Bishop Hill Boris on Piers
Jan 21, 2013 BBC Climate: Sceptics Boris Johnson, the shock-maned mayor of London, is waxing lyrical in the Telegraph about the virtues of Piers Corbyn and the possibility of a little ice age.
I am speaking only as a layman who observes that there is plenty of snow in our winters these days, and who wonders whether it might be time for government to start taking seriously the possibility — however remote — that Corbyn is right. If he is, that will have big implications for agriculture, tourism, transport, aviation policy and the economy as a whole. Of course it still seems a bit nuts to talk of the encroachment of a mini ice age.
But it doesn’t seem as nuts as it did five years ago. I look at the snowy waste outside, and I have an open mind.
I can't number myself amongst Corbyn's fans, having never seen any evidence to suggest that he can in fact weather forecast better than anyone else. A couple of years ago Roger Harrabin tried to organise a formal competition between the various weather forecasters, all of whom lost interest fairly quickly. This didn't encourage me to place any great weight on any of their claims.
Reader Comments (68)
Whether or not Piers Corbyn is right it is obvious to anyone (except possibly a few politicians and Greenies) that this country is badly prepared for even a few inches of snow. Money seems to be no object as far as "climate change" is concerned. Why not take just one tenth of the money being spent "combatting climate change" and use it to grit the roads, clear railway lines and airport runways etc.?
Below is a Met Office forecast, it's very general. Compare this to Piers' forecasts which tend to be very particular and much more useful, though I don't know how accurate they are. The problems are outlined here
http://news.bbc.co.uk/today/hi/today/newsid_9699000/9699478.stm
UK Outlook for Friday 25 Jan 2013 to Sunday 3 Feb 2013:
Many southern and central regions are likely to start dry on Friday but generally rather cloudy. However, more unsettled conditions with spells of rain, sleet and snow along with increasing winds, will develop over northwestern parts. The unsettled conditions are then expected to spread erratically southeastwards over the weekend. Remaining cold initially, with hard overnight frosts, widespread ice and some freezing fog patches, these locally slow to clear from some areas. However, during the weekend and into the new working week onwards, conditions are likely to turn rather less cold, as more-unsettled Atlantic weather systems push into the UK. This change to more unsettled conditions is likely to be accompanied by spells of snow in some areas. Later in the period, snow will be mainly restricted to upland areas.
An interesting piece. Leaving aside Piers Corbyn (about whom I am not knowledgeable enough to comment) I read it more as a call by an intelligent man for some questioning of the dogma without making himself sound like a heretic.
Nic L (comment above) - if Boris was genuinely intelligent he'd be seeking a specific and quantifiable test to compare forecasts. Some snow at winter = ice age as much as sun in summer = global meltdown.
It's not whether Piers can forecast better than any other forecaster but whether he can forecast further forward more accurately than the met office who for such a large and "trusted" organisation are woefully inadequate in anything over 5 days.
Met Office hindcasts and rolling forecasts are near 100% though.
So is Boris leaving the CAGW camp?
Boris Johnson...the Bertie Wooster of politics...but a clever version.
"Far be it from me to contradict all those scientist chappies but global warming just does not seem to be happening. Let us carry on doing the sensible things that make sense in any scenario and quietly drop the crazy ones dictated solely by the CAGW Theory until we know a bit more, eh? "
Sound sense delivered in Boris's inimitable style. I am no great fan of the man or his politics generally but he is spot on here. Every little helps.
I too have no evidence to validate Piers longer-range forecasts, but at least he puts his money where his mouth is - and his income is dependent on the success of his forecasts (unlike the Met Office whose failed forecasts result in them getting ever more funding and big bonuses).
Don't underestimate Piers. I was in the same physics class as him at Imperial and he was one of the really bright students. Annoyingly he was one of the very few who found it too easy and didn't need to study really hard.
Boris will become an ever more important figurehead in the Tory party. Leaving aside Corbyn, boris should be helped along. Maybe Lord Monkton can get to him. Maybe the bishop can send him a free book. If he is wavering he must be brought over to our side. He is very likely a future leader of the Tory party. He would be a massive asset in maybe five to ten years.
