Jump to content

  • Member Statistics

    17,507
    Total Members
    7,904
    Most Online
    SnowHabit
    Newest Member
    SnowHabit
    Joined

2014 NATL Tropical-Hurricane-Seasonal Forecast Contest

Roger Smith

Recommended Posts

  • Replies 193
  • Created
  • Last Reply
Roger Smith

Roger Smith

Posted
  • Author
Posted

Provisional scoring update for September 2014

 

 

September scoring based on 2/1/1 ... will be updated again if necessary. September score is worth 16 of the 50 points in the monthly portion.

 

NUMBERS IN RED ARE SCORES. MONTHLY FORECASTS FOR JUNE ARE SCORED to MAX of 4.0, JULY 6.0, AUGUST 12.0, SEPTEMBER 16.0. ... SEASONAL FORECASTERS WHO DID NOT ISSUE MONTHLY FORECASTS WERE GIVEN A PROVISIONAL BASED ON THEIR NUMBERS COMPARED TO "NORMAL 1989-2014" and THESE SCORES ARE REDUCED BY 50%. SLIGHT LATE PENALTIES APPLY TO A FEW SCORES.

 

<< scoring update October 1st, June, July, August and  Sept scores are in red

... new as of October 14th, provisional seasonal based on 8 6 2, will be updated if this changes and confirmed when final numbers available ... gives you a rough idea how your seasonals are doing (/50). >>

 

 

FORECASTER _________ SEASONAL ____ JUNE ___ JULY ____ AUG ____ SEP ____ OCT ____ NOV-DEC

 

____________________ S _ H _ M ____ SHM____ SHM_____ SHM ____ SHM _____ SHM _____ SHM_____ max

 

Normal 1989-2013_____ 14_7_3 27.0 __ 100 3.5 __ 100 5.5 __ 421 8.0 __ 531 7.0 __ 3 2 1 __ 0 0 0 ________

 

 

Uncle W _____________ 14_7_1 27.0  __ 100 3.5 __ 310 4.8 __ 530 5.0 __ 42112.0 __ 1 1 0 __ 0 0 0 _______

 

Roger Smith __________ 13_8_2 32.0 __ 000 4.0 __ 210 5.5 __ 32110.0 __432 9.0 __ 3 2 1

_______________________________________________________________________ cat 3 e NC late Aug

 

Jim Marusak __________ 13_6_1 34.0 __ 100 3.5 __ 100 5.5 __ 31010.0 __ 531 7.0 __ 2 2 0 __ 1 0 0

_______________________________________________________________________ cat 2 w-e across ncFL

 

Rjay ________________ 13_5_2 34.0 __ 100 3.5 __ 210 5.5 __ 421 8.0 __ 42112.0 __ 2 0 0 __ 0 0 0

 

IsentropicLift _________ 13_5_1 33.0 __ 100*1.8 __ 100*2.8 __21011.0 __ 532 6.0 __ 1 0 0

_______________________________________________________________________ cat 2 FL

 

Solak _______________ 12_5_2 39.0 __ 210 2.0 __ 100*2.8 __ 421*4.0 __ 421*6.0 __ 2*1*0*

 

NHC May outlook @___  12_5_2 39.0 __ 100*1.8 __ 100*2.8 __ 421*4.0 __ 421*6.0 __ 2*1*0*

 

Ground Scouring ______ 12_4_2 37.0 __ 100* 1.8 __ 200 5.5 __31010.0 __ 421*6.0 __ 2*1*1

_______________________________________________________________________ one 125kt or +

 

metalicwx366 _________12_3_1 33.0 __ 100 3.5 __ 110 6.0 __ 32110.0 __ 42112.0  ___ 1 1 1

 

Harry ________________11_5_2 43.0 __ 100* 1.8__ 100*2.8 __ 311*4.5 __ 421*6.0 __ 2*1*0*

 

Mike Ventrice __________11_4_1 40.0 __ 000 4.0 __ 100*2.8 __ 310*5.0 __ 421*6.0 __ 2*1*0*

 

CandymanColumbusGA __11_3_1 37.0 __ 200 2.5 __ 210 5.5 __ 31010.0 __ 32013.0 __ 1 1 0

