Jump to content
  • Member Statistics

    17,502
    Total Members
    7,904
    Most Online
    Weathernoob335
    Newest Member
    Weathernoob335
    Joined

January 2020 temperature forecast contest


Recommended Posts

Here comes the year 2020, do you believe it? I don't. (10 point penalty)

I will be beating the drums around the regional forums to see if we can improve our turnout, but as long as the regular folk want to keep going, the contest threads will appear until it's just me and three other people (after that, maybe not) ...

Our task as always is to predict the anomalies (relative to 1981-2010) for these nine wonderful locations:

DCA _ NYC _ BOS __ ORD _ ATL _ IAH ___ DEN _ PHX _ SEA

Since I hope to keep up the publicity over the new year's holiday, there will be no late penalties imposed until after Jan 3 0600z (which is late Thursday evening of Jan 2nd). This year (and this applies to the later start for January) the late penalties will be 1% for every portion of four hours late through first 36h (to 18z of 2nd normally, and 4th for January's contest). That takes us to 9 points possible deduction, and then it would be 1 more percentage point per hour until 91 more hours have passed (which takes us to 13z of the 6th). For January, I will increase these to 2% at whatever time is needed to reach that same absolute cut-off time. 

Good luck -- the 2019 contest final results should be available some time late on January 1st. It's a close fight between RodneyS and wxdude64, looking too mild for my chances. 

  • Like 1
Link to comment
Share on other sites

Well we have a soccer team at least (eleven entrants including all nine of the hardy crew). 

I am going to extend the free entry period one more day and post notices on all subforums, to see if we can encourage a few more entries. Post your own words of encouragement on my invitation threads, it might help. You can edit your forecasts too, I won't look at anything for a table of entries until Saturday. 

  • Like 1
Link to comment
Share on other sites

34 minutes ago, Roger Smith said:

Well we have a soccer team at least (eleven entrants including all nine of the hardy crew). 

I am going to extend the free entry period one more day and post notices on all subforums, to see if we can encourage a few more entries. Post your own words of encouragement on my invitation threads, it might help. You can edit your forecasts too, I won't look at anything for a table of entries until Saturday. 

I pinned it in my subforum.  

Also i think it'd be a good idea to post a reminder in the subforums at the end of each month. 

  • Like 2
  • Thanks 1
Link to comment
Share on other sites

On 12/26/2019 at 5:05 PM, Roger Smith said:

Here comes the year 2020, do you believe it? I don't. (10 point penalty)

I will be beating the drums around the regional forums to see if we can improve our turnout, but as long as the regular folk want to keep going, the contest threads will appear until it's just me and three other people (after that, maybe not) ...

Our task as always is to predict the anomalies (relative to 1981-2010) for these nine wonderful locations:

DCA _ NYC _ BOS __ ORD _ ATL _ IAH ___ DEN _ PHX _ SEA

Since I hope to keep up the publicity over the new year's holiday, there will be no late penalties imposed until after Jan 3 0600z (which is late Thursday evening of Jan 2nd). This year (and this applies to the later start for January) the late penalties will be 1% for every portion of four hours late through first 36h (to 18z of 2nd normally, and 4th for January's contest). That takes us to 9 points possible deduction, and then it would be 1 more percentage point per hour until 91 more hours have passed (which takes us to 13z of the 6th). For January, I will increase these to 2% at whatever time is needed to reach that same absolute cut-off time. 

Good luck -- the 2019 contest final results should be available some time late on January 1st. It's a close fight between RodneyS and wxdude64, looking too mild for my chances. 

DCA +6.0

NYC +5.0

BOS +4.5

ORD   0

ATL + 6.0

IAH +3.0

DEN -3.0

PHX -4.5

SEA -4.0

  • Like 1
  • Thanks 1
Link to comment
Share on other sites

Thanks for entering the Jan 2020 contest. Welcome to about half a dozen new and returning forecasters, hope you will continue to participate. And belated thanks to RJay whom I believe has been helping us out through 2019 by pinning these threads on a regular basis. So here are the forecasts. I have added a summary of scoring procedures for the benefit of our new entrants. Your forecasts appear in the same order as DCA forecasts from warmest to coldest, if those are tied, then NYC determines it, etc.

