Jump to content

  • Member Statistics

    17,606
    Total Members
    7,904
    Most Online
    ArlyDude
    Newest Member
    ArlyDude
    Joined

2017 North Atlantic Tropical Storm / Hurricane Forecast Contest

Roger Smith

By Roger Smith,
in Tropical Headquarters

Recommended Posts

Roger Smith

Roger Smith

Posted
Posted

To encourage a few more entries, and given that there seems to be little activity expected this week, the deadline will be extended to Tuesday June 6th at 23z. You could still enter after that but I will start assessing penalties at the following rates: seasonal will lose one point each day entered after June 6th and June monthly will move to the standard "portion of month elapsed" deduction which by 7th will be 20%. (this penalty will be waived for entries made before June 6th 23z -- note to any entrants already registered, you can edit your predictions up to that extended deadline, table of entries to be posted on June 7th from entries then placed in the thread.

 

After reviewing how the last two contests went, I will use the same format as last year.

 

Here's the simplified version of how to enter:

 

1. Enter your seasonal forecast (named storms / hurricanes / major hurricanes) in this thread before 06z June 1st, with 1% per day reductions in score for late entries. This portion of the contest is worth 50% of the total score but will also be tabulated in order of finish. Please note that in April, Tropical Storm Arlene made the count for 2017 1/0/0 before you even enter, so factor that into your prediction. Any storms that are named in May will also count towards the seasonal total but not the monthly portion of the contest. If you enter before a named storm is declared in May, either edit your post, or allow me to add the results of the named storm to your prediction (most people ensure that their seasonal forecast is the sum of their monthly forecasts).

2. If you want to enter a specific June forecast and other months later in the year, add those to your seasonal. Last year, I think all but one contestant entered all of their months in advance and one or two adjusted during the contest. One never entered any monthlies and was assigned a pro-rated version of the contest normal. NOV-DEC will count as one forecast, so you can enter monthlies for JUN, JUL, AUG, SEP, OCT and NOV-DEC. If you do not enter monthlies, you will be assigned provisional but fully-scoring estimates based on your seasonal weighted from this "contest normal" which is adjusted from last year which already had Alex and Bonnie, as Bonnie 2016 = Arlene 2017, the contest normal is set at last year's 16/8/3 minus 1/1/0 for Alex. The seasonal contest normal is set at 15/7/3. This is adjusted from the 1989-2016 average of 14/7/3. It will be adjusted again if necessary by whatever may happen later in May.

 

_________________________ SEASON __ APR _ MAY _ JUN _ JUL _ AUG_ SEP _ OCT_ N-D

 

Adjusted Normal 1989-2016____ 15_7_3 __ 100 _ 000__100__100__421__531__321__000

(note that N-D has an actual average of something like 0.4/0.1/0.01 so it rounds down to 000)

Your assigned monthlies will be adjusted from these as required, and you will see them in the table of forecasts -- if your June forecast does not match the provisional then the other months will not be adjusted, so your monthlies might not add up. Some of you will probably want to enter each month like last year, and you only need to do that if you don't want to use the already posted provisional.

3. Then if you want to update your monthly forecasts to fit your own estimate, the table will change to reflect your forecasts, but future provisionals will continue to appear unchanged and your ongoing total may continue to be different from your seasonal as a result. There is no requirement for monthly entrants to match their seasonals and monthlies. There will not be any scoring adjustments available for seasonal updates, you can make them, but the contest will not award any different scores for them.

4. Seasonal forecasts are scored from a maximum of 50 points, reduced by the average of one point per error in each category plus that number squared. Example, you predict 11/6/2 and reality is 14/7/4. You then lose (3+9)/2 or 6 points for your named storms, (1+1)/2 or one point for your hurricanes and (2+4)/2 or 3 points for your major hurricanes. Your score is then 50-6-1-3 or 40.

5. Monthly forecasts are scored the same way as above from totals available of 4 (June), 6 (July), 12 (August), 16 (September), 10 (October) and 2 (Nov-Dec) except that June, July and Nov-Dec error reductions are divided by two (e.g., if the normal method would give you 3.0 out of 6.0 for July, the adjusted method gives you 4.5. Monthly forecasts that you submit are subject to 1% reductions for every 3h late (the deadline is 03z of 1st each month except for June 06z). These late penalties will be doubled as soon as a named storm exists in the month in question. The June deadline, however, may be extended slightly to encourage more entries unless named storms are imminent. Expect an absolute cutoff by end of June 3rd with penalties after that.