 

Thewxmann ___________11_2_0 31.0  __ 200 2.5 __ 210 5.5 __ 410 8.0 __ 41012.0 __ 0 0 0 __ 0 0 0 ___

 

Eyewall 2005 __________10_5_2 46.0  __ 100*1.8__ 100*2.8 __ 321*5.0 __ 421*6.0 __ 2*1*0*

 

Isotherm _____________10_4_2 44.0 __ 100 3.5 __ 210 5.5 __ 311 9.0 __ 32114.0 __ 2*1*0*

 

Srain ________________10_4_2 44.0 __ 100*1.8 __ 100*2.8 __ 311*4.5 __ 421*6.0 __ 2*1*0*

 

wxmeddler ___________10_4_1 43.0 __ 100 3.5 __ 210 5.5 __ 31010.0 __ 42112.0 __ 1 0 0 (-5%)

_______________________________________________________________________ FL (TS or minor H)

 

Peabody Flood ________ 10_4_1 43.0 __ 100*1.8 __ 100*2.8 __ 310*5.0 __ 421*6.0 __ 2*1*0*

 

U Thant ______________10_3_2 41.0 __ 100 3.5 __ 210 5.5 __ 311 9.0 __ 531 7.0 __ 1 0 0

_____________ (9-2) __ 11_6_2 (44.0x.08=3.5)

 

superstorm93 ________ 10_3_1 40.0 __ 100*1.8 __ 100*2.8 __310*5.0 __ 411*6.5 __ 2*1*0*

 

 

Consensus (median) ___ 10_3_1 40.0 __ 100 3.5 __ 110 6.0 ___311 9.0 __ 42112.0 __ 2 1 0

 

 

TropicalAnalystwx13 ___ 10_2_1 36.0 __ 100*1.8 __ 100*2.8 __ 31010.0 __ 32013.0 __ 1*0*0*

 

wildweatherman179 ___ 10_2_0 34.0 __ 100*1.8 __ 200 5.0 __ 310*5.0 __ 410*6.0 __ 1*0*0*

 

wxmx _______________ 9_4_1 45.0 __ 200 2.5 __ 210 5.5 __ 421 8.0 __ 521 9.0 __ 2 0 0

 

Isohume / CSU #______ 9_3_2 43.0 __ 100#3.5 __ 100#5.5 __311#9.0 __321#14.0 __ 110#

 

blazess556 ___________ 9_3_1 42.0 __ 100 3.5 __ 100*2.8 __ 311 9.0 __ 321*7.0 __ 1*1*0*

 

hurricaneman _________ 9_3_1 42.0 __ 100 3.5 __ 210 5.5 __ 421 8.0 __ 32114.0 __ 1 0 0 __ 1 0 0

__________________________________________________________________________ 135kt FLL Aug 22

 

ncforecaster89 ________ 9_3_1 42.0 __ 100 3.5 __ 210 5.5 __ 311 9.0 __ 42112.0 __ 2 0 0

 

Stebo _______________ 9_3_1 42.0 __ 100 3.5 __ 210 5.5 __ 311 9.0 __ 32114.0 __ 2 1 0 __ 1 1 0

 

Icebreaker5221 _______ 9_3_1 42.0 __ 100*1.8 __ 210 5.5 __ 32110.0 __ 32114.0 __ 1 1 0

 

bkviking _____________ 9_2_0 36.0 __ 100 3.5 __ 100 5.5 __ 31010.0 __ 41012.0 __ 0 0 0 __ 0 0 0 ___

 

Amped ______________ 9_1_1 33.0 __ 100*1.8 __ 100*2.8 __300*4.0 __311*7.0 __ 1*0*0*

 

Pluffmud _____________ 8_4_1 46.0 __ 100*1.8 __ 100*2.8 __210*5.5 __ 321*7.0 __ 1*1*0*

 

GaWx ________________8_3_1 43.0 __ 100 3.5 __ 110 6.0 __ 31010.0 __ 42112.0 __ 1 1 0 ______

________________7-1__ 8_4_1 (46.0x.7=32.2)

 

andyhb ______________ 8_2_1 39.0 __ 100 3.5 __ 200 5.0 __ 310 9.0 __ 32114.0 __ 1 0 0

