
- [01:51]
D
Yes. The head of conservation for the Woodland Trust, Austin Brady, told me that some of the new dots represent disease that's actually been there for up to two years.
- [02:00]
E
As the dots on the map have sort of increased over the last few weeks, I think people have mistakenly assumed that that's a sign of the disease spreading rapidly. I think the reality is that it's a sign of the survey having been gradually rolled out across the country and the more dots on the map are just helping us to build a much better snapshot of where the disease is now and are not telling us Anything at all about how quickly the disease is spreading?
- [02:23]
D
Am I right to understand that actually at this time of year, the disease doesn't spread?
- [02:28]
E
That's correct, yes. The disease is spread by small spores that come out of the fallen leaves and parts of the stems of the leaves that have fallen to the ground. And those spores only really spread between about June to September, October. So at this time of year, there are no spores actively spreading. So the sense that the disease is popping up in other places is really because people are now more vigilant. They're looking for the signs and they're putting the dots on the map to give us a snapshot of where it is today.
- [02:55]
A
So we'll only begin to get an idea of how fast this disease is spreading when a second snapshot is taken?
- [03:00]
D
Yes. So in about a year's time, Austin says. But they do know from continental Europe's experience that on average, the disease can spread at a rate of 20 to
- [03:08]
A
30 kilometres a year, and they're worried they've felled 100,000 trees. And, Ruth, you can tell Austin if more need to go, I'm ready to do my bit.
- [03:17]
D
Tin. Put it away. There's a pair of scissors here, that'll do.
- [03:21]
E
If you could imagine something that looks more or less like a twig, about 18 inches long with some roots on the bottom, that's the kind of scale that we're talking about. And those kinds of saplings, the bulk of those that have been destroyed.
- [03:32]
D
So it might be more accurate to say 100,000 twigs have been pulled out of the ground.
- [03:38]
E
That's almost going to the other end of the scale, I guess. But I guess the most accurate thing to say would be that 100,000 saplings have been destroyed in nurseries or. Or removed from recent planting sites.
- [03:50]
A
So ancient ash trees aren't falling around the country already, but our landscape could change significantly. We know from the continent how bad it can get.
- [03:57]
D
Yes. Denmark is commonly held up as the country to watch. It's been dealing with the disease for some years. And one of the specialists at the Danish Nature Agency is Christina Dieter Ulrich.
- [04:08]
A
When we go to our seed orchards, which are grafted trees, we can see that approximately 90% of the trees are affected. And based on that and the observations in the forest stands, we give an estimate that approximately 90% of the Estrees here in Denmark are affected by the disease. And that's a statistic that the media has completely run away with.
- [04:30]
D
Yes. This stat that 90% of ash trees in Denmark are diseased has become a kind of poster stat for the ash dieback story. The media here is fixated on Denmark.
- [04:41]
A
The worst case scenario is always very tempting.
- [04:44]
D
It is. But Austin Brady from the Woodland Trust says the experts are taking a broader view.
- [04:49]
E
Looking across Europe, the range of figures in terms of how many trees have been affected and potentially will die does vary from round about 60% up to the 90% that's been quoted for Denmark. So it's more accurate to say that on average, 70% of trees across Europe have been infected. The other key bit of information we need to get a better understanding of is what proportion of the ash trees in different parts of Europe are resistant to the disease. We've heard figures that it may be as low as 2% in the UK, based on the experience of Denmark. But reports from parts of Poland suggest that they may have found up to 20% of the ash population to be resistant to the disease.
- [05:28]
D
Do you think to some degree the problem has been exaggerated?
- [05:32]
E
I think it's a situation where people are looking partly at the worst possible scenario, but also jumping to sort of some conclusions early on as to how the disease will behave in this country. And I suppose in a way one of the risks with the way the media handles the story is if they paint such a really awful black and bleak picture, then people give up hope of doing anything positive and that would be a shame.
- [05:53]
A
So the bottom line is that experts are still figuring out the scale of this problem, both in the UK and beyond. And the media has been reporting that 30% of Britain's trees could be affected.
- [06:04]
D
Yes, and that's not true because the disease only affects ash and actually only 5% of Britain's trees are ash.