The best entry method is to copy this template and add in your numbers (and name)

User Name ___ Seasonal (xx/xx/x) __ APR 1/0/0 _ MAY 0/0/0 _ JUN x/x/x _ JUL x/x/x _ AUG x/x/x _ SEP x/x/x _ OCT x/x/x _ NOV-DEC x/x/x

 

FAQ _ Should I include any May named storm in my seasonal forecast? __ Yes, you are basically forecasting the normal season plus the count for May (which is currently 0/0/0) and you should also include Arlene (1/0/0) from April 2017.

What if there's a named storm in May after I enter? __ If you don't edit in that change, I will edit it into your forecast by adding 1/0/0 to your seasonal forecast; if you want to stay at the original prediction despite the May storm and your monthly forecasts as submitted, edit in that request. Nothing gets tabulated until June 4th.

What if I (or the NWS) forecast a range? __ The contest will take the median of your various ranges and score those only. If you want to give decimals, we'll score from those.

What if I report on a forecast from elsewhere? __ Unless you are involved in making that forecast, then we'll score it but it won't count in the standings. If you are involved in it and want to enter it, make it clear that it's your personal entry.

What month does each named storm belong to? __ In this contest, each named storm belongs to whatever month it was first named in (according to the NWS conventions as to time zones etc). Then whatever that storm goes on to do will belong to that same month. So for example, a tropical storm named on Aug 31 will count in the August forecast for all of its exploits as a hurricane or major hurricane in September. The timing of any given storm receiving a number but not a name is irrelevant to the contest. As with last year, where there was one storm that had its hurricane phase in a different month than its tropical storm phase, an alternate scoring table was posted and the same entry won either way.

_______________________________________________________________

Good luck. Defending champ is Ohiowx. In 2015 it was Troy1234 with Stebo the winner in 2014.

 CSU forecast is 11/4/2. I will assign some monthly values to that one scaled down from contest normals. The NHC forecast will be entered as well, when I get the details. 

 

Link to comment
Share on other sites
kalasea

kalasea

Posted
Posted

My entry ...

Name: Kalasea, ___ Seasonal 2017  (18/13/6) __ APR 1/0/0 _ MAY 1/0/0 _ JUN x/x/x _ JUL 4/2/1 _ AUG 2/2/1 _ SEP 3/2/1 _ OCT 4/4/3 _ NOV-DEC 3/3/0

(named storms / hurricanes / major hurricanes)
 

Link to comment
Share on other sites
bigtenfan

bigtenfan

Posted
Posted
On 5/8/2017 at 8:21 PM, kalasea said:

My entry ...

Name: Kalasea,  ___ Seasonal 2017  (18/13/6) __ APR 1/0/0 _ MAY 1/0/0 _ JUN x/x/x _ JUL 4/2/1 _ AUG 2/2/1 _ SEP 3/2/1 _ OCT 4/4/3 _ NOV-DEC 3/3/0

(named storms / hurricanes / major hurricanes)
 

That is a pretty aggressive forecast. 3 majors in October!

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

kalasea

Posted
Posted
On 5/9/2017 at 6:39 PM, bigtenfan said:

That is a pretty aggressive forecast. 3 majors in October!

The end of Sept is calling for intense planetary weather thus my prediction for intense earth weather to follow. Galaxy and solar weather have affects and when the sun is weak the magnetics of earth are changed and let in more solar radiation there are many other factors as well. 

Link to comment
Share on other sites
Kurzov

Kurzov

Posted
Posted

Name: J. B. Mills - Seasonal (xx/xx/x) __ APR 1/0/0 _ MAY 0/0/0 _ JUN 1/1/0 _ JUL 1/2/0 _ AUG 3/2/1 _ SEP 3/0/0 _ OCT 2/1/1 _ NOV-DEC 1/0/0

Forecast:
Named Storms: 22
Hurricanes: 8 of 22
Major Hurricanes: 2 of 22

I'm a newb to this, so let's hope this works... it probably won't, if it'll be the way that my predictions usually go... :P

Link to comment
Share on other sites
Roger Smith

Roger Smith

Posted
  • Author
Posted

I just edited the instructions so that instead of NAME it now says USERNAME which is what I meant there, you folks who revealed your actual names should edit them out (not a requirement just a suggestion). Sorry about the confusion on that. 

My forecast will be

Roger Smith ___ 19 13 6 ___ 100 _ 000 _ 110 _ 210 _ 431 _ 543 _ 432 _ 210 _

Link to comment
Share on other sites
kalasea

kalasea

Posted
Posted
6 hours ago, Roger Smith said:

I just edited the instructions so that instead of NAME it now says USERNAME which is what I meant there, you folks who revealed your actual names should edit them out (not a requirement just a suggestion). Sorry about the confusion on that. 

