Jump to content
  • Member Statistics

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

January 2021 temperature forecast contest


Recommended Posts

Apologies, I'm a bit late getting this thread posted. Will set a more firm deadline for entries on the morning of Jan 2nd once we see how things are going with contest entries. 

As in past contests, predict the anomalies for these nine locations in F deg, I have to wait and see what the NWS plans to do about possible 1991-2020 averages coming into play, they may not be used in 2021. If anyone knows, post the info. 

 

DCA __ NYC __ BOS ____ ORD __ ATL __ IAH ____ DEN __ PHX __ SEA

Check the December thread for updates on contest results for 2020. 

Happy new year, let's hope it's an improvement over the one fading out. (Roger over and out)

  • Like 1
Link to comment
Share on other sites

On 12/29/2020 at 3:35 PM, Roger Smith said:

Apologies, I'm a bit late getting this thread posted. Will set a more firm deadline for entries on the morning of Jan 2nd once we see how things are going with contest entries. 

As in past contests, predict the anomalies for these nine locations in F deg, I have to wait and see what the NWS plans to do about possible 1991-2020 averages coming into play, they may not be used in 2021. If anyone knows, post the info. 

 

DCA __ NYC __ BOS ____ ORD __ ATL __ IAH ____ DEN __ PHX __ SEA

Check the December thread for updates on contest results for 2020. 

Happy new year, let's hope it's an improvement over the one fading out. (Roger over and out)

The new normals are implemented simultaneously worldwide. The WMO is scheduled to meet on the new normals on May 25, 2021. The new normals should then be adopted and implemented, probably starting in June. January-May anomalies might then be retroactively adjusted to fit the 1991-2020 base period.

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

  • Rjay pinned this topic

Nice to see everyone (just about) is planning to continue, I did mention a more lenient deadline as is our custom in early January, so anyone has the opportunity to edit these up to whatever deadline I eventually set (assume it will be no earlier than Jan 2 18z) ... and I'm going to post during the day tomorrow because I haven't had any time to look at guidance (believe it or not, I do look at it) :) 

-- and happy new year, still back in 2020 here in Pacific time zone, can't wait to escape it. 

(and any late lookers, you don't need to post anything until morning of Jan 2nd, if you would prefer to wait)

Link to comment
Share on other sites

Thanks for the forecasts and continuing to support the contest -- the 2020 contest results are in, congrats to RodneyS who cruised in at the end despite a late challenge from DonSutherland1 and hudsonvalley21. RJay also beat consensus for the year. 

Welcome back so_whats_happening, and good luck in 2021 everyone. 

I will do a table of forecasts tomorrow afternoon so as previously announced, the actual on-time deadline is Jan 2 18z and anyone can feel free to edit their submissions to then, I haven't made any notes. Also if any other American Weather forum members want to get in on the contest, post away (no late penalties this month until 18z 2nd). 

  • Thanks 2
Link to comment
Share on other sites

Table of forecasts for January, 2021

Apologies, a bit late getting this together ... welcome back to the new contest year. 

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

DonSutherland 1 ________________+4.1 _ +4.4 _ +4.6 __ +5.0 _ +2.5 _ +1.8 __ +1.8 _ +0.7 _ +3.5 __ +1.86

hudsonvalley21 _________________+3.8 _ +3.7 _ +3.9 __ +2.6 _ +1.9 _ +0.7 __ +0.6 _ +1.0 _ +1.9 __ +0.93

so_whats_happening ___________ +2.6 _ +4.3 _ +5.1 __ +2.5 _ +0.9 _ --0.8 __ +0.6 _ --0.5 _ --0.8 __ +0.24

RJay ___________________________ +2.5 _ +3.5 _ +4.5 __ +3.0 __ 0.0 __ 0.0 __ +1.5 _ +2.5 _ --1.8 __ +0.24

BKViking _______________________ +2.1 _ +2.0 _ +2.1 __ +1.2 _ +1.2 _ +0.7 __ +0.4 _ +0.7 _ +1.2 __ +0.44

wxallannj _______________________+2.0 _ +2.2 _ +2.6 __ +1.6 _ +1.7 _ --0.5 __ --1.2 _ --0.2 _ +0.8 __ --0.30

___ Consensus __________________+2.0 _ +2.0 _ +2.6 __ +1.6 _ +1.0 _ +0.4 __ +0.6 _ +0.7 _ +0.8

Roger Smith ____________________ +2.0 _ +1.2 _ +0.5 __ +1.5 _ +3.0 _ +2.5 __ +4.5 _ +3.5 _ +2.5 __ +1.06

wxdude64 ______________________ +1.6 _ +0.9 _ +0.8 __ +1.3 _ +0.6 _ +0.4 __ --0.5 _ --0.4 _ --0.3 __ --0.81

Scotty Lightning ________________ +1.5 _ +1.0 _ +0.5 __ +0.5 _ +1.0 _ +1.5 __ +0.5 _ +1.0 __ 0.0 __ --0.47

