Time to bet on Aprils UAH TLT. You all know the drill. If you don’t, ask.
[sockulator(../musings/wp-content/uploads/2011/UAHBets5.php?Metric=UAH TTL?Units=C?cutOffMonth=4?cutOffDay=17?cutOffYear=2012?DateMetric=April, 2012?)sockulator]
Cut off date: April 17!
This is, by the way, a new improved form. The previous form somehow permitted a mischievous person to enter no name and no email. (Or at least, do something that resulted in no name or email being stored in the database, which is not entirely the same thing.)
The effect was entirely non-dangerous to the blog, betting and the intertubes in general. However, it caused the display script to hiccup when showing final results. For some reason, the March betting results ended with the bet by ” “. I fixed the display by giving anonymous a name.
cut off date is?
April 17– sorry. I guess I was leaving that to your psychic powers.
“If you don’t ask.”
You don’t get?
James E, you missed the comma. Or did Lucia just add that?
Niels– I added it! I should have put up a note– but it was ohh… 11 pm.
Lucia –
It was hardly an SkS sneaky dishonest edit. It was a comma!!
I think by and large you should be free to edit for typos, spelling and clarity as and when you feel like it – we’re all a bit precious about the sacrosanct nature of an original post on the web.
***
But then again, if you want to put notes up any time you correct anything, that’s cool too.. 🙂
Anteros–
I usually try to remember to thank people who find my typos. Plus, anyway, James’s quip was funny. It made me laugh– but I went and changed and forgot to add a comment in response. Then that confused Niels.
Anteros:
To me, the determining factor has always been one of creating confusion. Could the unmarked edit create confusion? If so, it needs to be stated. If not, it doesn’t make much difference.
In this case, I’d say it needs to be stated because someone specifically pointed it out, and his comment doesn’t make sense if readers aren’t aware of the edit. Of course, failing to (immediately) make a notification is hardly some horrible offense. Slip-ups happen.
Brandon– I see it your way. I think most especially sneak edits shouldn’t make a previous comment look like the poster might have been out to lunch. In the case of corrections for typos, I think the best is to thank the person who corrected the typo in comments or to have a stated policy that comments about typos will be deleted after the typo is fixed. The only justification for the latter is to keep discussion focussed. But this is a post on letting people bet. Discussoin doesn’t need to be all that focused.
Plus, the issue of sneak edits in comments has been a lively one owing to SKS’s history of sneak edits that made commenters posting meaningful comments seem out to lunch after the post, comments and inline responses to comments by SKS were all changed. So I’d rather just acknowledge most typos corrections if they were triggered by someone making a comment– not doing so was my oversight.
I have sometimes found small typos myself and corrected with no particular comment. I’m not going to write an update announcing that I noticed I’d typed teh and corrected that to the.
Happily, I agree with everyone.
Had Lucia not inadvertently omitted a comma, we wouldn’t have had James’ quip and Niels’ astute guess that the comma had been surreptitiously added after James’ comment.
Brandon – I agree that the criterion of common sense is whether the edit has the potential for causing confusion (among other things) but intent is worth bearing in mind as well.
On this particular topic I should say that although there are a large number of things about SkS that I find quite offensive – particularly the dissembling and dishonesty – I thought John Cook’s explanation for how the post hoc editing came about was quite convincing. I think it was rather more innocent than it probably appears.
Anteros:
I’d be much more inclined to believe that if not for the fact their editing practices are simply dishonest. For example, I recently had a comment there moderated to remove a number of things, including calling using uncalibratable data upside down “nonsensical.” I was told to refrain from “personal characterizations.” At the same time, Tom Curtis flat out insulted me and accused me of being dishonest.
With behavior like that, I don’t trust any excuses SkS might give. Sure, what they say may have happened. On the other hand, maybe they (or at least some people at SkS) were fully aware of what happened and just didn’t mind.
I just want to wish Lucia and everyone a Happy and Blessed Easter. 🙂
Andrew
Thanks Andrew,
Mine started out exciting. We got a call. Jim’s Dad had a fall. I’m just back from the hospital and hospital visit induced errand running. (Jim’s Dad is ok. But the CT scan etc. sucked up time!)
I’m going up to Highland Park for my families dinner. Jim is mostly staying with his Dad but also visiting his mom.
Lucia, slightly off topic but do you (or someone else) know what happened to HadCRUT3 ? It is April 10 now, and there still is no update of the February data. Have they all gone home, until the data finally listen to their predictions?!
wilt
The February HadCRUT3 figure was published during late March.
The global figure was 0.192c.
Are you looking on the CRU site for this data?
If so, it might be better to use the UKMO site here:
http://www.metoffice.gov.uk/hadobs/hadcrut3/diagnostics/index.html
You have to click on “Mean of northern and southern hemisphere averages”, then scroll down to the monthly graph and click on the “Data file” link.
It’s a bit of a faff finding the data on the UKMO site, as it is well hidden, but they usually publish a lot earlier than CRU.
Ray, thank you for the link to MetOffice. They surely are hiding the data well. Mind you, I did NOT write that they try to hide the decline ….
Remarkably, however, the monthly average again is lower than the previous month, for the seventh time in a row now!
wilt,
Glad to help.