Table of forecasts for January 2020

FORECASTER __________________ DCA _ NYC _ BOS __ ORD _ ATL _ IAH __ DEN _ PHX _ SEA

Brian5671 ____________________ +6.0 _ +5.0 _ +4.5 ___ 0.0 _ +6.0 _ +3.0 __ --3.0 _ --4.5 _ --4.0

dwave ________________________ +4.3 _ +4.5 _ +2.2 __ +1.2 _ +1.7 _ +0.3 __ --1.4 _ --2.9 _ --3.0

RodneyS ______________________ +4.3 _ +3.5 _ +3.5 __ +4.2 _ +4.0 _ +3.5 __ +0.2 _ --1.7 _ +0.3 

RJay __________________________ +4.2 _ +4.6 _ +4.2 __ +2.0 _ +2.0 _ +0.5 __ --1.0 _ --2.2 _ --2.5

wxallannj ______________________+2.8 _ +3.0 _ +2.2 __ +0.7 _ +2.2 _ +1.0 __--0.7 _ --0.6 _ --0.7

yoda __________________________ +2.3 _ +1.8 _ +0.9 __ +0.4 _ +2.2 _ +2.0 __ --1.7 _ --0.7 _ --0.3

Don Sutherland.1 ______________ +2.2 _ +2.4 _ +1.3 __ +1.5 _ +1.0 __ 0.0 __ --1.0 _ --1.5 _ +0.4 

BKViking ______________________ +2.2 _ +2.0 _ +1.7 __ +0.5 _ +2.0 _ +1.3 __ --1.0 _ --1.0 _ --1.0

 

___ Consensus ________________ +2.2 _ +1.8 _ +1.3 __ +0.7 _ +2.0 _ +1.3 __ -1.0 _ --1.0 _ --0.4

 

JakkelWx ______________________+2.1 _ +1.8 _ +1.7 __ +0.7 _ +2.3 _ +2.2 __ --1.5 _ --0.2 _ --1.7

wxdude64 _____________________ +1.7 _ +1.2 _ +0.7 __ +0.8 _ +2.1 _ +1.8 __--1.1 _ --0.9 _ --0.6

Roger Smith ___________________ +1.5 _ +1.4 _ +1.2 __ +0.5 _ +2.0 _ +0.2 __ --1.0 _ --1.2 _ --0.5

Tom ___________________________ +1.5 _ +1.1 _ +0.9 __ +1.3 _ +1.8 _ +0.6 __ --0.5 _ --0.2 _ --0.4

hudsonvalley21 ________________ +1.4 _ +1.8 _ +1.1 __ +1.9 _ +1.6 _ +0.3 __ --0.7 _ --0.6 __ 0.0

Scotty Lightning _______________ +1.0 _ +0.5 _ +0.5 ___ 0.0 _ +1.5 _ +1.5 __ +1.0 _ +1.0 _ +0.5

rclab __________________________ +0.5 _ --0.5 _ --1.4 __ --0.9 _ +1.0 _ +1.0 __ +0.5 _ +1.0 _ +1.3

___ Normal ______________________ 0.0 __ 0.0 __ 0.0 ____0.0 __ 0.0 __ 0.0 ____ 0.0 __ 0.0 __ 0.0

Rhino16 _______________________ --0.3 _ +0.5 _ +1.0 __ --0.1 _ +1.1 _ +1.3 __ --3.0 _ --2.0 _ +0.7

____________________________________________________________________

In the table above, highest and lowest forecasts are color coded. Normal is lower for ATL and tied lowest for IAH. 

Scoring procedures:

1. For most months, scores are based on 100 for a perfect call, less 2 pts per 0.1 error. Scores that would be negative remain zero.

2. For months with anomalies greater than 5.0 (+ or --), the system is similar but for the amount by which the anomaly exceeds 5.0, that portion deducts only one point (at both the low and high ends). Example, you predict +3.5 and outcome is +6.0, your score would be 50 using the first rule, but is boosted by 10 since you lose only one point per 0.1 error from 5.0 to 6.0, making your score 60. Or if you score from zero up, you have 10 points by +1.0 then add another 50 for +2.5 further correct portion. In this way, anyone with the right anomaly sign always gets a non-zero score. If a month ends up with a greater anomaly than +10 (for example, March 2012) then scores are calculated from percentage of anomaly achieved (+8.0 against +12.0 would score 2/3 or 67).

3. If neither of those procedures result in a raw score of 60 or higher from one forecaster, then we go to a "minimum progression" of scores where the closest forecast gets 60, and all others get a pro-rated value down to a zero value for the least accurate forecast. However, any progression score that is lower than your individual raw score would convert to the higher raw score, so you can only gain, not lose, from this rule. As a result of this rule, some forecast always scores at least 60. This month, with 16 forecasts, the step function will be 4 points (60, 56, 52 etc ending in zero). Tied forecasts both or all score the highest of the step values (e.g., three tied for fourth best forecast this month would all get 48 points). 

4. Consensus and Normal are scored using all the same rules. Your rank and score assigned, however, is only within the group of forecasters. Consensus and Normal can be allotted intermediate step scores in the progression system, or they may be equal to a forecaster's score. Consensus forecast is the median rather than the mean (to reduce if not eliminate any influence of one or two outliers). I am going to keep track of two scores for Normal, one a contest score, and the other a raw score based on no adjustments from rule one, that way we have a log of the total departure from normal. 

