Advertisement
If you have a new account but are having problems posting or verifying your account, please email us on hello@boards.ie for help. Thanks :)
Hello all! Please ensure that you are posting a new thread or question in the appropriate forum. The Feedback forum is overwhelmed with questions that are having to be moved elsewhere. If you need help to verify your account contact hello@boards.ie

November 2014 Boards forecast contest

Options
13»

Comments

  • Options
    #62
    waterways
    Registered Users Posts: 373 ✭✭waterways


    Can we sell Con? :D


  • Options
    #63
    Rikand
    Registered Users Posts: 20,341 ✭✭✭✭Rikand


    waterways wrote: »
    Can we sell Con? :D

    Can sell him as a method, which is made up of all of our inputs. As a forum, our ability to predict the average of everything is astounding and we should be selling ourselves ;)


  • Options
    #64
    Joe Public
    Registered Users Posts: 1,778 ✭✭✭Joe Public


    Someone might be selling Con already under a pseudonym, how do we know? are we entitled to royalties? will MTC be looking for a bigger cut?


  • Options
    #65
    M.T. Cranium
    Registered Users Posts: 14,353 ✭✭✭✭M.T. Cranium


    I would not deserve much of a cut because Con would be slightly improved without my portion, looking at where I am in the tables.

    Can say this is a rather anomalous outcome, consensus is usually well placed in a forecast contest but rarely ahead of all forecasts. I run a contest on the American weather forum too, and consensus was sitting in 5th place out of 21 there. Unlike in this contest, I score "normal" as well, and normal is almost dead last, so that shows that the fiield have some skill in beating random chance.

    In previous years in our own contest, I think Con was running fairly well but more like third or so, would have to look back and see.

    In my own case, I'm pretty sure without looking back at the scoring that I need to retool my max and min forecasts, I probably get far fewer points from them than the average. In general I tend to be looking for more extreme ranges than nature plans to provide. If somebody has time on their hands, look back at IMT forecasts for Con and the 21 regulars here (anyone who has missed two or fewer and has a ranking in the annual contest). It would be interesting to see if Con has the lead there or whether it comes from the other categories. To some extent, Con has an advantage, they can't easily score less than the average in any column because of the guaranteed scoring and the rank order scoring methods for the bonus. In sheer mathematical terms, it would be very difficult for the field to beat Con although I have seen outcomes where consensus is as low as middle of the scoring table.


  • Options
    #66
    M.T. Cranium
    Registered Users Posts: 14,353 ✭✭✭✭M.T. Cranium


    So, had a look back at the past three months (autumn months) and the average scores that our regular forecasters posted in the various categories. Bear in mind, these are before any time penalties. They are listed in the annual scoring order but the autumn scoring was considerably different from that. Harps and delw had only two months, everyone else in this list played all three. The Zohan and sunflower3 are at the end of the list because they are not ranked in the annual contest. The maximum scores available are shown at top of the table, and bonus is set at 12 because each month has some additional scoring points available.

    I have listed the scores for Con Sensus and the average scores at the bottom. Why are they different? Not entirely sure, probably has something to do with median and mean being slightly different.

    FORECASTER ________ IMT __ MAX __ MIN __ PRC __ SUN __ bonus

    Max score ___________ 25 ___ 20 ___ 20 ___ 15 ____ 10 ____ 12

    Rikand _____________ 20.7 __10.0 ___6.0 ___6.3 ___7.7 ____7.0
    Harps ______________ 18.0 ___6.5 __10.5 __12.5 ___6.0 ____6.5
    Bsal _______________ 18.3 ___7.3 ___8.7 ___6.6 ___7.3 ____4.0
    omicron ____________ 21.7 ___9.7 __12.0 ___8.0 ___8.7 ____9.0
    Mickger844posts _____ 21.7 __14.0 __13.0 ___7.3 ___8.0 ____8.3
    Tae laidir ___________ 19.3 ___8.7 ___4.3 __12.7 ___7.0 ____5.0
    John mac ___________ 22.0 __11.7 __15.3 ___6.0 ___8.7 ____5.0
    Jpmarn _____________ 21.3 __12.3 __14.0 ___5.0 ___7.7 ____6.7
    kindredspirit _________ 21.3 ___9.0 ___6.7 ___7.3 ___7.3 ____5.3
    Pauldry _____________ 17.0 ___6.7 ___4.0 ___5.7 ___7.0 ____6.0
    rameire _____________ 21.0 __14.7 __13.3 ___7.7 ___6.0 ____6.6
    okla ________________22.0 ___9.7 ___9.3 __10.0 ___7.3 ____4.7
    dacogawa ___________ 21.0 __13.0 __10.7 ___7.7 ___7.3 ____5.0
    dasa29 _____________ 17.7 __11.0 __12.0 ___5.0 ___9.0 ____4.3
    waterways __________ 21.7 ___9.7 ___2.7 ___8.3 ____5.7 ____5.7
    jd _________________ 18.7 __10.3 __14.0 ___4.3 ____7.7 ____4.3
    200motels __________ 20.3 __13.7 __13.0 ___8.0 ____5.3 ____5.3
    DOCARCH __________ 21.0 ___7.3 ___9.0 ___6.3 ____7.7 ____7.0
    lostinashford ________ 18.3 ___8.0 ___4.7 ___8.3 ____7.7 ____5.7
    MTCranium _________ 19.0 __10.3 ___0.7 ___9.0 ____5.0 ____3.3
    Joe Public __________ 19.0 __13.3 ___6.3 ___9.0 ____7.3 ____4.7
    delw _______________ 4.0 ___3.0 ___1.0 ___8.5 ____7.0 ____7.0

    The Zohan __________ 23.7 ___9.7 __15.0 ___4.7____6.7 ____6.3
    sunflower3 __________ 22.7 __14.3 ___6.0 ___6.0 ____7.0 ____5.3

    Con Sensus _________ 22.7 __14.0 __12.3 ___8.0 ____7.7____7.0

    Average ____________ 19.9 __10.3 ___8.9 ___7.4 ____7.2 ____5.7

    _____________________________________________________

    What we can learn from this, I suppose, is where our forecasts need the most improvement and who is doing best in each category.

    Not sure what the annual numbers would show, this is just how we've been doing recently.


  • Advertisement
Advertisement