the Air Vent

Because the world needs another opinion

Ten Things Everyone Should Know About Hockey Sticks

Posted by Jeff Id on November 22, 2009

This is an updated repost from a September 11, 2008. It’s was a popular post then but is far more relevant now that we see some of the bias in the recently released files. It has to do with hockey sticks in general but the graphs are from data used in Mann08, a paper by Michael Mann who is a key figure in many of the CRU emails. This post should also answer some questions of the newcomers at the Air Vent.


Figure 1

Most of us have seen a ‘hockey stick’ graph like the one below indicating temperature. It and others like it have been repeatedly used to demonstrate the recent warming of the earth. But how was it created? Certainly there were no thermometers being recorded in 1000 AD.

Number 1 –

This data is considered the powerful evidence of man made climate change. This graph clearly shows it is hotter now than in the past. Scientists often quote the rapidity of the rise in the blade as the reason warming is man made.

Number 2 –

The red line is measured data from thermometers back to 1850. The red temperature data is from the HadCRUT series which was the subject of the legal FOIA requests being blocked by an apparent conspiracy between scientists and government employees in the recently released emails. The source of the data for the red temperature curve is a critical part of understanding man’s role in climate change. The GISS temperature record is a comparable dataset which has massive corrections added right onto thermometer data. The corrections are poorly validated and represent the majority of the warming signal for the past century. Still HadCRUT has a higher warming trend and despite a lack of transparency is more often the curve of choice in global warming publications.

Number 3 –

The rest of the data comes from proxies. Proxies are stand in measurements for temperature employing things such as tree ring width, mollusk shells, various kinds of coral and ice cores.

Number 4 –

There has been no verification or expirament that tree ring widths, mollusks or coral etc. measure temperature. Instead the concept for each proxy started with a basic theory that the trees or other things might change width with temp. The only verification is when the measured data from tree rings “correlates” after “calibration” to temperature. There have been no laboratory verifications of the concept.

Number 5 –

In creating Figure 1, the data from the proxies which match temperature trends are kept, and the remaining data are thrown away. About 60% of the data was thrown away in generating this graph above, the remaining 40 percent that did agree were from a variety of sources (one tree was used, the one next to it was not). Proxies are often very noisy data with little visible signal. See a typical proxy below, selected because it was the first data proxy of the series used to create the hockey stick.

Number 6 –

The retained (accepted) data was scaled (magnified) and offset (up and down) to provide the best fit to today’s temperature measurements. After the averaging, the red temperature line above is glued on the end to extend the blade of the hockey stick.

Number 7 –

The fact that 40% of the data when filtered has some similarity to temperature, is the sole rational for saying that tree ring width, and mollusk isotopes are actually temperature.

Number 8 –

Math techniques called CPS and EIV were used to generate the correlation and scaling of the proxy data for the graph above. A wide variety of other methods which do similar things have also been used.

Number 9 –

Statisticians and scientists have repeatedly discredited the calculations involved. Still ever more varieties are created and the main government branch for global warming Intergovernmental Panel on Climate Change has refused to relinquish this point. This continues despite repeated demonstrations that the statistics seem to be specifically designed to pick out hockey stick shaped proxies from even random red noise data. (see links above).

Number 10 –

Continued acceptance by the premier global warming scientists demonstrates a clear political bias at the top of the global warming community, leaving the world at the mercy of the conclusions of a large power hungry government organization. (this was written in Sept of 2009)

Now that the emails have been released including what appear to be discussions of bending data to fit the pre-determined conclusions, elimination of ‘skeptical’ editors and manipulation of IPCC rules to prevent the points here from being made, the truth of my statement is apparent.

What we can understand now.

This first graph, except for the red line at the end is not temperature, it is not science either but rather it is the output of a cleverly assembled statistical software designed to create hockey sticks. This method and others like it have been used to generate hundreds of hockey stick graphs. The sticks are created by small group of associated scientists with a preferred group of ‘favorite’ proxies using a variety of similar statistical methods. Now that we have these emails from the scientists themselves, it’s easy to see how they keep getting passed through peer review as evidence of global warming.

Proxy based temperatures cannot be trusted to reasonably define temperature history. Basing economic and energy policy on this curve and others like it are the height of irresponsibility.

But it will continue.