5. Late penalties are strictly applied once we get past this month. For 2020, these will be 1% for every four hours or portion late through 36h (possible 9% penalty to 18z 2nd) then a further 1% per hour.

  • Like 2
Link to comment
Share on other sites

Updated anomalies and projections:

____________________________ DCA _ NYC _ BOS __ ORD _ ATL _ IAH ___ DEN _ PHX _ SEA

(8th) ____ (7d anom) _______ +8.9 _+8.2 _+10.5 __+10.6 _+6.0 _+4.4 __ +7.1 _ --0.9 _+6.6

(15th) ___ (14d anom) _____+10.2_+10.0_+12.3__ +9.2_+10.5 _+5.7 ___ +4.1 _--1.4_ +1.0

(22nd) __ (21d anom) ______ +7.4 _ +6.9 _ +8.8 __ +5.3 _ +8.3 _+6.5 ___ +3.3 _--0.2_ +1.3

 

(8th) ____ (p14d) ___________ +7.0 _+6.5 _+8.0 __ +5.0 _ +4.0 _ +2.5 __ +3.0 _--1.0 _ +1.0

(15th) ___ (p21d) ___________+5.0 _+5.0 _+5.5 __ +4.0 _+5.0 _+3.0 ___ +1.5 _--1.0 _--0.5

(22nd) __ (p28d) ____________+5.0 _+4.5 _+6.0 __ +3.5 _+5.0 _+4.0 ___ +2.5 __0.0 _+1.8

 

(8th) ____ (24d) _____________+5.0 _+5.0 _+5.0 __ +1.0 _ +2.0 __0.0 ____0.0 __ 0.0 _ --2.5

(15th) ___ (31d) _____________+3.5 _+3.5 _+3.5 __ +2.0 _+4.0 _+2.0 ___ +2.0 __0.0 _+0.5

(22nd) ___ (31d) ____________ +4.5 _+4.0 _+5.0 __ +3.0 _+4.5 _+3.5 ___ +2.0 __0.0 _+2.0

(end of month final anoms) _ +6.4 _+6.5 _+9.0 __ +6.3 _+5.9 _+4.8 ___ +3.9 _+0.4 _+3.0

__________________ _ _______________ _ __________________

(8th) _ The month has started out very mild in all regions. Much colder air is pressing south from western Canada and will soon be impacting on those positive anomalies in ORD, DEN and SEA. The effects will be weaker or slower to arrive in the east and south. Even in the outlook period (days 8 to 16) the east coast will remain rather mild with occasional colder interludes, while the severe cold in central regions may begin to shift more to eastern Canada.

(15th) _ Large positive anomalies have continued to build over central and eastern regions while the west turned quite cold in the second week. Forecasts for this coming week begin to eat away at the larger anomalies and should see them reduced to contest-range values at some point around the 21st. This trend will peter out later in the month with a more variable regime expected. End of month projections show reduced but still fairly sizeable positive anomalies at most locations.

(22nd) _ The huge anomalies at mid-month have been somewhat reduced but not by quite as much as predicted a week ago. The trend from now to end of month looks rather close to average January values and a good chunk of the present anomalies will survive. Will post some preliminary scoring based on these estimates which are by and large higher than most forecasts except for western stations.

(Feb 1st) _ Sorry to report that life got rather hectic here and I never had a chance to create any provisional scoring. We're now into final scoring territory with the final anomalies rolling in overnight. Will be posting the scoring table soon, it is now final with all the anomalies confirmed (see table above) and the next post for the scores. At least in January I don't have the task of working on the annual scoring table (hurray for January, my favorite month). :)

 

  • Like 1
Link to comment
Share on other sites

  • 4 weeks later...

Final Scoring for January 2020

Scoring for BOS had to go to the "max 60" format since even the best forecast had a raw score of 50. Also DEN required minimum progression, the maximum raw score was only 42 (Scotty Lightning).

Scoring for ORD did not require a boost as RodneyS  prevented it with a raw score of 71. (final anom +6.3 means one point per 0.1 to +1.3 and two points per 0.1 beyond that to Rodney's max of 4.2, had anyone been in the 5.0-6.3 zone they would have been adding on a point per 0.1 there). Other scoring has a clear cut raw score above 60 to prevent any minimum progression scoring. This was a month where you could "cash in" on an extreme forecast for sure.  

^ Scores adjusted for "minimum progression" rule, max 60, scoring intervals of 4 points used. Your raw score counts if it's higher than the progression value. This applied to one score (05 rather than 04). This year I will be changing how I score Normal, it cannot receive points from minimum progression, that way we know what the actual "climate" score of Normal really is. In the old system, Normal would have scored 2 points at BOS, but this year, zero. For DEN, Normal would have scored 38 (being between assigned scores of 40 voided by tie at 44, and 36) but the climate score is 22 there. 