RodneyS ________________________+1.4 _ +1.5 _ +2.7 __ +2.7 _ +1.0 _ --0.3 __ +1.4 _ +0.5 _ +2.0 __ +0.13

Tom ____________________________ +0.9 _ +0.9 _ +0.8 __ --0.2 _ --0.2 _ +0.3 __ --0.2 _ +0.2 _ --0.1 __ --1.03

___ Normal ______________________ 0.0 __ 0.0 __ 0.0 ____ 0.0 __ 0.0 __ 0.0 ____ 0.0 __ 0.0 __ 0.0 ___ --1.30

____________________________________________________________________________________

____________________________________________________________________________________

Color coding shows warmest and coldest forecasts. Normal is colder than all forecasts at DCA, NYC, BOS.

New feature this year, "bias" is your average departure from consensus. (not an average of absolute difference,

but an average of the nine anomalies vs consensus -- could help you figure out your tendencies). 

With eleven forecasters and consensus being the median, that will be the 6th ranked forecast. 

------------------------------------------------------------------------------------------------------------------------------

Review of scoring rules

For most forecasts, the scoring is 100 minus two points per 0.1 deg error in outcome. 

When an outcome is more than 5.0 deg above or below normal, a smaller decrease is applied, at either end of the scoring zone between 0.0 and the outcome, where equal portions near 0.0 and the outcome lose only one point per 0.1 error. Example, outcome is 7.0, zones from 0.0 to 2.0 and 5.0 to 7.0 lose one point, a forecast of 3.5 which would normally score 30 gets 50 points because it loses 30 points 3.5 to 5.0, and 20 more 5.1 to 7.0. You always score some points by being on correct side of normal. Back in March 2012 we had an outcome greater than 10.0 (DCA +12) and scored that on a totally sliding scale from 0.0 to +12.0. 

The only other scoring rule is that the maximum raw score must be 60 or greater, when it is not, then all scores are adjusted on a "minimum progression" basis with the closest forecast scoring 60, and the rest eligible to score at intervals ... with eleven forecasts, there would be nine more scores at intervals of six, and then a bottom score of zero. However, any raw score that is higher than the minimum progression score is used instead. If the raw scores were clustered from 20 to 50, then in my example, 30 might be the lowest boosted score and the lower ones (24, 18, 12, 06, 00) would be higher raw scores instead. 

These are by the way the scoring rules that I have been applying for about ten years now. 

As to late penalties, I have been trying to balance fairness with fun, as we can see in the larger picture, this is not very easy to do. The late penalties kick in at 0600z of the 1st, and amount to one per cent reductions for each six hour interval (or fraction) that you are late, up to 36h (1800z 2nd, late penalty by then is 9%) and would increase by 1% per hour thereafter although we rarely get forecasts much later than 12-18 hours after deadline. I have been lenient in 2020 on health-related late situations. My philosophy on late penalties is not to be punitive but just to try to level the playing field in case guidance improves in some situations by the time a late entrant makes a forecast. If guidance improves, a 5% scoring reduction might offset the advantage gained. Anyone has the right to submit a late revision for one or more stations, and those would be scored by the relevant late penalty. It would not be applied to forecasts unchanged from on-time submission. 

 

Link to comment
Share on other sites

First and second editions of the anomaly tracker, for 10 and 20 days ... with final anomalies added on Feb 1st.

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

11th _ (10d anom) ___ +4.7 _ +3.6 _ +4.2 __ +7.1 _ +2.9 _ --1.5 __ +3.6 _ --1.1 _ +3.5

21st _ (20d anom) ___ +4.7 _ +5.4 _ +5.7 __ +6.9 _ +1.7 _ --1.2 __ +4.1 _ +0.9 _ +3.4

 

11th _ (p20d anom)__  +2.5 _ +1.8 _ +2.0 __ +2.5 _ +1.0 _ --2.0 __ +0.5 _ --1.5 _ +1.5

11th _ (p27d anom) __ +1.5 _ +1.0 _ +1.0 __ +1.0 __ 0.0 _ --2.5 ___ 0.0 _ --1.5 _ +1.5

21st _ (p31d anom) __ +3.0 _ +2.5 _ +2.0 ___+4.5 _ +1.5 __ 0.0 ___ 0.0 _ +1.0 _ +1.5

Feb 1st _(31d anom)_ +2.6 _ +2.2 _ +2.0 ___+5.4 _ +2.9 _+2.1 __ +3.2 _ --0.1 _ +1.8

__________________________________________________________________________________

snowfalls to date ____ 2.6 __ 12.6 __ 23.1 __25.4 _DTW19.5_BUF44.0 _ 20.1 _ 0.5 _BTV34.5

____________________________________________________________________________________

11th __ A mild start to January except in the southwest and Gulf coast where slightly below normal.

