February 2015 temperature forecast contest

[WEATHER53]

DCA is going to be right around -8, colder than 2007 which was the 2nd/3rd coldest Feb in 50 years.

[Roger Smith]

Well, cuz it's DCA, I was very conservative with my estimates past day seven. But if you're right all the scores will be the same anyway, except for ksammut down to Mikehobbyst who will lose some of your double points (those would shift from current  below --2.0 to below --3.0),

 

(edit Friday 11:10 a.m. EST -- Now I have changed the provisional to -8.0, as well as dropping ORD another degree. Minor changes there to the one score (mine) that had any double pointage, now nobody does).

[Roger Smith]

With the extreme temperatures this month, and in certain cases earlier too, I have been thinking about the scoring rules because there seem to be a few inequities even after I expanded the scoring ranges to provide at least some semblance of scoring when anomalies got past 5.0 in either direction. So what I will do, including January but not going back into 2014, is to use a "minimum progression" formula to determine scores if the normal method does not produce higher scores than the following ... and this rule will apply to all score sets and can bump up any particular scores so in some cases the main portion of scoring will stay the same but a few zero scores may get a slight adjustment. Penalties will then apply to these minimum progression scores so with that factored in the final results won't necessarily be this neat:

 

High score must be at least 70.

 

Second highest score must be at least 60.

 

Third highest score must be at least 55.

 

Fourth highest score must be at least 50.

 

This reduction by 5 will continue until it falls below this formula: 2n where n is the number of players remaining to be scored. At that point scores will reduce by 2 per forecaster with only the worst of all forecasts getting zero points. With the number of players we have now, that jog takes place below 30. Where we have ties, points are equal but then drop by a multiple of 2 or 5 depending on how many are tied. Normal and consensus will get whatever score is an equal distance between two actual forecaster scores if not tied with one of them. Example, normal lies between scores of 22 and 20, so it gets 21 points. Consensus is tied with 35 so it gets 35. Normal and consensus don't count in the progression. However, if a large jump occurs around where normal or consensus appear, as with normal in the NYC forecasts this month, I will score it by the logic of the system regardless of midpoint. NYC provides examples of where the minimum progression falls below some raw scores (basically from -2.5 down to -3.2 score higher from raw score than from adjusted score -- even as NYC sinks to -11.0 -- remember that in the old system scores gradually lose points to huge anomalies by a percentage reduction in case you're wondering why your -2.7 forecast gets 25 and not 27 points. Larger raw score drops took place for BOS as the provisional there fell from -10 to -12.

 

This will probably have a rather minor effect overall on scoring order but it will make the monthly scoring a bit more equal which may make the contest more fun for people (and that is always the main objective here). Let me know if you like this, feel neutral or dislike it -- if there are objections I can revise or abandon this change. Will not go back into January until I feel that this revised scoring method is generally acceptable but (later today) will be revising this month's scoring table in a previous post to reflect the system. The countdown also continues back a few more posts.

[WEATHER53]

With the extreme temperatures this month, and in certain cases earlier too, I have been thinking about the scoring rules because there seem to be a few inequities even after I expanded the scoring ranges to provide at least some semblance of scoring when anomalies got past 5.0 in either direction. So what I will do, including January but not going back into 2014, is to use a "minimum progression" formula to determine scores if the normal method does not produce higher scores than the following ... and this rule will apply to all score sets and can bump up any particular scores so in some cases the main portion of scoring will stay the same but a few zero scores may get a slight adjustment. Penalties will then apply to these minimum progression scores so with that factored in the final results won't necessarily be this neat:

 

High score must be at least 70.

 

Second highest score must be at least 60.

 

Third highest score must be at least 55.

 

Fourth highest score must be at least 50.