- [06:12]
A
Ruth Alexander. Thank you. You might be wondering whether there's been any upturn in fortunes for the Trumpetonshire economy while we've been away, but I'm pleased to say things are looking a little brighter. But there's been some bad news. The International Monetary Fund, the imf, has announced that it's made errors when forecasting the growth of a number of economies. And more to the point, the errors centre around a particular question. When governments cut spending or raise taxes, how badly does that affect the economy? If the IMF's right, austerity measures are doing more harm to the economy than we expected. And this is a bit of a blow, of course, because as you may remember, in my part time role as treasurer of Trumpetonshire County Council, I've had to make tough decisions. I decided to sack Dibble the Firefighter Times were tough. And anyway, the Trumpton Fire Brigade had never put out a fire. In the last series, we caught up with Dibble and desperate times meant he was trying his hand at a bit of adult modelling. But is this paying its way? Hello, Dibble. Are you still selling those pictures? No, no luck. Everyone seems to think it very funny. You would have thought pictures of a scantily clad former firefighter would sell well.
- [07:30]
F
Yes.
- [07:30]
A
It's a pity the people who come here don't think that way. Can you not get a proper job? No, it's been like that for many years. I think many years is an exaggeration. Dibble, you were only sacked last year. But if you're not making much money, what are you living on? All my savings. Well, I'm terribly sorry, Dibble. I hope you realise this isn't my fault. This is out of my control. I did write to you explaining everything. I've never seen such rubbish. It's a bit strong. Due to your silly mistakes. This is a terrible mess. Oh, dear. Crowd has started to gather outside. They're not happy. I think I need to call my trusted political advisor at the Town Hall. Hello, Trumpeton, Town Hall Town clerk speaking. Hello, Mr. Troupe, Tim Harford here. I have a problem. The IMF are saying that all this austerity might actually be making things worse.
- [08:16]
F
That's awkward.
- [08:17]
A
They're on that side and we're on this. It does look that way. They're not backing us in quite the way they were before. What shall we do? Every time we have trouble like this, we seem to phone up one man for help. Ah, yes. Chris Giles, Economics editor of the Financial Times. Chris is an expert on the Trumpetonshire economy and in the past he's been skeptical of claims that austerity is the explanation for our economic woes. Hello, Chris. What's the argument about here?
- [08:45]
F
The basic question is, what is the effect of spending cuts, public spending cuts or tax increases? How much do they wipe out of the size of the economy? Because we know for a fact that if you stop paying, someone like Dibble, you sacked him a while back, that cuts his income. And then we don't know how much it cuts his spending. And does this then pass on from one person to the next? Someone's income falls, that cuts their spending, that means someone else's income falls. So actually, for the economy, it has a very big effect. Or does it really stop at Dibble? You cut his income, his spending falls a bit, but maybe not that much. He maybe gets another job and has very Little knock on impact. So the multiplier tells you how much the initial change in public spending or taxation affects the economy as a whole.
- [09:38]
A
And the larger the multiplier is, the more serious the impact of austerity on the economy. Now, the IMF recently came out with a statement on multipliers. What do they say?
- [09:50]
F
Well, what they said was that previous estimates were too small. Previous estimates in most countries were about a half. That meant that for every pound one you cut public spending, the overall size of the economy fell by 50p. And they said, no, that's too small. And the numbers is somewhere between 0.7 and 1.7. If it's 1.7, that means for every pound you cut out of public spending, the size of the economy falls by 1 pound 70. And therefore that has a much bigger effect. And so if the IMF is right, then it means that austerity is going to be much more painful and might. This is a. Might even be counterproductive. You might cut the size of the economy so much that ultimately debt and borrowing rises.
- [10:37]
A
But even if you don't, I suppose the argument would be you're doing more damage than you have to. You might want to cut spending on some different path. Two questions. Do you believe the IMF with their revised estimates of the multipliers, and do you believe the way they went about making this calculation?
- [10:55]
F
I think it's very hard to say. I think there are plausible arguments for why you might think the multiplier is higher than it would normally be. And there are two main arguments for this. One is that interest rates in most advanced economies are zero. So monetary policy or interest rates can't offset fiscal tightening. So if you cut spending or raise taxes, you can't have an offsetting cut in interest rates to boost the economy. So that might make the effect bigger, but it's not certain. People don't agree on that. And the second argument is that if all countries are simultaneously cutting spending or raising taxes, then you can't effectively export some of the difficulty to other countries because they're doing it as well. The exporting works such that some of the spending of Dybl, for example, would go on imports. So if he bought a German car, if you sack him and he can't buy a German car, it's Germany that suffers. It's their economy that suffers, not the British economy or the Trumpeton economy.
- [11:55]
A
But in this case everybody is going through some sort of process of austerity at the same time.
- [11:59]
F
Yes. So the effect is likely to be
- [12:01]
A
bigger The IMF released a statistical analysis supporting their conclusions. What did you make of the IMF's work?