Don't underestimate Boris or Piers Corbyn. Listen past what they look like and how they say it. Neither of them is as dishonest as certain paople at Exeter and neither of them is quite so ... well .... you know.
I have followed Corbyn for some time - and changed my travel plans as a result in previous winters - and would say that his record of forecasting extreme weather events is quite good (as opposed to bread and butter forecasting where he clearly doesn't have the resources). The reason if true is that he has a specific theory of solar influence on the stratosphere. This would be relatively uncontroversial if its acceptance didn't open the Pandora's box of solar versus human impacts on climate. If you check his Weather Action web site he did forecast the very heavy snow of the last few days back in Dec and interestingly now has another forecast which we can all check. This is for a major snow fall - possibly worse than currently - around Jan 31 (check Weather Action for precise details).
I'd like to see the evidence that Corbyn has had a good record - it shouldn't be hard to lay out clearly the predictions certified being ahead of time against the actuality in the real world.
After all the Met office have suffered because they have at least let their predictions be seen and clearly documented and witnessed to be not very good.
Corbyn doesn't show the same ability. I'd like someone to show it clearly - and please, no references to his web site! I can't make head nor tail of it!
I don't see any evidence that anyone can improve upon the Lorenz chaotic horizon of 10 days before the complexity overpowers the knowledge of the initial conditions.
Piers certainly appears to get it right more often than the MO, and from a greater distance. I believe he predicted cold and snow for Jan 17-20th last month, and says there will be more of it at the end of the month and into Feb.
Probably difficult to compare with the MO, who stopped doing long-range forecasts after their mild winter and barbecue summer predictions...
I don't blame Corbyn for note taking part in Harribins propaganda game...which is exactly what it was. It was nothing more than an attemp to justify the MET Office forecasts and to denigrate the work if private companies.
Besides, if you were Corbyn would you trust the BBC?
Mailman
"Boris Johnson...the Bertie Wooster of politics...but a clever version." Jack Savage
Don't forget that Bertie won the Scripture Knowledge prize.
All I know is that the Met Office might as well look at tea-leaves/dried pinecones/seaweed and get just as good a forecast as their computers spew out.
As for Corbyn's forecasts, I simply do not know.
However what I do know is that he gets paid by results, unlike the shower (pun intended) at the Met Office.
I have been following Piers Corbyn's WeatherAction.com site for a few years. I have watched his forecast post event, & they do seem pretty good. I understand that he used to have bets with Hills who no longer indulge in these due to their mounting losses. He talks good sense imho! He works in the private sector where one has to stand or fall on one's own feet. He is still in business after many years which speaks for itself really. If he regularly got it wrong he wouldn't be in business. The Wet Office can get it wrong time & time again & nobody will officially censure them from top to bottom, as said on another thread, no national broadcaster (Beeb) would smite a national weather forecaster by using someone else, it would be unthinkable regardless of practicalities, like the straw man argument put forward by some a couple of years ago!!! The £234M a year WO just asks for more money claiming that more computing power is the answer, without even considering the possibility that their basic strategy & philosophy is right or not, they just don't care with their cushy salaries & pensions they self-perpetuate! I would like to ask the likes of Prof Julia Slingo or Dr Vicky Pope or John Hirst, just how many lives are they prepared to see lost through extreme cold, icy roads, etc, before they will consider re-thinking their Global Warming mantra, I want them to actually put a number to it, 5, 10, 20, 50, 100, 500, 1000?
Nicholas Hallam, as Gussie Fink-Nottle pointed out
Of course Bertie did also contribute an article to Milady's Boudoir on What the Well-Dressed Man is Wearing.
A five-day "forecast" is surely no such thing, right? AIUI, a forecast over such a short range is in fact largely observational. You observe what the weather is like five days' worth of travel out into the Atlantic and that's next weekend's forecast.