32 Responses to “Ten Things Everyone Should Know About Hockey Sticks”

  1. twawki said

    Let’s see whether Alan Jones picks this up on 2GB radio in Sydney tomorrow, he has been scathing about the government on this. Also hopefully there will be some editorial comment in the Australian, We need to write to our politicians en-mass this week to make them aware of what is going on – as they vote on the ETS – links to their email addresses on Steven Fielding’s site (also on mine). If we keep the comments coming on Andrew Bolts blog and also get more people to sign the online petition that Barnaby Joyce has then we can give this greater momentum. Additionally if people can link/and or comment about this site, Bishop Hill, Air Vent, Climate Audit, Andrew Bolt etc and any other relevant site on their tweeter, facebook, myspace etc – it will give greater traction and exposure! Get to work people!

  2. P Gosselin said

    If I may, for the drones from RC, CRU etc. who are perhaps keeping tab:
    My open statement on this: Posted at WUWT


  3. P Gosselin said

    It’s time that Jones etc. be demanded to resign.

  4. P Gosselin said

    That’s the spirit. I’ve at this all day.

  5. Mike said

    If the whole GW debate is political, and centred around softening us up for Global Governance (by the UN?) then we also have to understand that propaganda does not have to be ‘honest’, if the internet becomes too much of an embarassment, then these guys are going to lock down on the very media upon which these issues can be discussed. The internet will become just one more ‘bread and circuses’ distraction.

    It is therefore also time to look at how to preserve a free public network for perpetuity so that these climate issues, along with the similar issues in medicine (HIV -> Aids debate) and the neo-Darwin vs Inteligent Design debates can continue until scientific TRUTH emerges.

    Science is bogged down by the need for scientists to earn a living, and ‘truth’ has now become subjectively relative to the requirements of the money men who can ‘buy’ the scientific results they require from ‘hit-men’ with phd’s for hire.

    Global Warming is a scientific shibboleth erected on a pedestal of well financed mis-information supporting a brand of fuzzy truth which somehow seems to transcend the need for any intellectual rigour, and as such it carries neo-religious significance to many of its adherrants.

  6. Dan Hughes said

    Number 11-

    There are no connections between the quantity reported to be temperature and causality by physical phenomena and processes.

    Number 12-

    The change in the energy content of a thermodynamically heterogeneous system can be determined only under certain very special conditions by the change in the temperature of that system.

    Number 13-

    The responses of the components in any natural system are determined by the local-instantaneous state of the system at the location of the component.

    Number 14-

    Somewhere in all this stuff, there’s got to be area or volume or mass averaging of the measured quantities. I haven’t seen it yet.

    Snip at will.

  7. Jeff Id said

    #6 Not much snipping happens here so people need to be careful what they write.

    In fact less than 20 in 10,000 posts.

  8. Phil said

    Good summary, but surely the divergence issue (as well documented at CA and WUWT) should be somewhere in the top ten as well? How not only do they throw away the proxies that don’t fit, but they cut-off data from the proxies that do otherwise give the “right” answer (i.e. flat except in the 20th century) when they start to refuse to match temperature measurements by stubbornly dipping from the 1960s. Which is kind of serious given its only the alleged correlation between the proxies and temperature which allows any kind of inferrence to be made in the first place. (As one can tell from Jones et al’s efforts to “hide the decline”.)

    There’s also the rather amusing teleconnection farce, whereby they can write the widely documented (across thousands of miles) medieval warm period and little ice age out of history because they were “purely a Northern hemisphere phenomenon” and replace them with an inferred temperature record from 18 trees on one mountainside as apparently these could record temperature on a global basis!

  9. HalifaxCB said

    Jeff ID – Nothing to add other than thanks for getting the baby up and walking!
    A question – are there independent people now involved in doing the modelling number crunching to open up what’s true, and what’s not, re. this “science”? Just curious, as I’m a recently retired applied mathematician with too much exposure to statistics (both the use & misuse), and would happily donate some time/thought/computing to anyone willing to actually do this the way science is supposed to be done – out in the open, a darwinistic free-for-all…..Anyway, if you have links to such, could you post them?

    As an aside, loved the title from the other article, “Baby Steps”, I couldn’t help but think of the movie, “What About Bob?”….

  10. Erik said

    Thanks for the site all of the good posts and comments.

    I am not only interested in the climate debate, I am also a part time student of history.

    Could anyone recommend some good data on what we think we know/knew of the climate historically? There is a book by Lamb in my basket at amazon. Is it any good? Google is hardly a friend in this matter since all the previous knowledge comes from pre-hockey-stick days when there was no web.

    Not so long ago, the Medieval warm period was uncontroversial. If tree rings as proxies are unreliable, what do we know and how? Archeological evidence?