 

This reduction by 5 will continue until it falls below this formula: 2n where n is the number of players remaining to be scored. At that point scores will reduce by 2 per forecaster with only the worst of all forecasts getting zero points. With the number of players we have now, that jog takes place below 30. Where we have ties, points are equal but then drop by a multiple of 2 or 5 depending on how many are tied. Normal and consensus will get whatever score is an equal distance between two actual forecaster scores if not tied with one of them. Example, normal lies between scores of 22 and 20, so it gets 21 points. Consensus is tied with 35 so it gets 35. Normal and consensus don't count in the progression. However, if a large jump occurs around where normal or consensus appear, as with normal in the NYC forecasts this month, I will score it by the logic of the system regardless of midpoint. NYC provides examples of where the minimum progression falls below some raw scores (basically from -2.5 down to -3.2 score higher from raw score than from adjusted score -- even as NYC sinks to -11.0 -- remember that in the old system scores gradually lose points to huge anomalies by a percentage reduction in case you're wondering why your -2.7 forecast gets 25 and not 27 points. Larger raw score drops took place for BOS as the provisional there fell from -10 to -12.

 

This will probably have a rather minor effect overall on scoring order but it will make the monthly scoring a bit more equal which may make the contest more fun for people (and that is always the main objective here). Let me know if you like this, feel neutral or dislike it -- if there are objections I can revise or abandon this change. Will not go back into January until I feel that this revised scoring method is generally acceptable but (later today) will be revising this month's scoring table in a previous post to reflect the system. The countdown also continues back a few more posts.

If this is designed to smooth it out for those who have done poorly then that seems unreasonable especially since we are two months into it. This is a forecasting contest, not forecasting fun time.

[Roger Smith]

I think it may actually boost scores for the better forecasts more than the poorer ones, but what I was looking for was a method that would avoid situations where a large portion of the field score zero or nearly zero, yet there had been significant differences between forecasts that ended up with zero scores. It's basically just a retooling of the original concept of a wider error range for anomalous months, that was already being used. However, I will certainly be open to the full range of ideas about this, and I will mention this in the announcement for March so people can be alerted to the proposed change and read about it here.

[donsutherland1]

I think it may actually boost scores for the better forecasts more than the poorer ones, but what I was looking for was a method that would avoid situations where a large portion of the field score zero or nearly zero, yet there had been significant differences between forecasts that ended up with zero scores. It's basically just a retooling of the original concept of a wider error range for anomalous months, that was already being used. However, I will certainly be open to the full range of ideas about this, and I will mention this in the announcement for March so people can be alerted to the proposed change and read about it here.

Personally, I'd prefer to leave the rules as they are whether or not I would benefit from a change. The kind of extreme month that has led to such an outcome is rare.  This may be the first month such a situation has ever occurred during this contest. If the situation were fairly common, then I'd advocate a rules change, but that isn't the case.

[Roger Smith]

What about a much less intrusive change then, would say top score must be at least 50, second at least 40 and then count down to zero? This won't change things too much but it would continue to differentiate forecasts outside of scoring range. Right now we have a situation where somebody who is 7 degrees low on ATL gets the same zero as another person who is 10 degrees low. At least they would get a few points for their somewhat closer forecast. But I am easy on this, if the two of you who commented, and others, still don't like that small of an adjustment, I will gladly go back to the original scoring rules. I have to be candid here, my method would allow me to post guaranteed provisional scores in some cases, they just wouldn't have any way of changing if things got even worse or a bit better.

[Roger Smith]

Personally, I'd prefer to leave the rules as they are whether or not I would benefit from a change. The kind of extreme month that has led to such an outcome is rare.  This may be the first month such a situation has ever occurred during this contest. If the situation were fairly common, then I'd advocate a rules change, but that isn't the case.

 

 Last winter we had some doozies also, and March 2012 had some +10 to +12 anomalies in the east with scoring carnage. I can't remember if I was scoring then or Mallow was still doing the scoring. Either way, nobody scored much.

[Roger Smith]

 I should add, I totally get the concept that people don't like rules changes during play, but this is pretty early in the year and the differential in points would be rather trivial overall. It's more like an inflation of almost all scores so that these low-scoring months look a bit more like regular months. Anyway, if people accepted the modified version, I will firmly promise to lock these rules in and consider the whole system to be perfected. Same with late penalties, I had to modify those from time to time and I would prefer to have the same time penalties every month regardless of weekends, Super Bowls, Easter or any other distraction. This is why I relaxed the penalties thinking they could now fit every month.

 

But the philosophical question here is also, should we acknowledge rank order or an absolute error principle, allowing that we have already expanded the range of absolute error in the standard scoring method? I would say a blend of both. If X is a bit closer than Y, then X should get a few more points than Y. But it should be a fair fraction of what very close A and B score.