Up to a point anyone can do this. You'll have to make calls about whether the weather coming in from the east is going to shoulder aside that coming in from west, or whatever, but given access to satellite pictures, there are only so many possible outcomes over shortish horizons.
This is essentially how the D-Day meteorologists worked. They sent aircraft into the Atlantic to report on what the weather was like at various points west. These were compiled into a forecast from which they picked a date.
It makes sense for the Met Office to be reasonably good at this. They have lots of satellite photos to hand. They could still be rubbish at anything longer term, though, because there, they have to rely on some sort of model of weather patterns instead of data. And as ole Julia has told us, they do, and these are based on CAGW hysteria. So obviously they're going to be right only or largely by accident.
Whether Piers Corbyn is right more often than the Met Office would surely depend on whose model is less wrong, but it's not at all clear to me what sort of timescale would be required to establish this conclusively.
Nor is it obvious how you'd score such a comparison. If someone says it will be a rainy, windy summer, well, rainier than what, and by how much? And if it's one but not the other, is that nul points or half-marks? Should there be a third runner, in the shape of a default model which predicts nothing except that the immediate future will be much the same as the immediate past (that's the one I'd be backing)?
If it eventually turns out that the Met Office has spent however many years making long term forecasts on the basis of a totally bogus hypothesis, and that nobody else did any better, it does rather argue for closing down the Met Office. If it is possible to produce weather forecasts based on the entrails of a chicken, and they're simply as bad as everyone else's, what's the point, exactly?
I agree with Phillip Bratby ( 8:31 AM ) when he notes that the worse the Met Office get, the more money they can argue for, and get, for such as bigger computers. Corbyn uses, I think, a portable PC and will go out of business if his predictions are sufficienty wrong.
I think it would help governments allocate money more wisely in future if the Met Office could be split into two quite distinct entities: one for weather forecasting and one for climate speculation.
Whether the weather is cold,
Or whether the weather is hot,
We must weather the weather,
Whatever the weather,
Wether we like it or not.
Anon.
The Met Office commissioned a study by a consulting firm to put a value on the weather forecasting services delivered by the Met Office, based on estimates of lives saved, casualties avoided, damage avoided and so on as a result of correct weather forecasts. The study concluded that the annual value delivered was hundreds of millions.
I thought the report was a disgraceful exercise in dishonesty, as it took no account of work days lost, accidents resulting and lives lost as a result of incorrect forecasts.
To me, it provided confirmation that the Met Office sees propaganda and bullshit as its principal product.
I suggested something along those lines on a BH discussion thread: What should replace the Met Office?
It has been said that if you forecast tomorrow's weather as being the same as today's, you are marginally more accurate than the MO. If you then add a bit of common sense, you will be much better than the MO.
I wish somebody would do a trial on this - they might save us some money.
Boris constitutes a good weather forecast in himself - it's just that it's the political weather that his actions forecast.
A lot of people are asserting that Corbyn would go out of business if his forecasts failed.
Is there any sign that he actually makes any money? I have heard references to his laptop computer, which hardly suggests massive earning power.
When his name crops up, someone always mentions that he is not allowed by bookmakers to bet on weather events. Is this documented anywhere or is it another of those urban myths?
Certainly his long-range forecast, as reported, for January and Feb in the UK was rather too vague to be of much use, wasn't it? Where I am, we have had snow on 14th, 18th and 20th January. On those days, there were significant travel problems which would have led me to change travel plans if I had been intending to travel on those days. A general warning of freezing conditions with vague promises of snow for 2 weeks is not good enough for that sort of use, although it might be of use to say a farmer or someone who oes not need pinpoint accuracy - linking weather to a specific day.. Are the detailed forecasts any better?
And did he not forecast that the Olympics opening event would be rained off?
Maybe it's time to revert to the Man on the Air Ministry Roof, the massive computers seem to do no better. (For those too junior to know, he was the precursor of the Met Office.)