    I was born in Kiruna, in northern Sweden, and next to the church of Jukkasjärvi, there is a small chapel where corpses of those who died during the winter were stored until summer. From practical experience I can say that the Vikings that now lie in permafrost on Greenland were not buried in frozen ground.

    I posted the following on in heated debate on another forum:

    Finally, I looked up Histoire de France, Les origines from 1994. There on, pp. 139-140 I find a discussion of the “refroidissment brutal” (sudden cooling) that took place 600-400 BC. The author claims that the agricultural zone contracted in Scandinavia from 68° north to 60° north (Kiruna to Uppsala in Sweden which is 1250 km by road) because of a fall in the average temperature of 2 °C. I would assume that this has been confirmed through archaeological evidence since it must be straightforward to look for the evidence of what grew and didn’t grow.

    We even have memories of this in the Edda and from Pytheas of Marseille (2nd half of third century BC) regarding the ice in the northern seas. Was he also bought by oil money?

    Today, I manage to get to a section in volume two where the miserable climate of 1315-1317 is discussed.

    I recently read “Justian’s Flea” which suggests that the arrival of Justinian’s plague (640-750) which killed half the population of Europe was largely brought on by the colder weather that promotes the bacteria-flea-rat store of infection. Similarly, the Black Death could have been caused by the Little Ice age. (There was no plague between 750 and 1348).

  11. Erik said

    Small correction. The Histoire de France under Jean Favier started in 1984, well before hockey.

  12. Paul said

    Ok, about #4… That’d be “experiment”, not “expirament”… Don’t give them a diversion from the points if you don’t have to…

  13. MikeN said

    experiment is spelled wrong.

    You might want to post Jon Daly’s e-mail from the CRU archive, complaining about how tree rings are not valid.

  14. Espen said

    Erik: About the cold historic periods: Did you read anything on the the fascinating theory of “Bond events”? (

  15. Frank Lansner said

    Please notice another problem in hockey sticks:

    These are based on various different series smoothed together, and furthermore, the series are often 50 year smoothed.

    So when we look at the MWP in theese graphics, we look at MEDIUM temperatures og MWP. (faulty medium temps due to tree ring problems etc)

    So the leading scientists are comparing 1998 SUPER EL NINO year temperature with MWP MEDIUM temperatures. Compring an El Nino PEAK around year 2000 with average temps of MWP.

    This is so wrong an approach, that this has nothing to do with objektive science.

    K.R. Frank Lansner

  16. Erik said


    Thanks for that link. The Iron age cold era seems to be what was referred to in my book. I notice that they refer to , so I might get it. If you by it from, it is only £30 instead of $200 from the US site.

    I notice that it only talks about the North Atlantic. Half a year ago, I read the book by Singer and Avery.

    It “feels scientific” when you read it, which of course means nothing. It is a very good read though.

    They claim that there is 1500 year oscillation period observed globally, not just in the North Atlantic. I am on a business trip, far away from my library so I can’t look up any details.

    I also read about 100 pages, so far, in by Marcel Leroux.

    I managed to get it second hand, I am not that rich. I used to be a solid state physicist in an earlier life. After having looked at Leroux book and how complex the climate system(s) of the earth are, I am inclined to believe that we will never, at least with the brain and computing power we will have the next 50 years, be able to model the temperature sufficiently well to get the details of earths climate with sufficient precision.

    On the other hand, maybe we can get sufficient data from other sources? When I first heard of Svensmark’s theories of cosmic rays, it sounded outlandish, but if there is a relatively simple mechanism that amplifies the effect of the variation of sun activity, maybe we can crack the global problem.

    The experiment will start producing data next year.

    I notice that I am veering off the topic seriously…

  17. Briso said

    11. Quote from Mann08 : “Peak multidecadal warmth centered at A.D. 960 (representing
    average conditions over A.D. 940–980) in this case corresponds
    approximately to 1980 levels (representing average conditions over
    1960–2000).” EIV reconstructs a 40 year average.

    12. Red line is a 5 yr average.

    13. Mann’s paper shows that the highest 40 yr average in the recent past has not exceeded the upper 95% confidence level of the 960 AD 40 year average. Recent warming is therefore not unprecedented.

  18. Erik said

    I just found out something of a rather circular nature that I find rather depressing.

    I thought I’d get a book by Lamb on the history of the climate. Through reading today I learn that Hubert Lamb is “considered the Father of Modern Climatology” and also the founder of CRU Hadley and that he is likely to have been mortified knowing what is going on there now. Unfortunately, he is already mortified, quite literally, so he can’t comment.