 

torchageddon _________ 8_1_0 32.0 __ 100 3.5 __ 000 5.0 __ 200 9.0 __ 510 9.0 __ 0 0 0 __ 0 0 0

________________________________________________________________________ 45kt TS e/cFL Sep 9

 

Ed Lizard _____________7_3_1 42.0 _1.500 3.2 __ 100*2.8 __ 210*5.5 __ 211*8.0 __ 1*1*0*

 

biodhokie ____________ 6_3_1 40.0 __ 000*2.0 __ 100*2.8 __ 210*5.5__ 211*8.0 __ 1*1*0*

 

PSUBlizzicane 2007_____ 5_1_0 26.0 __ 100 3.5 __ 200 5.0 __ 21011.0 __ 32014.0 __ 1 0 0

_______________ 7-31 _ 7_3_0 (40.0x.4=16.0)

_______________ 9-2 __ 7_5_0 (45.0x.07=3.2)

 

_______________________________________________________________________________

 

monthly forecasts not associated with a seasonal forecast are not listed in the table, for your score look for somebody with the same forecast (not marked *). Let me know if your username changed and I should merge these monthlies with different names above.

 

* monthly forecasts not issued, provisional values assigned, 50% scoring penalty applies.

 

@ Isohume posted a range for seasonal, these are means from the range, and he attributes the forecasts to CSU. I have decided to score these Isohume-CSU provisionals at 100% because although they were not issued in monthly formats this will provide the highest available comparison scores for the mean CSU seasonal numbers in our contest format. I will probably add a table making this same assumption about all forecasters here although the official contest rules will determine the announced winner(s). This will let everyone compare "what if I had participated fully" paradigms.

 

# NHC May outlook added to table with provisional monthlies, Oct 18th.

 

Seasonal numbers are now shown provisionally based on 3/2/1 additional to reach 8/6/2. These values will change if a different seasonal total is actually reached. No in-season updates qualified to replace original forecasts under the rules. These provisional scores are based on the assumption that one more storm will develop late in October or in November. It could well be that 9/7/2 is the best actual estimate to use (a post has now been added with those scores). On that basis, final scoring is probably still quite wide open and the more active the season now becomes, the higher up into the table you might want to look for best scores. Quite a few of the best seasonal scores will be hampered by non-participation in monthly forecasts if that's the case.

 

 

 

 

Link to comment
Share on other sites
Roger Smith

Roger Smith

Posted
  • Author
Posted

You guys weren't late, the deadlines were extended to 06z Sept 2 due to Labor Day ... so now the penalty clock is ticking and I am inserting the provisional September forecasts into the table. Provisional means a scaled version of the normal monthly count related to each seasonal forecast. These are scored at 50% ... not even sure why now, seemed like a great idea in June.

 

Sort of like my seasonal forecast. :)

 

Okay then, anyone who would prefer to void the provisional and substitute your actual September forecast, at a slight discount (1% per hour starting 06z today), table will be edited.

Link to comment
Share on other sites
wxmx

wxmx

Posted
Posted

First contest rule (first post in this thread):

 

 

 

1. Post a forecast of seasonal totals, in the format ALL NAMED STORMS / HURRICANES / MAJOR H on or before June 1st (deadline June 2 06z). Late penalty will be 1% per day (06z to 06z) which means that if you enter the contest on July 1st you'll have a 30% penalty applied to this portion, etc. The seasonal forecast will be worth 50% of the total points.
Just realized that there's an implied shift in the rules from previous years, in that the rules did not state any opportunities to update seasonal forecasts after your first entry (which quite a few have already made). Thus, to amend that situation, you can update any time you want under the same rules as first entries, namely, a 1% penalty per day starting June 2nd. So if you change your mind let's say on July 21 and figure to score better at 50% of your forecast score than 100% of your original, make the amended forecast. 
Link to comment
Share on other sites
Roger Smith

Roger Smith

Posted
  • Author
Posted

Yeah well, the change in rules from previous contests is not as big as it might seem. I think previously we had some sort of scoring system where your changes counted for a small percentage, so if you improved it, you might get let's say 3 extra points by the September update (because your original would not have scored zero and your update might not score all five of the 5/50 it was then worth). And IIRC even last year we eliminated the October 1st seasonal update.