- [12:07]
F
Well, I wasn't very impressed because I didn't think it was robust to changing the number of countries they used in their statistical analysis or the year, the time period over which it. Which if it was a general result, it should be robust to all of these sorts of changes. And I went in, I replicated their numbers so I could get their result, and then I changed the number of countries you used because it wasn't obvious why they'd chosen the countries they'd chosen.
- [12:33]
A
So you took some away or you added some more countries?
- [12:36]
F
I took some away and added some in. I just tried to use as much data as there was on the IMF's website. And then because they based their results on forecasts made in 2010, I said, well, if this is a general result, the same result should apply if you take the forecast they'd made in 2011. And unfortunately, I think that's a much better test because the results just fall apart on that second test. And so it was a very specific result to the forecast the IMF made in 2010. And I don't think you can make a conclusion about the size of the fiscal multiplier based on some forecast the IMF made in spring of 2010.
- [13:15]
A
So let's be clear. Are you saying the IMF are wrong to say multipliers could be 0.7 or as high as 1.7, or are you just saying, well, they may be right, but we have no proof the latter.
- [13:25]
F
They might be right, but they certainly haven't proved it with this analysis.
- [13:29]
A
So, as treasurer of Trumpetonshire yet again, Chris Giles, you are providing me plausible deniability. I can say we don't know and I refuse to take the blame.
- [13:39]
F
Absolutely, there's no reason you should fall on your sword yet.
- [13:42]
A
Chris Giles of the Financial Times. I should say some other economists are far more convinced of the case that austerity is damaging. That, of course, is exactly why I called. Chris. You're listening to More or Less, and I'm Tim Harford. In the recent outbreak of deadly violence between Israel and Gaza, the weapons have included air to ground missiles, naval artillery, rockets and a concealed bomb on a bus. They've also included numbers. On Tuesday, for example, Avatar Leibovich of the Israeli Defense Force tweeted, unbelievable but true. One hundred and eleven Israelis wounded today. As we so often find on More or Less. While all eyes are on the number, you should really be paying attention to the definition. It's true that on Tuesday, 111 Israelis were treated by Israel's national ambulance Service. Actually, the final number was 112, but it's the definition of wounded that may surprise some loyal listeners. The ambulance service was quite clear. Three people were injured by shrapnel, 27 hurt themselves running for safe rooms and 82 were treated for shock. We don't have comparable figures for civilians in Gaza. Whether we're counting the cost of war or evaluating the effectiveness of the National Health Service, statisticians often find it's convenient to focus on deaths not because lesser injuries or psychological injuries don't count, but because death is rarely an ambiguous condition. Between the latest outbreak of hostilities on 14 November and the ceasefire on 21 November, five Israelis have died, including three civilians. Gaza medical officials say that 162 people have died in Gaza, more than half of them civilians. The exact numbers haven't been independently verified. If you've spotted a statistic you'd like us to think about or have comments or questions about the show, please email us at more or lessbc.co.uk. our website and the place to listen again and download the show is BBC.co.uk more or less while we're off the air. This doesn't stop us searching for dodgy statistics. Most of the time we find small errors and we just throw those back in the sea. Occasionally, we find bigger fish to fry, but what we found this time is the great white whale of maths mistakes a number so wrong it's staggering. The Daily Telegraph recently carried this. Only 100 COD left in North Sea. A dim future for cod, you might think. Now, this isn't a trivial issue over exploitation, and conflicts over fisheries cause major problems worldwide. So we ask Brian Buchan what he thought. He's a trawlerman who's been fishing in the North Sea for decades. It made me so angry. In 34 years in North Sea fishing, we've seen more cod this year than we've seen in the last 30 years. Then we get a headline from someone who probably doesn't even know what a cod looks like. This makes my blood boil. Our cod correspondent, Hannah Barnes, has been trawling for the truth. Hello, Hannah. Let's start with a headline, 100 COD left in the North Sea, from the Daily Telegraph. Could it possibly be true?
- [16:48]
G
Quite simply, no. And it turns out that the Telegraph took the story from the Sunday Times, which claimed slightly different that there are a hundred adult cod in the North Sea. A different claim, but still wildly wrong. And here's what happened. The Sunday Times got the story by looking at data from the International Council for the Exploration of the Sea. And it asked researchers from the British Government's Centre for Environment, Fisheries and Aquaculture Science, or cfas, for help with the numbers. But to quote cfas, the journalists misunderstood the data. For some reason, they decided to class an adult cod as a cod over the age of 13. In fact, that's not merely an adult cod, it's an ancient cod. And we shouldn't be surprised that there are very few COD over 13 because fewer than 60 have been recorded in the last 30 years. Just as we aren't surprised that there are hardly any humans older than 100. Here's Dr. Carl O' Brien from CFAS.