My forecast will be

Roger Smith ___ 19 13 6 ___ 100 _ 000 _ 110 _ 210 _ 431 _ 543 _ 432 _ 210 _

Would you let me know how to edit when someone has quoted me and used the full name...  I did edit my original post Thanks for your help!

Link to comment
Share on other sites
Roger Smith

Roger Smith

Posted
  • Author
Posted

NWS mid-range of forecast will be 14/7/3, fairly similar to our contest normal although they use a 30-year normal and say this would be above normal. 

I will leave the contest open for entries until there is some action so feel free to edit your posts, I am not making a table until at least Sunday June 4th. 

Link to comment
Share on other sites
Roger Smith

Roger Smith

Posted
  • Author
Posted

Table of entries -- North Atlantic tropical storm and hurricane seasonal contest

Note: Contest Normal is 14 8 3 based on 1989 to 2016 data but is adjusted to incorporate pre-season Arlene in April, this maintains the monthly values of Contest Normal. Consensus is the mean of forecasts from American Weather Forum participants and does not include outside agencies listed, or Normal. Any forecasts received without complete data are assigned representative values for scoring purposes. These are marked with asterisks. Contest rules allow you to submit revised monthly forecasts before any month begins without penalty. You are not stuck with these monthly forecasts but you are stuck with the seasonal forecast (as of June 7th table creation). Later forecasts are accepted but we have started the late penalty countdown of one point per day, which means no point in entering past July 27th when seasonal points (50) would be reduced to zero.

Note: sent Kurzov a message as their seasonal and monthlies are out of sync, in any case they have to end of June to provide updated monthlies from July to Nov-Dec. However, nobody is required to have monthlies that add up to a seasonal so these will play if not adjusted. I did edit July which reads 1/2/0 in the entry, to 2/2/0. Update _ have verified the completed forecast for pcbjr by pm. 

 

FORECASTER _______ Season  __ APR __JUN __JUL __ AUG __ SEP __ OCT__NOV-DEC 

 

Kurzov _____________22 8 2 ____1 0 0 _ 1 1 0 _ 2 2 0 _ 3 2 1 _ 3 0 0 _ 2 1 1 _ 1 0 0

Windspeed _________21 11 5 ___ 1 0 0 _ 2 1 0 _ 3 2 0 _ 4 2 0 _ 6 4 3 _ 4 2 2 _ 1 0 0
 

Roger Smith ________19 13 6 ___ 1 0 0 _ 1 1 0 _ 2 1 0 _ 4 3 1 _ 5 4 3 _ 4 3 2 _ 2 1 0

Kalasea ____________18 13 6 ___ 1 0 0 _ 1 0 0 _ 4 2 1 _ 2 2 1 _ 3 2 1 _ 4 4 3 _ 3 3 0

rockchalk83 ________ 18 10 5 ___ 1 0 0 _ 0 0 0 _ 2 1 1 _ 4 3 1 _ 6 4 2 _ 4 2 1 _ 1 0 0

dmillz25 ____________17 7 3 ____1 0 0 _ 2 0 0 _ 3 1 0 _ 4 2 1 _ 6 4 2 _ 2 0 0 _ 0 0 0

Stebo ______________16 9 4 ____1 0 0 _ 2 0 0 _ 3 2 1 _ 4 3 1 _ 4 2 1 _ 2 1 1 _ 1 1 0

NJwx85 ____________ 16 9 3 ___ 1 0 0 _  1 0 0 _ 2 1 0 _ 3 2 1 _ 5 4 2 _ 3 2 0 _ 1 0 0

NCforecaster89 ______ 16 8 3 ___ 1 0 0 _ 1 0 0 _ 2 1 0 _ 4 2 1 _ 5 4 1 _ 2 1 1 _ 1 0 0

 

Consensus* _________ 16 8 3 ___ 1 0 0 _ 1 0 0 _ 2 1 0 _ 3 2 1 _ 4 3 2 _ 3 2 1 _ 1 0 0

 

Ser Pounce __________15 9 3 ___ 1 0 0 _ 1 0 0 _ 2 1 0 _ 4 3 1 _ 5 3 2 _ 2 2 0 _ 0 0 0

 

Contest Normal ______ 15 8 3 ___ 1 0 0 _ 1 0 0 _ 1 0 0 _ 4 2 1 _ 5 3 1 _ 3 2 1 _ 0 0 0

