Betting

UAH October 2010: 3rd warmest October in Record.

24 Comments

Roy Spencer posted the UAH temperature anomaly for October: +0.416C.

For those wanting to compare to anomalies in the record this is:

  • Down from the September, 2010 anomaly of 0.603C.
  • Up from the October, 2009 anomaly of 0.362C.
  • Higher than the October, 1998 anomaly of 0.416C.
  • Lower than the October 2005 and 2006 anomaly of 0.466C and 0.423C respectively.

So, it’s hot enough to say it’s hot, but it’s not in contention to set any sort of record for October anomalies.

If we turn to 12 month averages, we see the 12 month average for UAH rose to 0.5104C when this months anomaly took the place of October 2009 in the average. There is still a chance UAH will set a new record for the high 12 month average of 0.5159C set in October 1998 and/or set a new record for the all time calendar year anomaly of 0.5130C. (Note: My temperature anomalies baselined using the values from Jan 1980-Dec 1999.)

The 12 month running averages are shown below; the multi-model mean for surface temperatures based on model runs forced using the A1B SRES are shown for reference:

And now for the most important part of the post: Who won quatloos?

This month, the three visitors who won, placed and showed netted quatloos.
These are KÃ¥re Kristiansen who wagered 4 quatloos the anomaly would be 0.422C, MikeP, who wagered 5 quatloos the anomaly would be 0.411C, and pdjakow who wagered 4 quatloos the anomaly would be 0.41C.

It also appears spambots with enticing names like bglnzjx placed bets. Spambots must have learned to add but don’t seem adept at guessing the UAH anomaly. Most spambots predicted the troposphere would cool to 0 which is lower than even Stephan predicted. But spambot feabqtcqy must be a real warmer– it entered an anomaly of 89C, which would correspond to the oceans boiling away.

The results of bets by both real people and spambots are shown below:

Winnings in Quatloos for UAH TTL October, 2010 Predictions.
Rank Name Prediction (C) Bet Won
Gross Net
— Observed 0.419 (C)
1 KÃ¥re Kristiansen 0.422 4 51.925 47.925
2 MikeP 0.411 5 51.925 46.925
3 pdjakow 0.41 5 22.15 17.15
4 Greg Meurer 0.41 5 0 -5
5 Jiri 0.43 3 0 -3
6 Sordnay 0.43 5 0 -5
7 Michael Hauber 0.433 5 0 -5
8 Vlasta 0.4 3 0 -3
9 Robert Leyland 0.441 4 0 -4
10 Pavel Panenka 0.389 3 0 -3
11 enSKog 0.375 3 0 -3
12 TroubleWithTribbles 0.47 4 0 -4
13 Pieter 0.368 3 0 -3
14 Layman Lurker 0.48 5 0 -5
15 John K 0.482 3 0 -3
16 Denis 0.356 5 0 -5
17 Don B 0.35 4 0 -4
18 bob droege 0.488 3 0 -3
19 Chuck L 0.497 4 0 -4
20 jeff id 0.505 5 0 -5
21 Tom Harrah 0.33 5 0 -5
22 ErnieP 0.329 3 0 -3
23 Neven 0.511 4 0 -4
24 jack mosevich 0.511 3 0 -3
25 AMac 0.512 2 0 -2
26 Troy_CA 0.32 3 0 -3
27 David 0.529 4 0 -4
28 March 0.53 2 0 -2
29 Boris 0.539 5 0 -5
30 John F. Pittman 0.557 3 0 -3
31 stephan 0.17 4 0 -4
32 bglnzjx 0 1 0 -1
33 dwhpxs 0 1 0 -1
34 msqpotset 0 1 0 -1
35 redfhncwvx 0 1 0 -1
36 njcorlrij 24 1 0 -1
37 hmueseayjza 71 1 0 -1
38 feabqtcqy 89 1 0 -1

The net winnings for each member of the ensemble will be added to their accounts.

24 thoughts on “UAH October 2010: 3rd warmest October in Record.”

  1. For arithmetic-capable spambots, their performance wasn’t very impressive.
    They’d probably squander their winnings on V1agra, anyway. So I suppose we should be glad that they choose to wager their quatloos here.

