How not to bet on the future

In an interesting coincidence, this article appeared in the New York Times a day or two after my bet was announced. The article, entitled "The $10,000 Question", concerned a $10,000 bet...but not mine. Instead, it was a bet on future oil prices, with one party arguing that the oil price will be above $200 by 2010, and the other arguing that it will be lower. Actually, the bet was only $5,000 v $5,000, so $10,000 in all.

Well, maybe I'm missing something, but we already have a large betting market on future oil prices...and it's helpfully called the "oil futures market". It suggests a price in the region of $60 by the end of 2009 (eg here, I don't really know where the best info is).

Therefore, anyone who thinks that the future oil price will be actually higher than that, can buy a future delivery of oil at that price. They don't need to actually be prepared to take a physical delivery of oil, they can simply sell the future contract nearer the date.

So, it seems to me that the person who bet on an oil price of over $200 was rather silly, because they could have got a much more profitable and lower-risk deal from the existing market. And that's before we even start with put and call options which could probably leverage a very large return on a small investment. If anyone else wants to place such a bet, please let me know - I will take willingly you on, hedge on the existing market and make a large risk-free profit.

A rather more attractive oil bet (from the "alarmist" point of view) can be found here. A drop to $25 by 2010 seems pretty unlikely right now.

Update
Ok, I found some futures and options data here. It looks like you can buy a "call option" for Dec 2010 oil at $70, for a price of say $4 (assuming I'm not misreading the site - it is not entirely clear to me). That $4 gives the owner the right (but not the obligation) to buy oil at the stated price and date. So if the oil price is $100 at that time, that nets a profit of $26 on a stake of $4. At $200, the return is $126. I'm sure that Matthew Simmons (who took the high side of the $200 oil bet) knows all this, so I conclude that he is giving away $5000 as a publicity stunt to push sales of his new book. Or maybe he's just trying to prove that doomsayers are stupid.

Anyway, I have no book to promote - yet :-)

Bet number 2

I'm delighted to announce that a second bet has been arranged, on essentially the same terms as my one.

It's for 500UKP, between Chris Randles (taking the warm side) and someone who posts as "snow hope" on the cold side. I don't know much about the protagonists, except Chris seems to be an active paticipant in the climateprediction.net project.

Some background to the bet can be found on this thread, and astute readers will spot that there appears to be another 500UKP on offer to anyone prepared to back the "warming" side of the argument.

I hope that someone will step forward.

Animal testing in medical research

I'm not a huge fan of animal testing. On the other hand, I can appreciate that it has its uses. So, although I'm all in favour of peaceful protest and even a bit of civil disobedience where appropriate, I'm disturbed to see that increasingly, de-facto policy in the UK appears to be being set by terrorists rather than reasoned debate over the ethical and practical pros and cons.

I wonder how long it will be before someone starts shooting doctors who perform abortions? Of course, the British have long preferred animals over children, so the AR movement probably has broader support than right-to-lifers :-)

The great egg mystery

Shortly after we came to Japan, a friend sent us a copy of an article written about the Japanese by AA Gill, which contained the following memorable phrase:

It's not that they're aliens; but they are the people aliens might be if they'd learnt Human by correspondence course.

Nowhere is this truer than in their cuisine.

There are few items of sushi that would not be improved by flashing them under a hot grill with a bit of butter and lemon juice. I strongly suspect that rotten squid guts (shiokara) and other similar monstrosities are just tricks to play on unsuspecting foreigners. But it's their treatment of eggs which I find particularly curious.

Firstly, there's the raw egg for breakfast. Before coming to Japan, I never imagined that the old party trick of telling whether an egg is cooked without breaking the shell would become an invaluable safety net in my life (spin the egg on the table; stop it momentarily and release; if it starts to spin again, it is raw). But at 4am after a poor night's sleep in a mountain hut, it is bad enough when the man next to you at breakfast stirs the egg into his rice and slurps it down noisily, without having to actually repeat the act oneself.