- [17:47]
A
COD start to mature at ages one and two and they're fully mature by age six. The bulk of the mature cod in the North Sea at the moment is constituted of fish which are actually younger than seven years of age. So where did the newspaper get the idea that cod reach adulthood at 13? In the same article, the Sunday Times states that cod can live to 25, which implies that a 13 year old cod is merely middle aged. It's looking forward to paying off its mortgage and getting its kids off to university.
- [18:14]
G
Well, leading marine ecologist Dr. Tom Webb thinks the journalist got that 25 figure from a website called Fishbase.
- [18:21]
A
Hannah, don't tell me that Fish Base is not a reliable source of fish related information. It would be a tragedy.
- [18:29]
G
Fear not, Tim, Dr. Webb says fish base is usually accurate, but the idea that cod in the north sea live to 25 is, well, a bit of a red herring.
- [18:39]
A
For each species it has a front page with some headline statistics. For instance, maximum age, which for cod is given as 25. But when you delve into where those statistics come from, it's all very well referenced in Fish base. And you can actually find that the estimate of 25 was not from a North Sea cod, it was from a cod in the Barents Sea. And the only estimate for a North Sea cod of maximum age is 11. So, given what we now know about when cod mature, just how many adult cod, as the Sunday Times would put it, really are in the North Sea? Remember, they said there were just 100. What's the right answer?
- [19:15]
G
Well, using the same data set the Sunday Times used, it's actually 21 million. And as for the Daily Telegraph's version,
- [19:23]
A
only 100 COD left in North Sea.
- [19:26]
G
Well, here's a more accurate rendering of that headline based Again, on the Same
- [19:31]
A
data, only 436,900,000 COD left in North Sea. Hannah, this could be a more or less record, the wrongest number we've ever reported.
- [19:41]
G
Yep, and I did ask the Sunday Times to respond to this. They told me that the article oversimplified a complicated issue which they now regret. They also said they did make it clear that they were referring to CODD over the age of 13.
- [19:54]
A
Well, thank you to the BBC's cod correspondent, Hannah Barnes. A listener, Lucas Morrow, emailed more or lessbc.co.uk, asking whether Fergie Time exists or is just another zombie statistic that won't die. Football fans may know exactly what he means, but let me When Sir Alex Ferguson's Manchester United team are behind towards the end of a match, do they get an extra helping of added time just when they need it to mount a comeback? Last weekend Norwich City were victorious against Manchester United, beating them 1 0. Our listener said he noticed the full time whistle was blown slightly later than the expected five minutes of added time. In this case, it still wasn't enough, but it has helped in the past. Now, some, mostly non Manchester United fans, I think it's safe to assume, accuse United of getting more added time than any other team. And they've dubbed this Fergie Time after the manager, Sir Alex Ferguson. Now, this implies that referees aren't doing their jobs properly. We asked former referee Graham Pole what he thought of that.
- [21:02]
B
When you're refereeing, you don't believe it. How would you be influenced by such a ridiculous concept? However, when you actually analyse and think psychologically what happens, the pressure that's on you when you referee at Old Trafford or at the moment the Etihad or the Emirates or Stamford Bridge, the pressure that is implied upon you must have an effect, even if subconsciously you're sitting there going, okay, there's a couple of subs, a couple of goals, that's wasted time, injury there, you know, three minutes, maybe four, you find yourself saying five. That's something which you then sitting down and analysing it, you kind of go, so where did that extra minute come from? That's when the subconscious can kick in and it's a very strong referee that can actually recognise that and make sure he doesn't fall for it.
- [21:43]
A
But let's look at the numbers. I spoke with Duncan Alexander from Opta Sports who collate all the data from the football leagues. He says the phenomenon of Fergie Time goes back to the very first Premier League season 1992-93, when Manchester United played
- [21:58]
H
Sheffield Wednesday, the game came to 90 minutes, the referee awarded quite a lot of injury time and they scored and went on to win the match and win the title. And I think it's a sort of bit of confirmation bias really because ever since then, every time United have been given quite a lot of injury time, it sort of flagged up in people's heads. They've gone, well, you know, United have got more Fergie time again.
- [22:19]
A
Well, let's have a look at the numbers. I suppose you want to look at second half injury time because no one really cares how much time is added on at the end of the first half because you're in the middle of the game. What do we find? What do the numbers tell us?
- [22:31]
H
This season United have had the most. So that makes it true?
- [22:34]
A
Well, it's true.
- [22:35]
H
It's the truth.
- [22:36]
A
Fergie time is a fact. The numbers don't lie.
- [22:39]
H
Well, that might be the case, but then last season they had the least.