  3. In the interests of promoting discussion, how does the mean of the entries compare with the published result?

    Are we a delphi oracle?

    Ignoring the spam bots (and summing stephen is not a spam bot!), I compute the mean as:

    0.431 C (with standard deviation of 0.084)

    So the ensemble-mean does not match the actual value 🙂

  4. Robert–
    Hmmm…. I do have all the numbers in a database.

    I guess I can test whether the multi-model mean is consistent with the outcome and also whether the outcome falls within the ±95% range of all predictions.

    That might be worth doing.

    But I need to do this with and without spambots, right?

  5. 🙂

    Yes please, no spambots.

    And so, are we (your loyal participants) a valid oracle?

    cheers,
    Robert

  6. It would be, sort of, like The Wisdom of Crowds, except 31 of us are probably not a crowd. Unless we were all in an elevator.

  7. Hate to lose some quatloos and the fame that goes with them, but I think the published value is .419 instead of .416, making Kare first.

  8. Lucia,

    Don’t you usually put confidence limits on the graph (yellow dotted lines) to demonstrate the significance, or lack of significance, of the difference between the IPCC projected value of ~0.2C per decade and the measured value?

  9. How come Greg didn’t win any quatloos, while that spambot pdjakow won with the same guess?

    pdjakow 0.41 5 19.553 14.553
    Greg Meurer 0.41 5 0 -5

  10. Hal– Why do you think pdjakow is a spambot?

    The reason pdjakow won and Greg Meurer did not is pdjakow bet earlier. Ties go to the person who bets first and you win according to rank. So, Greg is #4, and was edged out.

  12. Hate to lose some quatloos and the fame that goes with them, but I think the published value is .419 instead of .416, making Kare first.

    Hmmmm… I better recalculate!

  13. And so, are we (your loyal participants) a valid oracle?

    I think to test various hypotheses, I’m going to have to concoct some uncertainty intervals. . .

  15. Boris–
    It’s hard to say for sure.

    I think people program spambots to fill out all sorts of forms. They may just be entering random stuff in the fields, and these values passed through my script. I don’t know for sure the ‘bots add– they may just be entering things more or less at random to some extent.

    The ‘bots may even “learn”. They may even be trying to crack into dreamhost etc. I do program to make sure weird scripts can’t be submitted and propagate through, but that doesn’t mean bots won’t try.

    There have been several days when spambots hit my blog hard.

    I’m going to go in and make the answers include decimal places. Possibly that will repel a higher fraction of the ‘bots.

  17. Winning quatloos can be hard sometimes. I was only 0.01 degrees out and I only came 7th.

  18. So, it’s hot enough to say it’s hot, but it’s in contention to set any sort of record for October anomalies.

    I think you meant to include a “not” in there somewhere.

  20. @Hal

    I didn’t know that i’m spambot 🙂 But in fact i was very close to winners in last three bets (Jaxa minimum, september UAH, october UAH).

    Best Regards
    spambot 🙂

  21. Hi all
    I forgot to introduce my self . Iam a compulsive gambler LOL .
    This season I entered AMOS weather tipping competition
    http://tipping.amos.org.au/dist/pages/index.php#static/home

    Lucia , you may have to register to see how every round is neatly done with no hickups I saw this my 1st try here .
    After initial time input it will practicly be maintenance free , all you have to do is to enter new month value , one click and all is done .
    It wont just tell you the round winner in a second , also accumulated errors ( in this case one error would be 0.01C ) for the whole year
    Iam pretty sure Luke from AMOS would give you this program .

  22. Vlasta–
    No matter what, the program would need to be modified to run inside the wordpress blog software. I also don’t want people to need to register. I’m just pondering the best change to deal with ‘bots. I think the best change is for me to query the database to see if the person betting either a) bet before and/or b) has submitted a comment. I also think I need to code so if they fail either of those, I record the IP and block them as spammers.

  23. AARRGGHH!!

    I was close to adding an additional .001 to my number, but rounded instead. 5 Quatloos wasted.

    Oh, the shame!

Comments are closed.