The projections are based on the assumption of a modified cold pattern, not exceptionally cold, but somewhat below average

in most regions for the rest of the month to the 27th at least. Despite the widespread below average temperatures, not a lot of

snow on offer other than perhaps lake effect and a few transient 1-3" events. 

 

21st __ The milder pattern continued anyway in the east, modified cold became modified mild with no snow on the ground near

the coastal plain. Similar comment for DEN and to a lesser extent (with no snow expected) PHX and SEA.  Forecasts for the

southeast fared better though. The end of month forecasts are based on the assumption of near normal averages in

most places across the south now to 31st, trending to around -4 average anomalies closer to the Canadian border.

 

Snowfall contest amounts are updated above also. Many of those had added almost nothing until late January. Current totals

are valid end of day Jan 31st. 

 

Provisional scoring will follow using the end of month projections generated. 

27th _ ORD adjusted from +3.0 to +4.5 as cold end of month not as deep on current charts. 

31st - 1st Feb _ Monthly final anomalies have been added.

 

Link to comment
Share on other sites

  • 2 weeks later...

Final scoring for Jan 2021

These scores are based on the final anomalies in the previous post.

 

FORECASTER _________________DCA_NYC_BOS__east_ ORD_ATL_IAH__cent_ c/e___DEN_PHX_SEA__west___TOTALS

 

DonSutherland 1 _______________ 70 _ 56 _ 48 __ 174 __ 96 _ 92 _ 94 __ 282 __ 456 __ 72 _ 84 _ 66 __ 222 _____ 678

RodneyS _______________________ 76 _ 86 _ 86 __ 248 __ 50 _ 62 _ 52 __ 164 __ 412 __ 64 _ 88 _ 96 __ 248 _____ 660

BKViking _______________________ 90 _ 96 _ 98 __ 284 __ 20 _ 66 _ 72 __ 158 __ 442 __ 44 _ 84 _ 88 __ 216 _____ 658

Roger Smith ____________________88 _ 80 _ 70 __ 238 __ 26 _ 98 _ 92 __ 216 __ 454 __ 74 _ 28 _ 86 __ 188 _____ 642

___ Consensus __________________88 _ 96 _ 88 __ 272 __ 28 _ 62 _ 66 __ 156 __ 428 __ 48 _ 84 _ 80 __ 212 _____ 640

wxallannj _______________________88 _100_ 88 __ 276 __ 28 _ 76 _ 48 __ 152 __ 428 __ 12 _ 98 _ 80 __ 190 _____ 618

hudsonvalley21 _________________76 _ 70 _ 62 __ 208 __ 48 _ 64 _ 72 __ 184 __ 392 __ 48 _ 78 _ 98 __ 224 _____ 616

Scotty Lightning ________________78 _ 76 _ 70 __ 224 __ 06 _ 62 _ 88 __ 156 __ 380 __ 46 _ 78 _ 64 __ 188 _____ 568

wxdude64 ______________________80 _ 74 _ 76 __ 230 __ 22 _ 54 _ 66 __ 142 __ 372 __ 26 _ 94 _ 58 __ 178 _____ 550

so_whats_happening __________ 100 _ 58 _ 38 __ 196 __ 46 _ 60 _ 42 __ 148 __ 344 __ 48 _ 92 _ 48 __ 188 _____ 532

RJay ___________________________98 _ 74 _ 50 __ 222 __ 56 _ 42 _ 58 __ 156 __ 378 __ 66 _ 48 _ 28 __ 142 _____ 520

Tom ___________________________ 66 _ 74 _ 76 __ 216 __ 00 _ 38 _ 64 __ 102 __ 318 __ 32 _ 94 _ 62 __ 188 _____ 506

___ Normal _____________________48 _ 56 _ 60 __ 164 __ 00 _ 42 _ 58 __ 100 __ 264 __ 36 _ 98 _ 64 __ 198 _____ 462

 

____________________________________________________________________________

EXTREME FORECAST REPORT

High score must go to either the extreme warm or cold forecast, or second most extreme, for a station to qualify.

ORD _ DonSutherland1 (+5.0) was a lot closer than anyone else to the confirmed outcome (+5.4).

ATL _ Roger Smith (+3.0) wins here as ATL finishes on +2.9 or higher.  

IAH _ Don Sutherland (+1.8) has a win as IAH finished +2.1 ... Roger Smith (+2.5) has a loss by two points.

DEN _ Roger Smith (+4.5) took a win here (DEN confirmed +3.2). 

(PHX narrowly missed out with third coldest forecast high score)

The results in total are records of 2-0 for DonSutherland1 and 2-1 for Roger Smith. 

Four months qualified, all on the warm end of the forecasts. 

__________________________________________

Congrats to DonSutherland1 and all the crew after quite a close first month generally speaking, narrow range of scores. 

(note _ edit on March 3 was to correct small addition mistakes in scoring revealed by work on annual scoring after Feb)

 

 

 

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...