 

I hope a few others give me some feedback on this. Reaching a friendly consensus would be good here. It's sort of like golfing early in the year, and wondering what exactly is ground under repair. I was out on Saturday and we came to the conclusion that the entire golf course was ground (not) under repair.

[donsutherland1]

 Last winter we had some doozies also, and March 2012 had some +10 to +12 anomalies in the east with scoring carnage. I can't remember if I was scoring then or Mallow was still doing the scoring. Either way, nobody scored much.

If that's the case, then it deserves a closer look, IMO. Perhaps at the end of February, you can post the scores under the current rules and then what they might look like under the proposed rules change. Then, see what people think.

[Roger Smith]

Yes, I can do that ... as you may have noted, the provisional scores are posted (back a few posts now) and today I adjusted them to the smaller changes that have evolved from this discussion. In doing so, I found that it was becoming almost like smoothing out the old system for expanded scoring ranges, except that it allowed in a few points near the bottom end (mostly under 10 points) to differentiate the bad from the awful if you see what I mean (and believe me, I've been there, not so much recently). So it's hard to find an adjustment now that is more than five points in the higher scores (per station) and ten in the lower scoring range. Part of my concern was that I suspected one or two people dropped out because they kept getting blanked when they had not had the very worst forecast in the group. I'm not sure why we lost metalicwx366, they were doing well.

[WEATHER53]

If that's the case, then it deserves a closer look, IMO. Perhaps at the end of February, you can post the scores under the current rules and then what they might look like under the proposed rules change. Then, see what people think.

I agree Don. Lots of long winded statements about why a contest score should be changed after it has started. If the case could be made briefly that might speak for itself as to validity.

[Roger Smith]

I think I will abandon this proposed change and go back to my standard scoring, and carry several zero scores even though the forecasts are over a range. The more I tinker with this, the more it appears necessary to apply almost the same rules to prevent too much differential change in current scoring range, all in order to give a few people a handful of points instead of zero.

 

So, thanks for joining the discussion. The scoring as I type this still reflects the second proposal made above, but later today I will go back to the old system (edit, done) in the provisionals. To illustrate how little difference it was making, I only had to move two scores in the table when I changed to the new proposed scoring, but I was noticing a general upward shift of 30-40 points in the first case and then back down 10-20 for some in the second case. (edit -- so when I moved things back, once again it was mainly just a general drop of 30-40 points).

 

Bottom line is, the whole thing is merely tinkering with the concept of scores based on absolute error and it never really introduced much of an overall change to results anyway.

 

So meanwhile, how about these anomalies? These are now updated through Monday 23rd in the countdown, check them out.

[WEATHER53]

Thank you Roger for seeking input and thank you for running this contest.

[Roger Smith]

<<<<<< === ANNUAL SCORING FOR 2015, JAN-FEB === >>>>>>

 

 

... For best viewing, set to 100% to prevent line overflow ...

 

FORECASTER ___________DCA_NYC_BOS __Classic__ ORD_ATL_IAH__Expanded__TOTAL ___ Best scores

 

 

Tenman Johnson ________ 86 _133_126___ 345 ____ 99 _ 94_156___ 349 __ 694 _ 011 002 _ 0 _ 0 __ FEB

Mallow ________________133 _100_ 99 ___ 332 ____100_107_136 ___ 343 __ 675 _ 100 000 _ 0 _ 1 __ JAN

ksammut ______________113 _ 96 _102 ___ 311 ____ 50 _ 84_142 ___ 276 __ 587

Isotherm ______________101 _122_113 ___ 336 ____ 72 _ 92 _ 78 ___ 242 __ 578

RodneyS _______________ 98 _ 80 _ 80 ___ 258 ____ 48 _ 96 _120___ 264 __ 522 _ 000 010 __ 0 _ 0

Damage in Tolland ________99 _ 92_108 ___ 299 ____ 96 _ 56 _ 42 ___ 194 __ 493
Roger Smith ____________ 53 _111_133 ___ 297 ____ 50 _ 76 _ 68 ___ 194 __ 491 _ 000 100 _ 0 _ 0
wxdude64 ______________74 _ 78 _ 75 ___ 227 ____ 38 _ 76 _130 ___ 244 __ 471

..