Alan the Brit
it is also the consequences of energy policy based on the MO's version of science that needs to be taken into account in terms of lives lost. I asked DECC recently how many deaths from hypothermia they had factored into their energy policies. Surprisingly??? I got a stereotyped letter that answered questions that appear to me to have been raised by someone else. Nothing on fuel poverty or its consequences.
Of course people don't die within the bubble!
@ diogenes
Good points. It does point up the difficulty of establishing what's an accurate forecast. If someone forecasts two weeks of snow in January and none in February, and instead we get a fortnight of snow spanning the end of January and the start of February, that was actually quite a good forecast, although wrong in respect of either month.
Piers Corbyn - well I tend to agree with the Bish, but it is also true that Corbyn was pointing to colder winters at the time that the MO and IPCC were telling us that the Northern Hemisphere would see increasingly mild and wet ones. Since he is a Sun watcher that might not be a surprise, and he wasn't the only one, just the one with the highest profile.
I agree with Roy's remarks in the first comment.
It is surely time for empirics to start making a come back over a thoroughly discredited theory. Time for science to return to climate science. There are signs that it is starting to happen, but I think we have much delusion and activism to overcome yet. Too many people who are making a tidy living out of unjustified alarmism. Still a huge number of "useful idiot" foot soldiers in the MSM and elsewhere.
I also think that Alarmists think they can tell as many lies as they need - Hurricane Sandy, Arctic sea ice "melt" etc. because the MSM will repeat them and the people will be scared. BUT what I don't think they understand is that such fibbing has two other effects, it makes sceptics more certain that it is an exaggeration and it is also noted by genuine scientists, even those in the AGW field. As Pointman's article underlined recently -
http://thepointman.wordpress.com/2013/01/18/on-climate-science-and-all-those-leaks/
Thank you for your comment which quotes one of P G Wodehouse's most amusing characters.
(Jan 21, 2013 at 10:31 AM | Registered CommenterJonathan Jones). I'm a great fan of P.G. with about 90 of his most humourous books on the shelves a few feet away. My favourite one has to be the collection of short stories in "The Clicking of Cuthbert"
It was reading about PiersC's exploits with William Hill that helped get me interested in this subject many years ago now. As a professional gambler myself who has also been banned from William Hill I was aware that if true this was no mean feat so became intrigued.
So the MET Office set the odds for an event, Hills add on their substantial margin and yet PiersC beats it CONSISTENTLY to win enough cash for Hills to close his account. If true, in my opinion this is a BIG THING.
It wouldn't necessarily mean Piers is any good (he may be bad but the MO very bad!) but it would mean he is considerably better than the MO for these type of events. So we would have empirical evidence (years ago) that the MO forecasting methods are duff (at least for these types of event). I've tried to find out more about this (e.g what type of events) but failed. Perhaps Richard Betts knows about it. Given the significance I'm surprised more hasn't been made of this.
I presume PiersC is his own PR department but if I was he I would get a signed letter from Hills to prove the claims and wave it around in the MO's face at all opportunities.
The forecasters don't have to be right every time (so it is easy for detractors to find a failed prediction for anyone) but they do have to show "skill", i.e. better than pure chance over a sufficient timescale.
Since I first read about it I've argued for an ongoing controlled test to see if any of these organisations do show any signs of skill. It would be difficult to set up but not insurmountable. Specify timescales (yearly, seasonal, monthly, 5-day, 1-day), measurements (temperature, precipitation, sea levels, tornadoes etc). I remember the Harribin suggestion and would also like to know what happened to that.
The MO must have figures internally to show how they're doing at all forecasting timescales surely? Anecdotally on forecasts longer than a few days they seem way worse than pure chance to me but show the figures and let's see.
We could have had plenty of years worth of data by now and a picture would soon emerge of who knows what they're talking about. Hmm, that's probably why we haven't seen it.
Please, please please you get no credibility for critising piers corbyn unless you state that you have purchased a forecast and measured it by some predetermined parametres which are similar to those used by the UKMO. Stop criticising from your guesses of what he forecasts.