 

FORECASTER ________________ DCA_NYC_BOS__east__ORD_ATL_IAH__cent__ c/e __DEN_PHX_SEA__west ___ TOTAL

 

RodneyS ______________________ 72 _ 55 _ 52^___ 179 __ 71 _ 71 _ 74 __ 216 __ 395 __ 52^_ 58 _ 46 __ 156 _____ 551

Brian5671 ____________________ 96 _ 85 _ 60^___ 241 __ 00 _ 99 _ 64 __ 163 __ 404 __ 04^_ 02 _ 00 __ 006 _____ 410

RJay __________________________ 70 _ 77 _ 56^___ 203 __ 27 _ 31 _ 14 __ 072 __ 275 __ 36^_ 48 _ 00 __ 084 _____ 359

wxallannj ______________________42 _ 45 _ 48^___ 135 __ 07 _ 35 _ 24 __ 066 __ 201 __ 44^_ 80 _ 26 __ 150 _____ 351

dwave ________________________ 72 _ 75 _ 48^___ 195 __ 12 _ 25 _ 10 __ 047 __ 242 __ 16^_ 34 _ 00 __ 050 _____ 292

 

___ Consensus ________________ 30 _ 21 _ 32^___ 083 __ 07 _ 31 _ 30 __ 068 __ 151 __ 36^_ 72 _ 32 __ 140 _____ 291

 

BKViking ______________________ 30 _ 25 _ 40^___ 095 __ 05 _ 31 _ 30 __ 066 __ 161 __ 36^_ 72 _ 20 __ 128 _____ 289

JakkelWx ______________________28 _ 21 _ 40^___ 089 __ 07 _ 37 _ 48 __ 092 __ 181 __ 12^_ 88 _ 06 __ 106 _____ 287

hudsonvalley21 ________________ 14 _ 21 _ 24^___ 059 __ 25 _ 23 _ 10 __ 058 __ 117 __ 44^_ 80 _ 40 __ 164 _____ 281

Scotty Lightning ________________10 _ 05 _ 05 ___ 020 __ 00 _ 21 _ 34 __ 055 __ 075 __ 60^_ 88 _ 50 __ 198 _____ 273

Don Sutherland.1 ______________ 30 _ 33 _ 32^___ 095 __ 17 _ 11 _ 04 __ 032 __ 127 __ 36^_ 62 _ 48 __ 146 _____ 273

yoda __________________________ 32 _ 21 _ 16^___ 069 __ 04 _ 35 _ 44 __ 083 __ 152 __ 08^_ 78 _ 34 __ 120 _____ 272

Tom ___________________________ 16 _ 11 _ 16^___ 043 __ 13 _ 27 _ 16 __ 056 __ 099 __ 48^_ 88 _ 32 __ 168 _____ 267

rclab ___________________________05 _ 00 _ 00 ___ 005 __ 00 _ 11 _ 24 __ 035 __ 040 __ 56^_ 88 _ 66 __ 210 _____ 250

wxdude64 _____________________ 20 _ 12 _ 08^___ 040 __ 08 _ 33 _ 40 __ 081 __ 121 __ 20^_ 74 _ 28 __ 122 _____ 243

Roger Smith ___________________ 16 _ 14 _ 28^___ 058 __ 05 _ 31 _ 08 __ 044 __ 102 __ 36^_ 68 _ 30 __ 134 _____ 236

Rhino16 _______________________ 00 _ 05 _ 20^___ 025 __ 00 _ 13 _ 30 __ 043 __ 068 __ 04^_ 52 _ 54 __ 110 _____ 178

___ Normal _____________________00 _ 00 _ 00 ___ 000 __ 00 _ 00 _ 04 __ 004 __ 004 __ 22 _ 92 _ 50 __ 164 _____ 168

==============================================================================

Extreme forecast report

All nine locations qualify for an extreme forecast award for the warmest forecast(s), even PHX which only finished at +0.5.

Brian 5671 scoops four of them (DCA, NYC, BOS and ATL.

RodneyS has two for ORD and IAH. 

RClab has two for PHX and SEA (PHX shared with Scotty Lightning and Normal which had the highest score but cannot win solo).

and Scotty Lightning has a second one for DEN. 

Tom tied for top score with the extreme forecasts so gets a win also for PHX. 

================================================================================

Congrats to RodneyS and Brian 5671 for their top scores in various sections, also well done to RClab for the high score in the west. 

Although two locations (ATL, IAH) will be adjustable for anomalies, all scores will  move in lock-step. 

  • Like 1
Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share

  • Recently Browsing   0 members

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