Consensus _____________ 95 _ 58 _ 68 ___ 221 ____104 _ 71 _ 68 ___ 243 __ 464

..

metalicwx366 *__________ 70 _ 92 _ 96 ___ 258 ____ 66 _ 96 _ 42 ___ 204 __ 462 _ 001 010 _ 0 _ 1

BKViking _______________73 _ 62 _ 91 ___ 226 ____ 93 _ 78 _ 50 ___ 221 __ 447

Tom __________________ 88 _ 64 _ 80 ___ 232 ____ 82 _ 84 _ 44 ___ 210 __ 442
donsutherland.1 _________ 96 _ 70 _ 78 ___ 244 ____ 69 _ 71 _ 48 ___ 188 __ 432
Absolute Humidity *______ 96 _ 68 _ 58 ___ 222 ____ 94 _ 72 _ 30 ___ 196 __ 418
Quixotic1 ______________ 76 _ 49 _ 60 ___ 185 ____ 78 _ 50 _ 92 ___ 220 __ 405

..

Normal ________________ 92 _ 46 _ 42 ___ 180 ____ 70 _ 96 _ 58 ___ 224 __ 404 _ 000 010 __ 0 _ 0

..

Carvers Gap ____________ 63 _ 39 _ 38 ___ 140 ____104_ 60 _ 80 ___ 244 __ 384
blazess556 _____________ 86 _ 30 _ 30 ___ 146 ____110_74 _ 41 ___ 225 __ 371
hudsonvalley21 _________ 98 _ 43 _ 47 ____188 ____ 42 _ 71 _ 70 ___ 183 __ 371
wxallannj ______________ 84 _ 42 _ 48 ____174 ____ 97 _ 42 _ 50 ___ 189 __ 363
SD ____________________82 _ 26 _ 28 ___ 136 ____ 96 _ 76 _ 48 ___ 220 __ 356 _ 000 100 _ 0 _ 0
Midlo Snow Maker _______ 55 _ 26 _ 23 ____104 ____ 98 _ 61 _ 88 ___ 247 __ 351
SACRUS *______________ 78 _ 36 _ 40 ___ 154 ____ 88 _ 68 _ 06 ___ 162 __ 316
Stebo _________________ 38 _ 00 _ 00 ____ 38 ____104 _ 28 _ 84 ___ 216 __ 254
Rjay __________________ 50 _ 16 _ 22 ____ 88 ____ 54 _ 46 _ 32 ___ 132 __ 220
N. of Pike ______________56 _ 18 _ 28 ____102 ____ 58 _ 48 _ 06 ___ 112 __ 214
H2OTown__Wx *________ 56 _ 28 _ 20 ____104 ____ 06 _ 56 _ 04 ____ 66 __ 170
mikehobbyst* __________ 43 _ 46 _ 46 ____ 135 ____ 10 _ 04 _ 00 ____ 14 __ 149 _ 111 000 _ 1 _ 0

Maxim ________________ 23 _ 25 _ 24 _____ 72 ____ 00 _ 00 _ 00 ____ 00 ___ 72
hockeyinc *_____________ 00 _ 00 _ 00 ____ 00 ____ 30 _ 26 _ 00 ____ 56 ___ 56
Uncle W *______________ 00 _ 00 _ 00 ____ 00 _____00 _ 30 _ 10 ____ 40 ___ 40

 

 

Western contest 

 

FORECASTER __________ DEN_PHX_SEA__ TOTAL (all 9 in brackets, = ranks) __ best scores

 

 

Isotherm _______________140_145_155____ 440 ____ (1018) = 1

Midlo Snow Maker ________150_125_147 ____ 422 _____ (773) = 7 __________ 1 0 1 __ JAN

donsutherland.1 _________ 152_113_155____ 420 _____ (852) = 4

Roger Smith ____________ 168_ 83_155____ 406 _____ (897) = 3 ___________ 0 0 1

Damage in Tolland _______ 116_141_ 77 ____ 334 _____ (827) = 5 ___________ 0 1 0

Mallow _________________ 38_133_149 ____ 320 _____ (995) = 2

N. of Pike ______________ 147_ 95 _ 66 ____ 308 _____ (522) =19

wxdude64 ______________ 140_ 85 _ 29 ____ 254 _____ (725) = 8

..