I know he closely controls his website comments but so what and the other comment about "does he make money". Well, I have not seen his bank accounts but he rents an office in London (not the pretty part), he organises conferences at the House of commons and he buys laptops and he employs people so I don't suppose you can do that on social security even in the UK.
So buy a forecast, do the analysis and come back and discuss it. I for one will be very interested.
Does Corbyn have any other income other than that through "Weather Action"?
If not he is totally dependent on getting things right, more often than wrong.
After all would you pay for something you could get free, if it were just as good?
A five-day "forecast" is surely no such thing, right? AIUI, a forecast over such a short range is in fact largely observational. You observe what the weather is like five days' worth of travel out into the Atlantic and that's next weekend's forecast.
This is one of the better comments today. The MO 5 day forecasts are only moderately reliable if the weather is predictable. That is if the jet stream is stable. When the jetstream becomes unstable their 1 day forecast can be utter nonsense. Their monthly forecasts are useless information because of the nature of their model which chucks out mathematical probabililties. These probabilities tend to be 30%, 30%, 30% of each type of weather (wet, dry, cold, warm).
Piers on stilts?
The MO must have figures internally to show how they're doing at all forecasting timescales surely? Anecdotally on forecasts longer than a few days they seem way worse than pure chance to me but show the figures and let's see.
They have figures but as I said recently, they are measuring themselves. How do you think they do? Brilliantly of course. The fox counting the chickens comes to mind.
As I have said before, the question of whether a forecast is "accurate" is not a simple thing. You have to define what constitutes accurate and I have not been able to find this for the MO. I was referred to a web page full of indices etc which claimed to show that they were very accurate but nothing to say whether they get within a degree or ten of their forecast, over what area and for how long.
The only solid fact I have found is that for wind in the Inshore Forecast "accurate" means +/- 1 beaufort force against a forecast which can span 4 BF, again no information about where and how long.
E.g. if 24 hr forecast said "Winds 4-7" and the wind at some time during that forecast was somewhere between 3 and 8 (!), would that be recorded as "Accurate forecast", albeit useless?
There are a number of webpages out there where people have conducted their own limited accuracy studies of local MO forecast vs actual, they all seem to get results around 50% which I suspect means random chance.
This an old article from Wired magazine:
http://www.wired.com/wired/archive/7.02/weather_pr.html
It's an interview with Corbyn from 1999 and it does give an idea as to how Weather Action make money. It also touches on Dr Dennis Wheeler's confirmation of Corbyn's technique. A long but interesting read.
@ NW
This sounds a bit like the situation whereby the train companies are able to claim 99% of trains run on time by redefining 'on time' as meaning 'less than half an hour late'.
Johnson writes: "I look at the snowy waste outside, and I have an open mind."
So does Penn Jillette, American magician and entertainer. In the link below he echoes much of what Johnson says-- to wit, he doesn't know what's going on either. He does it with a strong dose of profanity, however. Very amusing. The first minute twenty are taken up with an introduction. But it's worth the wait:
Penn and Teller
This was first linked by a poster named Canman at WUWT on Saturday.
Boris Johnson rowing back from a position of luke warmist to pragmatic agnosticism should be a cause for some celibration amongst sceptical bloggers.
Ed Miliband the ex Labour Climate Secretary and his now shadow cabinet were responsible for the imposition of the AGW policies that are ruining our country today, whilst Cameron the faux greenie and Clegg the buffoon continue to support a huge earner for their immediate families and friends.
If, as we all hope, a denouement of this scam occurs in the next two years, only Boris, Nigel Farage and the handful who voted no in 2008 will stand untainted to fill the vacuum amongst the wreckage of British politics.
As for Piers Corbyn, I must confess to having benefitted from timely bits and pieces of long range forecasts that were blabbed by subscribers on various Blogs over the past few years. I cannot remember a single snippet that was not advantageous to me.
Piers may not sing from the same hymn sheet as some of you, but climate theories are many and varied as might be expected when the argument boils down to .7 of a degree in a chaotic system comprising as many variables as government funding might allow.
