wxmx Posted October 28, 2014 Share Posted October 28, 2014 1/0/0 Link to comment Share on other sites More sharing options...
ncforecaster89 Posted October 28, 2014 Share Posted October 28, 2014 1/0/0 Link to comment Share on other sites More sharing options...
GaWx Posted November 1, 2014 Share Posted November 1, 2014 Nov-Dec: 0/0/0 Link to comment Share on other sites More sharing options...
hurricaneman Posted November 1, 2014 Share Posted November 1, 2014 November-December 1-0-0 Link to comment Share on other sites More sharing options...
Rjay Posted November 1, 2014 Share Posted November 1, 2014 November-December 0/0/0 Link to comment Share on other sites More sharing options...
wxmeddler Posted November 1, 2014 Share Posted November 1, 2014 Nov/Dec0/0/0 Link to comment Share on other sites More sharing options...
Roger Smith Posted November 3, 2014 Author Share Posted November 3, 2014 The table back in post 174 has been updated -- scoring is now based on the seasonals and NOV-DEC 2-pt conversions (O-maha) fitting the current 8/6/2 seasonal count with a 0/0/0 finish. If that changes, the table will change along with the storm evolution(s). Otherwise, it would appear that Stebo has the lead in the contest. I scored Isohume's unsubmitted monthlies full-value since I wanted to provide the best possible comparison for the FSU forecast, so that one has done quite well, however, in terms of unsubmitted monthlies added to seasonals, Pluffmud has the lead. Isotherm, Gawx, wxmeddler and ncforecaster89 all have some chance of overtaking Stebo depending on how things pan out, and I would come out on top if there are 2 more storms and 1 is a hurricane. At least that's what my excel program says when I change the numbers and look at the scoring tables (ranks and everything, this is almost like a real contest now). When all is said and done, I will post the results in more readable formats, the table at present has all the junk left in the trunk, but it's relatively easy to compare seasonal forecasts, the current numbers (out of 50) are in blue and are the first numbers you'll see going across the page from the names. The overall contest totals are the last numbers in each row. Everything else relates to monthly scores and the forecasts that generated them. The overall contest totals for those who did not enter all months are shown like this -- as an example, 63.0 (84.0) and what that means is that so and so has a total of 62 points from their seasonal (which might have been 42) and half the value of the unsubmitted provisional monthlies that I calculated from a percentage of the normal values (which are in the first entry). Now those people might have actually done better or worse if they had tried each month -- some tried a few, some never entered the monthly section at all. But those half-values were given out mostly to encourage people to get into the monthly section back in the summer when the points missed were not that numerous. For the entries without any actual monthly forecasts, I am not sure if the totals mean much at all. Meanwhile, the number in brackets would compare those scores, if they had been earned by actual forecasts, with those scores that are based on actual forecasts. This tells you how your seasonal forecast might have set you up in general terms, but maybe the more relevant comparison is how your seasonal forecast did compared to others. Generally speaking, unless we get a surge of late activity, the best seasonals were around the bottom third of the table but very few people anticipated the high percentage of storms that would become hurricanes (75% at this point), the average was probably closer to 50% or lower. If we get a surge of activity (meaning perhaps 2 storms, 1 hurricane) then the best seasonals will move up closer to consensus or higher, the trade-off being that the high end would have better hurricane prediction and the middle would have better storm counts. Not sure if there will be another post for a while, unless the dormant hurricane machine turns on again. But there will certainly be an end-of-contest report around the Christmas-New Year period. Thanks for playing, hope you enjoyed the ride. Link to comment Share on other sites More sharing options...
Roger Smith Posted January 5, 2015 Author Share Posted January 5, 2015 The scoring in post 174 is now confirmed as the final contest scoring. Congrats to Stebo who won with a total score of 82.0. That score would have been matched by the FSU numbers submitted by Isohume using the assumption that the seasonal forecast would generate equally weighted monthly forecasts. However, those were not submitted. Meanwhile, Ga_wx was a very close second at 81.5. Isotherm was third at 81.0. Icebreaker5221 also beat consensus at 79.8 points. Both Isotherm and Icebreaker missed one month and their scores include the half-value provisionals that are given to all who miss any monthly forecast. Their actual "earned" scores were 77.5 and 78.0. Consensus was 79.5. See the table in post 174 for the full list of scores. The best seasonal forecast by the scoring system was a tie (46 out of 50 possible points) for Eyewall 2005 (10,5,2) and Pluffmud (8,4,1). There is a list of high total monthly scores in the table. In terms of actual forecasts submitted, the highest monthly portion of the contest was also Stebo at 40 out of 50. Roger Smith and Ga_wx tied for second at 38.5. The list contains some other scores that rank between Stebo and these two, but they include one provisional value or a set of assumed values. Finally, a list could be drawn up for potential total scores given full rather than partial provisional monthly scores. You can look that up in the table, but the highest possible score would have been 87 for Pluffmud and 86 for Eyewall 2005. Thanks for playing, hope you enjoyed the contest. I hope that if we have another one, everyone who makes a seasonal forecast will supply a monthly. Well done to those who supplied a list in advance, your names appear in italics in the table. Link to comment Share on other sites More sharing options...
GaWx Posted January 6, 2015 Share Posted January 6, 2015 Congrats to Stebo! Thanks, Roger for running this. This was a lot of fun! Link to comment Share on other sites More sharing options...
Stebo Posted January 6, 2015 Share Posted January 6, 2015 Congrats to Stebo! Thanks, Roger for running this. This was a lot of fun! Thanks! I agree this is a lot of fun, maybe next season we can have something nice like 15/8/4 just perk things up a bit. Link to comment Share on other sites More sharing options...
Roger Smith Posted January 6, 2015 Author Share Posted January 6, 2015 I should mention that I added the NHC median forecast but they actually predicted a range and the reality fell just within that range, I believe. The results also show that the FSU forecast (if I read the post correctly) is as good as any indicator for this past season,. My seasonal forecast, on the other hand, would have been good in some other unspecified year. I needed the five nuisance name-waster tropical storms in mid-ocean that have bulked up other recent seasons. Link to comment Share on other sites More sharing options...
Rjay Posted January 8, 2015 Share Posted January 8, 2015 Thanks! I agree this is a lot of fun, maybe next season we can have something nice like 15/8/4 just perk things up a bit. Too bad you couldn't win fantasy football . Link to comment Share on other sites More sharing options...
Rjay Posted May 7, 2015 Share Posted May 7, 2015 It's time. Link to comment Share on other sites More sharing options...
Roger Smith Posted May 8, 2015 Author Share Posted May 8, 2015 Yes, I mentioned in the ongoing 2015 thread that a contest would be announced soon. If this event off the SC coast gets a name then we will just have to swallow that as background noise because the contest will be open for entries until June 1st (and beyond if you want to take a late penalty). Monthly forecasts will start with June. All this might mean is that the seasonal forecast can't be worse than 1/0/0. We already have 1914 tied if this gets a name. Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.