Consensus ______________106 _77 _ 67 ____ 250 _____ (714) =10 ___________ 1 0 0

..

ksammut ________________96_ 47 _ 95 ____ 238 _____ (825) = 6

mikehobbyst*____________ 70_ 89 _ 75 ____ 234 _____ (383) =23 ____________ 0 1 0 __ FEB

hudsonvalley21 __________105_ 67 _ 55 ____ 227 _____ (598) =13

Wxallannj _______________ 92_ 65 _ 49 _____206 _____ (569) =16

..

Normal _________________110_ 56 _ 38 ____ 204 _____ (608) =11

..

RodneyS ________________ 66_ 81 _ 49 ____ 196 _____ (718) = 9

H2OTown_Wx*___________ 70_ 50 _ 76 ____ 196 _____ (366) =24

Maxim*_________________ 40_ 85 _ 67 ____ 192 _____ (264) =27

Quixotic.1 _______________ 52_ 72 _ 57 ____ 181 _____ (586) =15

Carvers Gap ______________94_ 31 _ 47 ____ 172 _____ (556) =17 ____________ 1 0 0

Tom ____________________56_ 78 _ 30 ____ 164 _____ (606) =11

BKViking ________________ 60_ 53 _ 39 ____ 152 _____ (599) =12

metalicwx366 *____________00_ 72 _ 56 ____ 128 _____ (590) =14

SD _____________________ 54_ 36 _ 28 ____ 118 _____ (474) =20

Absolute Humidity *________ 08_ 46 _ 54 ____ 108 _____ (526) =18

Rjay ____________________ 30_ 44 _ 20 _____ 94 _____ (314) =25

Uncle W *________________ 26_ 46 _ 18 _____ 90 _____ (130) =28

blazess556 _______________ 22_ 41 _ 26 _____ 89 _____ (460) =21

SACRUS* ________________ 04_ 24 _ 48 _____ 76 _____ (392) =22

Stebo ___________________ 12_ 17 _ 00 _____ 29 _____ (283) =26

hockeyinc* _______________ 00_ 16 _ 08 _____ 24 ______ (80) =29

 

(not entered in western)

 

Tenman Johnson __________________________________ (694) =10

 

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

 

* have entered one of two months

 

Posts are cross-checked against excel file for accuracy ... updated Mar 1 at 1245h EST

[Roger Smith]

No worries, you should enter the western portion and get a few more points since you are already in the top ten of the composite without even entering. Quite a few in the field have had a lot of trouble with DEN, which always seems to be near the dividing line between western warmth and east-central cold.

 

I have just finished posting revised total scores for the two months after finding one or two errors in my pre-excel work (just finished a cross-check, found a couple of fairly large errors one helpful to a score and one not so great, but mostly just round-off stuff from the percentage reductions) and will have a post up later today with provisional winter seasonal totals.

 

 

Thank you Roger for seeking input and thank you for running this contest.

 

[Roger Smith]

Four Seasons Award 2014-15 -- Winter (Dec, Jan, Feb)

___________________________________________________________________

 

Points are awarded as follows: 10 for the win, 7 for second, 6 for third, etc, but 1 point for each of 8th, 9th and 10th.

 

These final scores are listed in the order of the main contest then adding any others who placed in either the western or in the "all nine" categories (new this year). Scores are shown to give you an idea whether you might have any chance of moving up or down. Consensus is scored but does not replace an actual forecaster, and Normal, well thanks for playing and drive safely.

 

FORECASTER _______________ Main contest ____ Western contest ______ All nine

 

Mallow _____________________ 965 __ 10 _______ 526 __ 4 _________ 1491 __ 10

Isotherm ___________________ 772 ___ 7 _______ 626 __ 10 ________ 1398 ___ 7

Tenman Johnson _____________770 ___ 6

Midlo Snow Maker ____________ 749 ___ 5 _______ 616 ___ 7 ________ 1365 ___ 6

 

metalicwx366 ________________732 ___ 4 _______ 270 ___ -- ________1002 ___ --

Roger Smith _________________731 ___ 3 _______ 506 ___ 3 ________ 1237 ___ 4

Absolute Humidity ____________ 724 ___ 2 _______ 236 ___ -- _________960 ___ --

BKviking ____________________715 ___ 1 _______ 341 ___ -- ________1056 ___ 1

  

Consensus ___________________686 ___ 1 _______ 452 ___ 1 _______ 1138 ___ 2

 

Damage in Tolland ____________ 665 ___ 1 _______ 530 ___ 5 _______ 1195 ___ 3

Quixotic.1 ___________________ 659 ___ 1 _______ 399 ___ 1 _______ 1058 ___ 2

RodneyS ____________________ 658 ___ --_______ 354 ___ -- _______ 1012 ___ --

ksammut ____________________ 655 ___ -- _______366 ___ -- _______1021 ___ 1

Tom ________________________648 ___ -- _______316 ___ -- ________ 964 ___ --

wxallannj ____________________ 643 ___ -- _______388 ___ -- _______1031 ___ 1

Donsutherland.1 ______________ 632 ___ -- _______ 610 ___ 6 _______1242 ___ 5

SD _________________________ 604 ___ -- _______304 ___ -- ________908 ___ --

hudsonvalley21 _______________ 571 ___ -- _______429 ___ 1 _______ 1000 ___ --

blazess556 ___________________567 ___ -- _______285 ___ -- ________ 852 ___ --

SACRUS _____________________568 ___ -- _______214 ___ -- ________ 782 ___ --

N. Pike ______________________552 ___ -- _______468 ___ 2 ________1020 ___ --

H20Town_Wx ________________520 ___ -- _______ 334 ___ -- ________854 ___ --

wxdude64 ___________________471 ___ -- _______ 436 ___ 1 ________ 907 ___ --

Stebo _______________________479 ___ -- _______ 201 ___ -- _______ 680 ___ --

 

(all other scores are lower and do not score in this contest)

 

[Carvers Gap]

Roger, thanks for doing this!  It has been a blast.  Do I get extra points for picking the Super Bowl winner and score?  Just kidding.  Are you going to have a February(only) final ranking?  Thanks!

[Roger Smith]

Scroll back onto page one, the February scores are posted and will be updated but with the huge negative anomalies and the positives out west, I would not be expecting much change in them now. DEN scores could change order slightly. Everything else is pretty much nailed down in terms of order, the absolutes can still change a little.

[Roger Smith]

Notes:

 

1. The countdown to final anomalies can be seen in post 28.

2. The final February scoring can be seen in post 30.

3. The final annual scoring and winter four season awards can be seen a few posts back on this page (2) of the thread.

 

I have migrated the snowfall contest scoring summary and here it is updated to March 1st 11th. ATL continues to try to produce a measurable snowfall but so far just four trace days so the date is not yet established. IAH has also had its share of cold rains but no measurable snow. PHX had that one chance on Jan. 1st and missed out, nothing close since (Feb anomaly near +7 most of the month).

 

First (measurable) snowfall contest

 

Error point scoring system. For most stations, one error point per day. For ATL and IAH, one error point every two days, maximum of 20. Any "no snow" predictions at those two stations given 20 points if snow occurs, otherwise zero error points and all others get 20. For PHX, three forecasters predicted snow, their error dates divided by 5 and maximum of 5 error points for any of them if it snows, otherwise, they get 10 points and rest of field get zero.

 

THESE ARE NOW PROVISIONAL FINAL SCORES ASSUMING IT DOES NOT SNOW AT IAH or PHX --

if it does snow at either location, some of these scores will change.

 

 FORECASTER _________ DCA____NYC____BOS____ORD____ATL____IAH____DEN____SEA____PHX

  

bkviking _________65 _12-1720_11-28 2_11-28 1_11-10 5_12-2020_ no ** 0_11-08 3_11-25 4_12-1010

mallow __________90 _12-2215_12-0812_11-21 6_11-19 4_01-0720_ no ** 0 _11-15 4_12-2829

Damage in Tolland _96_12-0829_11-29 3_11-29 2_11-2712_01-0220_02-0320_11-17 6_12-03 4

cpick79 (N. of Pike) 96_12-30 7_12-05 9_11-0225_11-15 0_01-1520_ no ** 0_11-15 4_12-3031
SD _____________ 97_12-2413_12-0812_11-18 9_11-11 4_01-1320_ 01-2720_11-07 4_12-04 5 02-0910