A link to the other UKMO datasets is here:
http://www.metoffice.gov.uk/hadobs/index.html
Normally you have to click on the dataset and then the “various diagnostics” links on the next page, to get the uncompressed data files.
I somehow think that the run of lower monthly figures is about to come to an end with March!
wilt–
I subscribe to a reporting tool that tells me when Hadcrut is posted– but it’s for ‘3’. I need to find the best page for ‘4’. But the 10th is not that late for HadCrut to come out.
Lucia,
April 10th would not be late for the March HadCRUT3, (in fact it would be very early), but it would be late for the February figure!
Lucia,
HadCRUT4 isn’t technically out yet, but will be available here when its released: http://www.metoffice.gov.uk/hadobs/hadcrut4/data/download.html
In the mean time you can get a pretty close approximation by using HadSST3 and CRUTEM4 weighted by the respective ocean and land areas of the earth.
Zeke–
I’m not desperate for the values. 🙂 I figure NOAA, GISTemp can hold down the fort for a while.
Ray–
Yes. April 10 is late for February values.
Hadley is being late during the transition. I don’t think this is “hiding” data in anyway. Most of us know there is some sort of transition going on between “3” and “4”. Personnel are likely not making getting monthly values for “3” out a time critical task.
I got an alert for GISt’s “table C”. March was up. I usually don’t r un scripts until they post on their other page. (Also, sometimes table C seems to change … Or at least it has in the past. I’ve seen loonie numbers appear for … oh… an hour and then vanish. So, I wait for the slightly more official page. I suspect Table C is on some sort of automatic script and not watched.)
Lucia,
Just to clear up any confusion, the February HadCrut3 figures *have* been published, on the UKMO website, but not the CRU site, which is not unusual. See above.
Lucia, can you please give us a yell when the cut off date approaches? I missed out the last round by a few hours. I hoped to check on the most recent data on UAH before making an informed guess but I forgot to bet altogether 🙁
sHx–
Do you mean write a post on April 16th warning you?
Lucia, April 15 or the 15th of any month would do fine.
On the other hand, it’d be even better if you open the bets 1/3 into the month or even a little later. That’s what you used to do in the past. Now you open the bets too early even though it remains open for longer but it also makes it easier to forget to bet.
Anyway, never mind about this month. I’m just gonna take my chances now based on last month’s value lest I forget to bet again. Ho hum! 🙂
sHx–
I always thought my timing was just haphazard. 🙂
When bets open and close early, people need to be more skilled. If they close late, they need to go watch Channel 5 and then figure out the correlation, and guess.
test
test again.
Lucia
Have you considered a new “bet” incorporating “CAGW/CACC” as it is actually happening against computer models. Not just temperature, which is merely one of the variables.
How about best available multi-variate empirical data vs. best model projections?
Just a thought.
lucia (Comment #94277),
I’ve been watching Channel 5 and my guesses haven’t even been close.
skeptikal,
what math are you doing on it?
Skeptical– Channel 5 is still generally better than nothing. But it may no longer be much better than knowing last months anomaly!
lucia,
what do you mean “no longer”? was it ever? I think so, but I don’t know why it would have changed recently. if anything i think it correlates more than years ago when more older satellites were flying?
BillC,
I haven’t been doing any math on it yet. I’ve just been using the previous months anomaly as the starting point and watching Channel 5 to see if it looks like it’s going up or down from that.
I’ve only been playing this guessing game the last couple of months. I’ve downloaded lots of data and I’m still trying to work out what’s useful and how to best use it. So far, the best equation I’ve come up with only gets me an accuracy of plus or minus 0.2 which is no better than a WAG.
Skeptikal,
I more or less agree…just note that you can’t tell from looking at Ch 5 what the anomaly is doing as much as you can tell what the actual temp is doing, which isn’t much help. Of course if you look at say 2011 vs 2012 and you see what the anomaly was reported for 2011 in that same month, you are doing a mental calibration and may be able to get as close as anything you could do with the numbers anyway.
For those who are interested, here is my quick analysis of the betting for April:
MAX 0.650
MIN -0.054
MEAN 0.138
MEDIAN 0.132
STD DEV 0.105
MEAN 1-33 0.134
MEAN 34-66 0.142
Total no. of bets was 66, with very little difference between mean and median. There was a slight upward trend during the course of the betting, although that would have been greater without the bet of 0.65c during the early part of the betting. While the +/- 1 SD range is between +0.033c and +0.243c, my own estimate, based on the current AQUA CH5 situation and possible trends to the end of the month is that the most likely range is between +0.11c and +0.32c. Still, the majority of bets seem to be in with a chance at this stage.
Ray–
Do those include Anteros? I doubt it because I snuck that in later. Also… there is a bot bet. (I think it’s a bot.) Did you include that?
Lucia,
No my figures didn’t include Anteros and did include the bot, since you included it in the list and had a question mark.
Anyway, the figures including Anteros and excluding the bot are:
MAX 0.650
MIN -0.054
MEAN 0.140
MEDIAN 0.132
STD DEV 0.104
MEAN 1-33 0.143
MEAN 34-66 0.137
MEAN PLUS 1 SD 0.244
MEAN MINUS 1 SD 0.036
The changes now mean that the betting decreased slightly between the first 33 and the second 33, but that is influenced by IainT’s bet of 0.65c, early on in the betting.