However, the mystery egg that I had with my dinner tonight and which inspired this blog entry was not raw, but soft-boiled. Of course we all know and love the western-style soft boiled egg, with firm white and runny yolk for dipping toast into. The preparation of this elegant culinary masterpiece has been the subject of much research, eg there are a bunch of formulae here, and another method hinted at on that page is to use a carefully temperature-controlled water bath at about 63-64C, at which temperature the white will congeal but the yolk will not.

Tonight's egg, however, was a complete mystery. The white was very runny, in fact barely cooked (mix of translucent and white, that poured easily). However, the yolk was firm! Not absolutely hard and crumbly, but a long way from being toast-dippable.

So, since the white congeals at a lower temperature than the yolk, how on earth did they do that? Some might say the really interesting question is why did they do that, but I've learnt not to even try to go there...

The Bet

A few people have asked about the details of the bet I've arranged with Bashkirtsev and Mashnich. Both Jim Giles (Nature) and David Adam (Guardian) basically got it right, but here it is again in full.

• We will compare global surface temperatures in 1998-2003 with those in 2012-2017 (6 year average in both cases), using the USA National Climatic Data Center data (which can currently be found here for annual and here for monthly analyses). (Update 13 Feb 2007: data seems to have moved to here.) (Update Jan 2010 now it's here)
  
• If the temperature rises over this interval, B+M will pay me $10,000 (in total). If the temperature drops, I will pay them $10,000 (again, in total).

It really is as simple as that. There are no exemptions for volcanoes, large meteorite strikes or nuclear winter. I recognise that this introduces a small additional risk against me: such short-lived perturbations could cause a few years of anomalous cooling in an otherwise warming trend, but I can't think of any corresponding external forcing event which could cause a few years of anomalous warming (El Nino could have a modest effect). There is also no allowance for anthropogenic emissions, as any plausible changes in these can hardly have an effect over such a short interval, even if one accepts that they have a significant effect at all (which B+M presumably would dispute). There is also no explicit agreement as to what we should do if the NCDC ceases to exist - I would expect to transfer to another near-equivalent body (eg CRU in UK, NASA GISS - there is virtually no chance of any of these disagreeing on the sign of the temperature change).

At the moment, this bet is essentially a "gentleman's agreement". B+M (M, in particular, CCed to B) expressed their acceptance of the bet over 2 emails to me, and also confirmed this directly with Jim Giles. I suppose it would make sense to actually exchanged signed bits of paper about it (and will try to arrange this), but I suspect that in the event of a default, chasing such a modest gambling debt internationally through the courts would be effectively impossible anyway. I'm also confident that B+M are honourable and honest people who will not try to wriggle out of paying up if they lose. Note that the loser will almost certainly see it coming a few years off and have a long time to save up.

Obviously, for a proper futures market, it would be necessary to pay up-front to a recognised market, where the money would have to be invested in a sensible (interest-bearing) place and controlled by a trusted third party who would eventually also judge the bets. At the moment, none of the obvious on-line betting markets seem interested due to the time scales involved, but I haven't given up hope.

More on Spencer and Christy

I'm feeling a little guilty, because my blog (and in particular this page) ranks very highly on a google search for spencer and christy - for reasons entirely unrelated to climate science. So I suppose even though it is not my speciality I should at least plot the new data and point to more useful urls.

But until I get around to that, you'll have to make do with this interesting snippet from an article by Roy Spencer on the Marshall Institute web-site, commenting on the 3 11 August sciencexpress papers:

I only hope that the appearance of these three papers together, with considerable overlapping of authorship, does not represent an attempt to make measurements fit theoretical models. For when this happens, actual measurements can no longer fulfill their critical role in independent validation of climate models. Ideally, measurements would be analyzed with no knowledge of what any given theory predicts they should be.

The only overlapping I noted was Mears and Wentz being co-authors on the Santer et al paper. I've not read any of the papers yet, as I do not have access to Sciencexpress, and do not know in detail what input they had. But in the absence of any evidence of malfeasance, it seems like a rather unnecessary slur. Talk of "actual measurements" and "theoretical models" also seems rather loaded terminology given the amount of processing that the "actual measurements" undergo - processing that has been shown to be faulty in the case of S+C's previous work, even if they have now got it right (which is still very much a matter of debate - Mears and Wentz get 0.19C/decade, versus S+C's 0.12C/decade).

Interestingly, this paragraph does not appear in the otherwise word-perfect TechCentralStation version.

Update (26 Aug)
This page is now well up the google rankings and being found by people who are probably hoping to find some interesting discussion. But I don't have anything much to say about the new data - it's not my speciality, and anyway Science has not yet seen fit to let us hoi-polloi see the papers (something I might grumble about in more depth later). I suggest any accidental readers go to Realclimate and Deltoid for more detailed discussion. Kevin Vranes thinks I am gleeful about the errors - well, whatever. I do like to see the gap between reality and perception narrow a little.

Guardian Unlimited | The Guardian | Climate change sceptics bet $10,000 on cooler world

The betting story is in the mainstream press now, via this article by David Adam in the Guardian.

I'm pleased to see he explicitly mentions the futures market idea, which is an important part of the bigger picture. With Piers Corbyn apparently also keen to get involved, there may be opportunities for others to make some money, too.

So, I'm eager to hear from those on all sides of the debate who are prepared to put their money where their mouths are.

An index to quantify an individual's scientific output

I saw a discussion about this paper while looking through the latest Nature News. It introduces a new measure of scientific output, called the h index, where h is the largest integer such that a researcher has (co-)authored h papers each with h or more citations. As such, it accounts for both the volume and the importance the author's body of work. It's an interesting idea, but I have two specific criticisms (and a more general comment).

The first is a general one about the paper: when proposing a new measure of scientific output, it would seem appropriate to actually compare it quantitatively against the alternatives. The author "argues" (his term) that his new measure is preferable to 4 other common alternatives, but provides no hard evidence that it actually is better (eg in discriminating Nobel winners from professorial/tenured/non-tenured staff). Indeed, the evidence presented seems to rather undermine the measure - some of the Nobel winners have very mediocre scores, well below the author's threshold for a "successful scientist". But of course, I don't know how poorly the alternatives perform...

The second is a more detailed one concerning the measure itself. I'm sure that my opinion is completely biased, because the measure seems particularly ungenerous to me:-) The problem is that no effort has made to account for the individual contributionthat a researcher makes to each paper. As it stands, the measure is very generous to people who work in large groups who share wide co-authorship, and very stingy to those who largely work alone. I was surprised to note that none of the 4 pre-existing criteria, which the author provides for comparison, account for this factor either. In the UK (at least where I used to work, in a NERC laboratory) it was standard practice to allocate a proportion of a paper to each contributor - eg a 2-author paper might be shared 60%-40%, or 80%-20%, depending on the magnitude and importance of the relative contributions. Performance assessments can then be based on "paper equivalents", with one single-author paper being equivalent to say 2x40% + 1x20% contributions on 3 separate papers. NERC promotion guidelines also explicitly referred to single-author and/or first-author papers as being particularly valuable. According to NERC's guidelines, I was reasonably competent. By the h index, I suck. I therefore conclude that the h index is faulty :-)

I suggest the following simple adaptation: count "paper equivalents" rather than total papers. h would then be the largest integer such that the total contribution of the researcher to papers with h citations, is at least h. Where this percentage contribution information is not readily available, a sensible decreasing function of percentage authorship with position in author list could no doubt be devised - perhaps an exponential, or the nth person being awarded a score in proportion to 1/n (2 authors get 67%:33%; 3 authors get 55%:27%:18%).

More fundamentally, is there any evidence at all that such measures are useful predictors of future performance, rather than merely a recognition of past achievements? This is of course crucial if such measures are intended to be a used as a guide to resource allocation (as the author suggests) rather than solely as a reward system. How did the Nobel winners look, 5 or 10 years into their careers? It should in principle be straightforward to check whether this, or any other measure, has any useful predictive ability. I'm not aware of any attempts to demonstrate this, but I haven't looked. Comments?

Climate sceptics place bets on world cooling down : Nature

I've been away on holiday for a week, and have a bit of catching up to do. The most exciting bit of news (from my point of view) is that this Nature article by Jim Giles has just appeared. I'm grateful for his help in getting the bet with Bashkirtsev and Mashnich formalised.

I'll have more to add later as time allows...

Food miles

A hilly 13km cycling trip carrying a couple of melons on a hot summer's day really brings home the meaning of food miles!

Perhaps I should explain. There's a roadside fruit stall that sets up once a week on the route we cycle to work. Japanese fruit is really good, and we like to pick up a selection of whatever is in season. The stall mostly sells fairly local produce which seems to cover a wide range of items, but grapefruit is strangely absent - they are all imported from Florida, which seems odd for a country with such a strong tradition of citrus growing (eg satsuma). There also don't seem to be any local suppliers of that staple food of cyclists, the banana.

This NewScientist article from a few months back describes some interesting research on the hidden environmental costs of food production in the UK. It demonstrates that the total air and ship miles of imported food is actually an extremely small part of the overall problem, with lorry transport in the UK doing the lion's share of the damage. The real surprise to me, though, was the fact that the average person's shopping trips, 221 per year of 6.4km length (and almost all by car) causes about half as much harm as the lorries do. On reflection though, a 40-tonne lorry has a much larger payload as a proportion of vehicle weight than a 1-tonne family car with a bag of shopping, and the car trip will also take place in congested urban environments, so the per-tomato environmental cost of each mile in the car will be much higher than for the lorry.

Back to the melons. The sheer effort of transporting them a few miles also made me think about the extent to which agriculture relies on oil as a raw input. A bit of surfing finds energy input:output ratio estimates of about 10:1 overall from some alarmist "peak oil" type sites (ie 10 times as much fossil fuel energy is used in production as is created in food calories), and more industry-friendly estimates for the most efficient foods (bread) give a more optimistic picture, such as this estimate of an energy ratio of 1:5 in favour of food output. Probably the truth is somewhere between the two. In any case, even the most optimistic estimates underline to what extent our society has been based on ever-increasing increasing supplies of cheap energy.

Jolly hockey sticks

It seems to be a popular theme in septic circles that the IPCC TAR is basically equivalent to the MBH98 "hockey stick". Of course the reason for this is obvious - according to this mindset, any flaw in the MBH98 work would directly "disprove" global warming.

Over in sci.environment, one of the resident septics regularly brings up the subject of the IPCC's overuse of the hockey stick, and presented these two images as evidence:


(originals here and here).

As Coby Beck noticed, the second of these pics is actually a duplicate of the first, flipped from left to right, cropped and blurred.

Professor von Storch used the same picture for the title page of a presentation he recently gave (powerpoint file):


from which it can be seen that the picture originates with this International Herald Tribune article, which describes:

The new climate report, more than 1,000 pages in length, was the work of 123 lead authors and 516 contributing experts. It is the most comprehensive study ever made of the global warming phenomenon.

This doesn't exactly square with the septic viewpoint that the IPCC TAR was really based on a 4-page paper with 3 authors. But no matter. My real interest here is not the TAR itself but the extent to which the MBH98 picture has been used to promote climate alarmism.

I can now exclusively reveal the results of my painstaking investigation which took, ooh, at least several minutes. The very same graph was used by Bush in his famous "We must think about doing something, maybe, once I've got this pretzel problem sorted" speech:

It was also presented in Australia at the International Conference on Wallabies and Climate:


and it made an appearance on Red Nose Day too:


I'd be grateful for any further sightings of the hockey stick in the wild.

Living in the shadow of death

(Hey, I spend most of my life trying to think rationally - surely I'm allowed a bit of melodrama in a blog title.)

Japan is one of the most seismically active places in the world, reckoned to suffer about 1000 earthquakes a year (and 20% of all the world's big ones). Regularly, there is one big enough to cause damage and injury - since we arrived in 2001, there have been big quakes in Hokkaido (8.3 magnitude) and Niigata (6.9). The former occurred some way from population centres, and there were no direct fatalities, but the latter killed about 50 people, injured 5,000 and over 100,000 had to be evacuated from ther homes. Just a few years previously, the Great Hanshin Earthquake killed about 6,400 and caused 10 trillion Yen of damage (100 billion US$), the costliest natural disaster in history. There are numerous smaller quakes which cause small amounts of damage and limited injury - falling furniture and broken windows and the like.

Of course, what everyone round here is waiting for is the next really big one - the overdue Kanto (Eastern region) Earthquake. The recent 6.0 magnitude Tokyo quake provided a reminder to all of us, without doing any real damage (I was up in a mountain hut and missed the excitement). The last big one, in 1923, killed about 140,000, mostly through firestorms racing through the streets of close-packed wooden houses. It's anticipated that a magnitude 7 shock (which seem to turn up every 70 years or so - you do the maths) could kill about 10,000, many in fires in some vulnerable older wooden housing in Tokyo. In a city of 30 million, that's not really such a huge proportion. For us foreigners, vigilante death squads are likely to be a greater threat (and in case anyone thinks such behaviour is obviously a thing of the past, Tokyo's notoriously racist mayor gave a speech a few years ago urging the euphemistically-named Self-Defence Force to be ready to round up foreigners in the event of a quake, cos we are liable to riot).

Earthquakes are of course unrelated to climate change, but typhoons are a little closer to my area of research. It's certainly not my field, but from what I've read, there doesn't seem to be any really solid evidence of anthropogenically-induced increases in strength or frequency of tropical storms. There are plausible arguments (supported by modelling studies) that such effects may occur, but the natural variability is too high to see anything in the data. This very new paper has stirred up the debate a bit, but it's surely not going to be the final word on the subject.

Last year Japan was hit by 10 or 11 typhoons, nearly double the previous record of 6, and way above the typical 3 or so. The country seems to cope pretty well with them. When one is forecast to come close, people in the area tend to shut down and stay home for the day as much as possible. The infrastructure - drainage and buildings - generally copes very well, and there is rarely more than the odd person who ventures out at an inopportune time, or who gets buried under a landslide. Sorry if that sounds flippant - but it really doesn't seem to be a big deal usually, just a few hours of disruption to trains which seems quite nostalgic for us UK ex-pats. Last year, there was a particularly bad one (Typhoon Tokage) which killed at least 80 - that really was unusual (worst in 25 years), but still only represents about 3 days' worth of roadkill or (astonishing but true) a typical day's suicides here. I think it's important to keep a sense of perspective on the risks we face.

Natural disasters are not unique to Japan, of course. San Francisco is probably also due a big earthquake, and Florida has been getting plenty of hurricane action recently. UK residents have to make do with Tim Henman's Wimbledon exploits, and the England cricket team :-) The combination of typhoons and earthquakes keeps the power of nature high in the public consciousness here. Once a year we have "Disaster Prevention Day", when there are plenty of practice drills etc - of course we cannot prevent the event itself, but the consequences can be ameliorated both by increasing the robustness of the infrastructure (not just making sure that buildings are strong and fire-resistant, but being careful to keep large bookcases and other heavy furniture away from beds) and by having plans for the immediate aftermath, eg storing a few days worth of food and water so as to relieve pressure on the emergency supplies. Japan is a rich country, which means that the financial costs of natural disasters appear very high, but it also means that the human cost is relatively small - compare and contrast the effects of a recent hurricane which hit Haiti and the Dominican Republic on its way to Florida for a graphic demonstration of this. Roger Pielke Jr's blog has lots of posts and links to his papers on the significance of vulnerability to hurricanes, and the Bam earthquake in Iran killed about 25,000, despite only having a magnitude of about 6.5 - the mud-brick houses simply turned into heaps of earth, burying people alive. By being reasonably well prepared, we can usually prevent a disruption from turning into a disaster.

By the way, rumour has it that the Earth Simulator is built to be highly earthquake-resistant, but they didn't worry so much about the adjacent buildings that the researchers use...