 

RJay _______________ 15 8 3 ___ 1 0 0 _ 1 0 0 _ 2 1 0 _ 4 3 1 _ 4 3 2 _ 2 1 0 _ 1 0 0

NWLinnCountyIA _____ 15 6 3 ___ 1 0 0 _ 1 0 0 _ 1 0 0 _ 4 2 1 _ 4 2 1 _ 3 2 1 _ 1 0 0 

Matt Petrulli _________ 15 6 2 ___ 1 0 0 _ 1 0 0 _ 2 1 0 _ 4 2 1 _ 5 2 1 _ 2 1 0 _ 0 0 0

..... NHC (mid-range) __14 7 3 ___ 1 0 0 _ 1 0 0 _ 1 0 0 _ 3 2 1 _ 5 3 1 _ 3 2 1 _ 0 0 0

Yoda _______________13 8 3 ___ 1 0 0 _ 0 0 0 _ 2 1 0 _ 3 2 0 _ 4 3 2 _ 3 2 1 _ 0 0 0

ldub23 _____________ 13 8 3 ___ 1 0 0 _ 1 0 0 _ 1 1 0 _ 3 2 1 _ 5 3 2 _ 2 2 0 _ 0 0 0

radarman ___________ 12 5 2 ___ 1 0 0 _ 1 0 0 _ 2 1 1 _ 3 1 0 _ 3 2 1 _ 2 1 0 _ 0 0 0

pcbjr _______________ 11 5 2 ___ 1 0 0 _ 0 0 0 _ 0 0 0 _ 2 1 0 _ 3 2 1 _ 4 2 1 _ 1 0 0

 

..... CSU ____________ 11 4 2 ___ 1 0 0 _ 0 0 0 _ 1 0 0 _ 3 1 1 _ 4 2 1 _ 2 1 0 _ 0 0 0

________________________________________________________________

* Consensus is derived from means for 17 forecasts not incl NHC, CSU, Normal, listed below to

 one decimal (rounded in table above)

 

Mean _________ 16.0,8.4,3.4 _ 100 _ 1.0,0.2,0 _ 2.1,1.1,0.2 _ 3.4,2.2,0.8 _ 4.4,2.8,1.6 _ 2.8,1.6,0.8 _ 0.8,0.3,0


 

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

Roger Smith

Posted
  • Author
Posted

Now appears July will finish 2 0 0 thanks to the surprise appearance of Emily near Tampa Bay on the 31st. Under contest rules any further intensification of Emily in August will count for the July monthly values but so far the advisories hold it at TS strength when it emerges into the Atlantic. If that proves to be the case, then scoring for the various entries in the contest would be revised to the following:

2 0 0 __ 6

1 0 0 __ 5.5

2 1 0 __ 5.5

2 1 1 __ 5.0

2 2 0 __ 4.5

0 0 0 __ 4.5

3 0 0 __ 5.5

3 1 0 __ 5.0

3 1 1 __ 4.5

3 2 0 __ 4.0

3 2 1 __ 3.5

4 2 1 __ 2.5

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

Roger Smith

Posted
  • Author
Posted

(posted while Harvey still at TS intensity) August currently sits at 3/2/0 and if Harvey attains hurricane status before landfall it will be 3/3/0 (NHC now upgrading to possible major, so could reach 3/3/1) with the season soon to be 8/3/0 and possibly 8/3/1 by Saturday. 

This month, Franklin was briefly a 'cane, Gert almost made it to cat 3 but not quite, and Harvey is being forecast to reach cat-3 intensity before landfall.

(update) August now sits at 3/3/1 and season 8/3/1. There is some chance of further action (Irma) before the end of August. Under contest rules, if Irma forms in August, its later stages will count for August even if they only develop in September, but an alternate score will be provided (this happened last year between August 31 and September 1). (update 30th) _ Irma formed although not from the expected "potential TD 10" but in the eastern Atlantic. The same comments apply and appear to be likely to verify, so August will likely finish either 4/3/1 or 4/4/1 and possibly 4/4/2 depending on what Irma does in September. And as already mentioned, alternate scoring tables will be posted for the count within months.

For years with ten or more named storms, the lowest ratio of hurricanes to named storms was .143 (2 of 14) in 2013, followed by .231 (3 of 13) in 1931. For years with fewer than 10, the lowest ratios are zero (0 of 5) in 1907 and (0 of 1) in 1914, and .200 (1 of 5) in 1905.

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...