Don Keiller
This might be a useful starting point for researching Mr. Corbyn.
http://www.sourcewatch.org/index.php?title=Piers_Corbyn
E.g. if 24 hr forecast said "Winds 4-7" and the wind at some time during that forecast was somewhere between 3 and 8 (!), would that be recorded as "Accurate forecast", albeit useless?
If I was going out on my boat and I was expecting a force 3, force 8 would be very annoying. Perhaps that's what that woman did around the sw england in the autumn when she disappeared in a force 8-9 in her newly bought boat.
This is one man's attempt to analyse the reliability of Oiers Corbyn's forecasts.
http://www.democracyforum.co.uk/environment-energy/93140-how-reliable-weather-action-forecasts.html
Anyone else got better data?
They do that 4-7 thing a lot over the summer and it's no use whatsoever for passage planning. Also, the 48 hour forecast very rarely resembles the 24 hr one 24 hours later, so it's not a lot of use either.
Local knowledge and the location of highs and lows is more useful.
I have a suspicion that the MO models treat the earth as some variation on a smooth sphere and don't take account of the effect of the landmass on what happens. If you take their wind forecast and adjust for the amount and height of land between you and it then it starts to more closely resemble reality. Likewise for rain shadow effects.
I remember the Harrabin proposal for a contest re: forecasting skill. I thought at the time that he was a singularly poor chairman of such an effort. Heartlessly, I thought that it would be very easy to force "the challenger" onto Met Office turf, rather than Corbyn's strength, which appears to be long-term forecasts of extreme events. And that was before anyone opened the hornet's nest that is the scoring system in such a contest, as others have pointed out.
And I have to say, I was tempted to put in an FOI request at the Met Office for correspondence with Harrabin to find out exactly what the proposed term were, and what contact there was between the Met and Harrabin prior to the announcement. But I didn't.
Having read the extract of the mayor's musing (below), I checked his birth date on Wiki. He was born in 1964. Now what happened in the winter of 1962-63?
Sometimes we have very long cold winters. Sometimes we have long, hot summers. It is called natural variability.....
"I remember snow that used to come and settle for just long enough for a single decent snowball fight before turning to slush; I don’t remember winters like this. Two days ago I was cycling through Trafalgar Square and saw icicles on the traffic lights; and though I am sure plenty of readers will say I am just unobservant, I don’t think I have seen that before. I am all for theories about climate change, and would not for a moment dispute the wisdom or good intentions of the vast majority of scientists."
I would also note that the wikipedia page states that WeatherAction used to be listed on AIM but was de-listed in 1999 because of persistent losses. It also shows cases where forecasts were right and wrong. I cannot find any trace of WeatherAction on Companies House so cannot attest whether he employs staff or makes money currently.
Over to the fanclub for any facts.
Let's face it, back in December, i thought, we will probably have snow in January. For the last 4 years, without checking the records, I recall that we have had snow here for a period of about 2 weeks every year, between mid-December and mid-February. I did not even need a laptop to make that prediction.
Early Weather Action (Solar Weather Technique) skill was independently verified in a peer-reviewed paper by Dr Dennis Wheeler, University of Sunderland, in the Journal of Atmospheric and Solar-Terrestrial Physics, Vol 63 (2001) p29-34.
It's behind a paywall, but someone who read the paper summarised as follows:
Essentially he assessed Corbyn's predictions of gales in the UK over a two year period. He used a chi-squared test to work the probability that Corbyn's results might have occurred by chance and found ...
"the all year forecast success rates have only a 1 in 1000 probability of occurring by chance. It must be recalled however that this figure is derived by inclusion of the summer season data when ‘success’ rates are inflated as a result of the marked seasonality in the incidence of gales [ie its easy to predict 'no gales' in the summer]. In comparison the more informative forecast success rates for the September to April period have a random probability of 1 in 125. The corresponding figure for the winter only period (October to March) is much lower at approximately 1 in 5.
Then in the conclusion, he mentions this Yet more compelling is the finding that the system successfully forecast, several months in advance, the four (arguably five) most active and damaging of the storms that occurred in the survey period.