    Putting two and two together, I now also remember reading an entry on Real Climate where they complained about Lamb’s book. And these guys are his successors.

    Someone said that “there is no such thing as progress in the arts”. It seems that in climatology, things are going backwards.

  19. TurkeyLurkey said

    Hey Jeff,
    Thanks for your ongoing efforts.

    So, here’s my question;
    Are the ‘proxy’ values averaged with a different timescale than the red ‘temperature’ line?

    I don’t want to overstate the fraud case when I am discussing this with my non-sceptical relatives over pecan pie…


  20. Jeff Id said

    #19 By averaged, do you mean filtered?

  21. TurkeyLurkey said

    Well, yes, as I understand them, every (finite impulse response} (low-pass) filter operating on sampled discrete data points could probably be expressed as a ‘running average’ with appropriate weighting of the samples.

    But despite my relatively high degree of training, ‘filtered’ didn’t have the same visceral impact as ‘averaged’.

    Anyway, I rephrase; Are the Temperature Data (the provocative RED curve’ smoothed by a filter of equivalent time response as the other colored strands of spaghetti? ( and I intend no slight of spaghetti by this analogy).

  22. Jeff Id said

    #21, I believe they are similarly low pass filtered in the end, however the original data have different levels of autocorrelation and the CRU red curve has a lot of mysterious processing beforehand – so my answer is yesish. 🙂

  23. Jeff Id said

    #21 If you do have substantial training, the (hockey stick posts) link right below the header of this blog has some code for how HS are made.

  24. TurkeyLurkey said

    OK I took a look to the Mann et all ‘defoliated’ 2008 PNAS paper.

    Hmm, if the temperature series is subjected to a 40 year low-pass, how does a 0.8 degree uptick roar through that filter?

    What was the magnitude before the filter?

    What am I missing here?
    One other thing; what is the deal where non-dentro series have ‘divergance’ ? is that telling us that the temperature record is wrong?

  25. Jeff Id said

    Well to me divergence means – ain’t really temp. If it were, some lab experiments could prove it was temp, but all we have is correlation for an excuse. So correlation (if you’re familiar) loves a strong upslope. That’s why hadcrut’s higher slope is so important to dendros. There are comments to that effect in the emails also. Basically high correlation means upslope so any squiggle with an upslope is temp.

    That’s why I still laugh at this post:

    Non-dendro divergence simply means inconvenient downslope and in the case of Briffa MXD, the proper climate school response is chop it off and call it temp anyway. Consider the fact that if correlation is what proves temp and many series “diverge” during modern time (post 1850) how can we be certain of the history?

    Answer: We can’t, but that doesn’t stop them.

  26. TurkeyLurkey said

    BTW, when I saw the link, I promptly snagged it friday and right away sent pdfs of the email to the GB program, even snagging one of the references to him.

    Looks like he’s maybe among the first to take it to television.

    The radio version seemed to be well informed, and doing a ~decent job of conveying the whole complicated mess to regular folks.

    Time will tell.

  27. marius said

    there were no thermometers in ancient times (at least not any good ones), but the temperature is being established out of some old ice caps. you can always ignore, ignorance is the domain of you-alikes.

  28. Jeff Id said

    #27, I don’t think that ignorance is a reasoned assessment. Can you point to any single fact above and find a problem?

  29. Briso said

    TurkeyLurkey said
    November 23, 2009 at 4:20 pm
    OK I took a look to the Mann et all ‘defoliated’ 2008 PNAS paper.

    Hmm, if the temperature series is subjected to a 40 year low-pass, how does a 0.8 degree uptick roar through that filter?

    What was the magnitude before the filter

    Amen brother!

  30. […] January 31, 2010 I have been reading over the blog posts of Steve McIntryre and Jeff Id and others about the nuances of various constructions of the hockey stick.  I’ve been […]

  31. […] The Climategate furore shows no sign of dying down and the big questions now are “Was it a whistleblower?, “Is there more to come?” and “What next?”.  While the skeptical blogs are thriving on discovery of long-suspected foul-play, the exchanges on climate ‘insider’ sites such as RealClimate seem to be from a different planet, and take a “so what” smooth-it-over theme; however, these waters are boiling and no amount of oil will calm the maelstrom. (For anyone who is new to the scientific controversy over climate data I would highly recommend Jeff Id’s WordPress site “The Air Vent” where he currently has a reposting of his excellent Ten Things Everyone Should Know About Hockey Sticks). […]

  32. security door lock FOR LIENKUAN…

    Ten Things Everyone Should Know About Hockey Sticks « the Air Vent…

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: