2021 North Atlantic hurricane forecast contest

[ncforecaster89]

Hi Roger!  Thanks for all the work you put into this effort. Just wanted to note that all of us below the 20/9/4 numbers should be listed a position higher (e.g. after BKViking, since they have the same forecast as WxWatcher007).  By counting both, as equal, we get unfairly knocked down a place…due to our efforts to provide a distinct forecast from everyone else.  
 

Edit: This post isn’t meant to disparage BKVing and it’s not a big deal, necessarily.  I’m just big on fairness in general.    

[Roger Smith]

Well the convention in other contests that I enter (and one that has been going for many years and that is not scored by me in one case) is to treat all similar forecasts as separate entities and in all probability BKViking (who is a regular in the temperature forecast contest) never noticed that he was making the same forecast as wxwatcher007, or else did notice and agreed that was the call to make. The convention in those other contests is to rank the identical forecasts from earliest to latest entry which I did not do here (yet). 

I see your point but just to let you know that this scoring convention is fairly well established in weather contests and not something I introduced here for the first time. 

(later edit _ I went and had a look, realized that in fact it would make no difference to anything if I went with this suggestion, so I have edited in the changed ranks and also removed some gaps in the table that served no actual purpose. ... so now it is ranked the way NC89 suggests.)

Also by the way there was no actual reason for that gap in the scoring table, it wasn't on my screen but what I see and what gets posted sometimes not the same thing. I can probably just edit the table to show the ranks the way you suggest with an asterisk noting that two forecasts are at the same rank. But be prepared if you enter other contests in other places, that's how they normally handle identical forecasts. I suppose the cases I had in mind would deal with duplicate values on a regular basis, for example, one temperature value and fifty to a hundred entrants, you will always have clusters of multiple entrants at one value. And in my experience you finish ranked behind all of them if their errors are smaller. But what can you do when there are more entrants than plausible entry points? With this contest the complexity is greater so that 30 people could all choose unique forecasts. On the other hand, I don't think there is anything basically wrong with placing an identical forecast, if this is what you think is going to happen, it's your best effort and the fact that somebody else thought the same thing might not be any contributing factor. Some people in past years have repeated the NHC or CSU numbers, would we penalize that or simply view it as prudent forecasting, after all, whose numbers should be best, me at my desk here in the middle of nowhere, or some established expert(s) being paid millions to do this sort of work? (so why do some of us, well let's not go there) ....

 

 

[Roger Smith]

If there are no further tropical storms this month, the contest will end on December 31st and the winner at this point would be wxdude64 (98.5 points) -- the prediction of 19/7/4 hit both the hurricanes and majors right on. Those who had the right number of named storms (21) included second place JKeithLee, Newman and hudsonvalley21 while the only other person to guess seven hurricanes was Iceresistance. Nobody in the contest predicted fewer hurricanes than what we ended up getting in 2021. About half the field selected four majors.

See the table posted Sep 24 (and edited since then) for the full standings. 

If there should happen to be one more tropical storm and no further hurricanes, then wxdude64 would share the win with IntenseWind002. 

Will post again on January 1st or 2nd (sobriety having been restored) with a confirmation of this post or any necessary changes. Thanks for playing. 

[Roger Smith]

The contest has officially concluded at 21/7/4 and the winner is wxdude64, congrats. 

The full scoring summary can be seen back a few posts, as nothing has changed since then.