Stebo ___________97_12-1621_11-20 6_11-20 7_11-19 4_01-1220_no ** 0_11-12 1_01-0638

RodneyS ________104_12-0334_12-1216_12-06 9_11-2914_01-2020_no **0_ 11-12 1_11-1910

Quixotic 1 ______ 104 _12-2017_12-1014_11-1413_11-15 0_01-0120_12-3120_11-15 4_12-1516

 

Consensus ______ 107_12-1621_12-05 9_11-21 6_11-19 4_01-0520_01-2120_11-12 1_12-2526_no *** 0

 

H2OTown_WX ___108_12-1621_12-05 9_11-24 3_11-22 7_01-1220_01-3120_11-20 9_ 12-1819

Roger Smith _____109_11-2839_11-25 1_11-1215_11-0510_12-0520_01-2520_11-11 0_12-03 4

Isotherm _______ 111_12-1423_12-03 7_12-03 6_11-23 8_01-1520 _no ** 0_11-18 7_01-0840

SACRUS ________115_12-2116_12-0711_11-1611_11-10 5_12-1720_12-2020_11-17 6_12-2526

donsutherland.1 __115_12-0730_12-0711_11-23 4_11-22 7_01-0520_no ** 0_ 11-12 1_01-1042

metalicwx366 ____116_12-0235_12-02 6_11-28 1_11-19 4_12-2420_01-2120_11-11 0_12-2930

wxallannj _______118_12-1720_12-0711_12-01 4_11-23 8_01-1420_01-2120_11-12 1_01-0434

blazess556 ______120_12-0928_12-0913_11-21 6_11-23 8_01-0620_no ** 0_11-13 2_01-1143

hudsonvalley21 __122_12-2611_12-1721_12-03 6_11-24 9_01-1820_01-2020_ 11-11 0_01-0335

wxdude64_______127_12-0829_12-01 5_11-27 0_11-16 1_12-3020_01-1920_11-12 1_01-1951

Midlo Snow Maker_135_11-2740_11-1511_11-1512_11-14 1_12-2220_01-1020_11-12 1_12-2930

Tom __________ 145_12-1720_12-1519_11-0225_11-16 1_01-0520_ no ** 0_11-17 6 _01-2254

Winters Grasp ___ 150_11-2047_11-0224_11-1413_11-14 1_12-0520_01-1020_11-10 1_12-2324

Absolute Humidity_151_12-0433_12-05 9_12-05 8_11-15 0_01-0520_ 01-3020_11-11 0_01-2961

OrangeburgWx ___157_12-0136_11-0125_11-0621_11-10 5_11-2720_12-2820_11-07 4_12-1516_02-2610

ksammut _______172_12-1819_12-1822_12-1922_12-0621_01-2120_01-1920_12-0423_12-2425

____________________________________________________________________________________

 

(only comparable to first six scores, tied 9th place in that portion) 

 

UncleW _________ 78 _12-1621_12-1014_11-30 3_11-15 0_01-0120_12-3120

 

 

Consensus values are medians, IAH favored for snow by 16/25 so a date was computed, PHX consensus no snow

 

Results will be calculated for both the "original six" and the full slate.

 

Entries checked Nov 3 ... no edits please

 

Error points thru end of contest assuming no snow IAH or PHX (ATL also but all have now gotten to 20 points with their predictions anyway) -- (table was updated Jan 6 when it finally snowed at DCA, Cpick79 now N of Pike had the least error points (7), still waiting for ATL, IAH and PHX which missed out when parts of Arizona got hit on New Years Day) ... IAH and ATL, error days divided by 2, and PHX where it's error days divided by 5 to a maximum of 5 if there is snow. Once you get to 5 at PHX (for the three who have a forecast there) you hold at 5 then if there is no snow you get 10 which is what the rest of us will get if it does snow there. We are now at a point where all players have 20 at ATL anyway, and all forecasts of snow at IAH now have 20 (or will have by March 15th in one case). Everyone predicted snow in ATL so if it doesn't snow there it won't make any difference to the contest scores.

 

LEADERS

 

1 BKviking 65

2 Mallow 90

3 Damage in Tolland and N of Pike 96

5 SD at 97

(see table above for rest of rankings) 

It's now official -- BKViking cannot be overtaken and has therefore won the contest.