- [22:42]
A
Ah, okay. That would make things a little different. Yes, I suppose actually we might be more interested in situations where United are losing or I suppose even when United are drawing, they're supposed to be a top team, they expect to win most games. So when United are not winning, what do we find? Do the referees add on extra time there?
- [23:01]
H
Well, I had a look at the length of games for Manchester United over the last three seasons. In games they lost, they had an average of 4 minutes and 37 seconds compared to 3 minutes and 18 seconds when they won. And draws was about in the middle. So you can see there that games they lost, they've got more time. So whether that is a case of the defending team sort of wasting time to try and hang on to what would be a fairly notable victory or whether it's because referees are influenced by the fact that Manchester United are losing. We can't necessarily draw from the data, but it is, you know, that is true that they have had more time when they've lost.
- [23:37]
A
How does that compare to the other teams?
- [23:39]
H
Well, I looked at the other so called top teams, I mean Arsenal, Chelsea, Liverpool, Man, City and Tottenham over the same period past three seasons. And other than Chelsea, all of them had longer. Their games were longer on average when they were losing, when they lost the match. So you know, if there is a bias inherent with referees, it's not just towards Manchester United. It looks like it's to all big teams.
- [24:00]
A
But can we account for other events in a match? Some matches have more substitutions or longer goal celebrations, which would increase the added time. Or when a big team such as Manchester United are losing or even merely drawing, their opponents could be desperate to hang on to their position. They might waste time, the goalkeeper might hold onto the ball. Gabriela Lebrecht is from Decision Technologies, who also analyse sports statistics.
- [24:22]
C
We try to calculate the stuff that the referees should be adding on extra time for like substitutions or yellow or red cards or goals. And then you're left over with a certain amount which seems to be affected by.
- [24:34]
A
The referee doesn't have any obvious explanation. It's not to do with substitutions, it's not to do with goals scored, but it is possibly to do with home team advantage. The home team is losing, the home team is strong, that kind of thing. So what did you find there?
- [24:48]
C
If the home team is winning, then the extra time is cut by 46 seconds.
- [24:54]
A
The ref just calls time early, the fans are whistling, the home fans want the game to be over and the referee caves in.
- [25:01]
C
That would appear to be the effect.
- [25:04]
A
So it's the converse of the Fergie time hypothesis.
- [25:08]
C
Yes.
- [25:08]
A
Now, is there anything you've seen that does correspond to this idea that strong teams that are losing get extra time with the aim perhaps of helping them catch up?
- [25:19]
C
That's also an effect within the model. There's an interaction term that if a strong team is losing at home, they get more time than if a strong team was losing away.
- [25:31]
A
Okay, so you've tested two things together. It's a strong team and the strong team is losing at home versus a strong team and the strong team is losing away. And so if the strong team is in this position at home, you're saying there's Fergie time. At that point, there's a bit of extra time to catch up.
- [25:48]
C
However, if they're playing away, then it doesn't seem to exist as much this Fergie time.
- [25:53]
A
I'm also interested, how long does the ball just spend out of play in a typical football match?
- [25:58]
C
Average match, about 34 minutes.
- [26:00]
A
So there's no question that the referee is actually adding on the time that the ball's not in play. It's a much more informal thing than that.
- [26:06]
C
What's interesting in terms of how much does the referee watch the clock if, for example, a substitution is made in extra time, they get a lot more time added to the clock than they would if the substitution was made in normal time.
- [26:20]
A
Okay, so the ref's paying attention.
- [26:22]
C
Well, he feels the fans anger if he doesn't add on enough time.
- [26:26]
A
Okay. And all of these effects that you have described here, they're all statistically significant?
- [26:32]
C
Yes, that's correct.
- [26:33]
A
Now, of course, numbers do not lie. Your analysis is entirely unbiased, completely objective. Are you a Manchester United fan?
- [26:41]
C
That's correct.
- [26:43]
A
Interesting. Looking at the numbers, it would seem former referee Graham Pole is right. A version of Fergie time does exist, not only for Manchester United, but for all the top Premier League teams. So there's Mancini time, Wenger time, Villas Boas time and Benitez time. Or whoever's in charge of Chelsea this week, thanks to Gabriella and to Duncan. Now, Sir Alex Ferguson isn't the controller of Radio 4, so, sadly, there's no added time for us. For more information on our stories, you can visit BBC.co.uk more or less and keep your questions and ideas coming in to More or Less BC that's all we have time for this week. Goodbye. Thanks for listening.
- [27:26]
E
More or Less was presented by Tim Harford.
- [27:28]
A
The producer was Wesley Stevenson and the
- [27:30]
E
editor was Richard Varden.