 

Under these rules for 2014 there probably wouldn't be any advantage to an update (unless it scores very high and your original scored zero, then you would still even today get 7% of the 50 points instead of zero, so in that way, the new rules are not that different from the divided scores at optional updates) but also, you'll get your extra points from a more accurate monthly outlook as September is worth 16/50 of those points. In other words, you're probably gaining more on the field if you're correct about an active September, by (.a.) making a forecast and not taking the 50% provisional, and (.b.) gaining possibly half or more of those 16 points.

 

However, I will note your updates for later discussion and we can do a post-mortem on what they would have scored under rules used in previous years. I honestly don't think they would change total points very much, and this way, the contest is very much anchored by pre-season (early season?) seasonal forecasts.

 

Here is a "what if" to illustrate using the changes made by PSUBlizzicane2007 and assuming the current seasonal update is the final outcome.

 

SEASONAL FORECASTS ~June 1 ___________ 7-31 update _______ 9-2 update ____________ TOTAL SCORE ( /50)

 

2014 contest ______ 5/1/0______________________ 7/3/0 ___________ 7/5/0

 

2014 scoring ________ 37 ____________________ 47 x .40 _________ 50 x .1 _______ 37

(highest of these...)____37 ____________________ 18.8 _____________ 5.0

 

2013 scoring ________ 37 x .7 _________________ 47 x .2 _________ 50 x .1 _______ 40.3

(sum of these...) ____25.9 _____________________ 9.4 ____________ 5.0

 

Note that although the updates improved the 2013 type score by 3.3, the changes have not been mathematically very large under our scoring system and therefore no update made after June 27th in the 2014 system would have improved the score. Updates can only improve scores for really low-scoring seasonals.

 

Now let's compare an actual outcome of 11/7/2 to find out what if any improvements to score would occur with these same forecasts in 2013 ...

 

2014 scoring ____  5 _____________________________ 27 x .40 _________ 34 x 1.0 _______ 10.8

(highest of these..) 5 ______________________________10.8 _____________ 3.4

 

2013 scoring ____ 5 x .7 __________________________ 27 x .2 _________ 34 x .1 _________ 12.3

(sum of these...) __3.5 ____________________________ 5.4 ____________ 3.4

 

In that hypothetical situation, your July 31 update improves your score by 5.8 and the 2013 system would have improved it by 7.3. Also in this situation, any update made before August 29th could have improved the score slightly.

 

Bottom line is, the 2014 scoring will only penalize late seasonal updates by a small number of points and those can be gained back by participating in the monthly, for example, if 5/3/1 was correct and your seasonal implied 3/2/0, you would gain 5 points. If the actual turned out even higher your improvement would gain more points (as the difference in squared errors mounted).

Link to comment
Share on other sites
  • 2 weeks later...
Roger Smith

Roger Smith

Posted
  • Author
Posted

Month now 2/1/1

 

Seasonal now 5/4/1

 

It's an "oddsmakers delight" on who might win, place or show in this contest (either seasonal or combined) -- almost everyone seems to remain in the hunt but much different scenarios are required from now to end of season, one group need almost nothing more to happen, another group need a few storms and a third group will require very active conditions late September and October.

Link to comment
Share on other sites
  • 2 weeks later...
Roger Smith

Roger Smith

Posted
  • Author
Posted

So, the scoring table is updated and your entries for October are being edited into it as they arrive. I will say 3/2/1 just to get a bit deeper into the hole I already dug for myself. :) ... for money I would probably say 0/0/0.

 

Had a look at the few entries regarding maximum landfalling U.S. cane, well, slight pat on the back there for contest host, right location, bit strong and way late. Other ideas no closer in most regards, so I guess that's the "winner" so far. Jim Marusak and Isentropic Lift could be considered joint second place as they had cat-2, the correct intensity, but not too close on location or timing. Has to be said that there could yet be a change in this assessment if a major hurricane makes landfall in October.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

Go to topic listing

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...