 

(original six)

(1) BKviking 48, (2) Mallow 57, (3) stebo 58, (4) N. of Pike 61 (5) Isotherm 64 (6) Donsutherland.1 at 72, (7) wxdude64 and blazess556 are at 75, (9) SD and Uncle W 78, (11) H20Town_Wx 80.

 

(western three)

scores if no snow at PHX, Roger Smith 4, Damage in Tolland 10, RodneyS 11, BKViking 17, SD 19, Quixotic1 with 20.

[Roger Smith]

Scoring is now complete for February, and Tenman Johnson took the month (see post 30). Annual and four seasons award updates are posted just above the recently migrated snowfall contest. In the annual race, it's a narrow lead for Tenman Johnson over Mallow. In the west, it's Isotherm leading, after Mikehobbyst took February, his first entry. For the winter portion of the four seasons award, Mallow took the top points in both the main contest and the overall nine, while Isotherm was tops in the west.

 

The snowfall contest continues but we can declare a winner since BKViking cannot be overtaken now. ATL and IAH have yet to record any measurable snow despite a number of trace amounts. PHX had one shot at a snowfall on New Years Day but missed out. Whatever now happens, BKViking could not finish with more points than 85 and everyone else is over that score already. So congrats to BKViking, the race for second is still quite open.

[Roger Smith]

Extreme forecasts ruled again, eight of nine qualified, DEN was in there until the 20th as it went into free fall and ended near the middle of our range. So here are the updated standings with January merged ...

 

In January and February, these months and forecasts have qualified for an extreme forecast decision. 2nd extreme is only cited where it can win under these rules. No entry there means that an extreme forecast has won the month. This was shown last year as an "A" type win and the others were shown as "B" -- this year I am changing the format but not the actual rules.

 

Month ___________ Location ____ Anomaly _____ Extreme forecast ____ 2nd extreme ___ W-L decision

 

Jan 2015 __________ NYC _______ -2.7 _________ -3.5 Roger Smith ___ -2.7 Tenman ___ (L) RS _ (W) TJ

Jan 2015 __________ BOS _______ -2.9 _________ -3.3 Roger Smith ___ -2.7 TJ, met ___ (L) RS (W) TJ,m

Jan 2015 __________ IAH _______ -3.3 __________-5.9 Rodney S _____ -2.7 Tenman ___ (L) Rod (W) TJ

Jan 2015 __________ DEN _______ +3.2 _________+2.9 Midlo ________ (+2.5) ________ (W) Midlo

Jan 2015 __________ PHX _______ +2.2 _________ +2.8 Midlo ________ +2.0 Damage __ (L) Mid (W) Dam

Jan 2015 __________ SEA _______ +3.1 _________ +2.9 Midlo ________ (+2.4) ________(W) Midlo

Feb 2015 __________ DCA _______ -8.7 _________ -3.7 Mikehobbyst ____ (-3.5) ________(W) Mikehobbyst 

Feb 2015 __________ NYC _______-11.4 _________ -5.3 Mikehobbyst ____ (-4.3) ________(W) Mikehobbyst

Feb 2015 __________ BOS _______-12.7 _________ -5.9 Mikehobbyst ____ (-5.2) ________(W) Mikehobbyst

Feb 2015 __________ ORD _______-13.1 _________ -4.5 Roger Smith ____ (-3.0) ________(W) Roger

Feb 2015 __________ ATL _______ -6.8 __________-3.1 Tenman Johnson _(-2.0) ________(W) Tenman

Feb 2015 __________ IAH _______ -3.8 __________-2.2 Tenman Johnson _(-2.1) ________(W) Tenman

Feb 2015 __________ PHX _______ +5.9 _________ +4.9 Mikehobbyst ____(+4.7) ________(W) Mikehobbyst

Feb 2015 __________ SEA _______ +5.3 _________ +4.5 Roger Smith ____(+3.9) ________(W) Roger

 

__________________________________________________________________

 

Tenman Johnson _____ 5-0

Mikehobbyst _________ 4-0

Roger Smith _________ 2-2

Midlo Snow Maker ____ 2-1

Damage in Tolland ____ 1-0

metalicwx366 ________ 1-0

RodneyS ____________ 0-1