Activity is tracked in terms of https://en.wikipedia.org/wiki/Accumulated_cyclone_energy. This Wikipedia page has a list of previous North Atlantic hurricane seasons by ACE.
If the value is 10 or more ACE away from one of the cutoff values, this market resolves in January 2025 to whatever value is listed on Wikipedia. If it is close to a cutoff, the market resolves after the reanalysis of all storms is completed (usually in April) to account for postseason adjustments in ACE.
North Atlantic sea surface temperature anomalies, which are correlated with tropical cyclone activity, are currently (as of February 20) extremely high and can be tracked here: https://www.tropicaltidbits.com/analysis/ocean/cdas-sflux_ssta_atl_1.png
The last ten years had the following activity:
Below normal: 2014, 2015
Near normal: 2022
Above normal: 2016, 2018, 2019, 2021, 2023
Hyperactive: 2017, 2020
The least active season in recorded history was 1914, while the most active season was 1933.
See also my El Nino / La Nina market for hurricane season: https://manifold.markets/SaviorofPlant/will-it-be-el-nino-la-nina-or-neith. El Niño tends to suppress Atlantic tropical cyclone activity, while La Niña tends to enhance it.
Related questions
Colorado State University released a discussion of the recent inactivity: https://tropical.colostate.edu/Forecast/2024_0903_seasondiscussion.pdf
Quote that stuck out to me the most: "Only one other time in the satellite era has the Atlantic not produced any named storms between 13 August and 3 September. That other year was 1968."
There's no forecast update, but they still expect an above average season
@SaviorofPlant "So at this point, while it would be highly unlikely for CSU’s seasonal hurricane forecasts to exactly verify, there is still potential that a hyperactive season could be reached."
I interpret that for this market this as sounding like 'Above normal' as expected (slightly?) more so than hyperactive.
Figure 29 I found useful relative to Figure 15.. I wish they had a version of Figure 15 climatologically to see whether instability will return to at least 2023 levels by end of September....
The current wind shear and forecasts means we likely won't get a STS from that low in that NW Atlantic for the Francine market...
@parhizj Another quote I found valuable was this:
"One of the big issues plaguing the Atlantic has been the monsoon trough being so far
north that the easterly waves emerging from Africa are coming off at too far north of a
latitude to reach hurricane-conducive conditions. However, as the hurricane season
progresses, the climatological African Intertropical Front shifts back southward in
September (Figure 27).
Suggests we should still expect high activity in mid-September into early October, which is consistent with the 10-day model forecasts. Given we're already at 40 ACE, we don't need that much additional activity to get into the above-normal range.
I came across a forecaster that does project real time ACE that bettors might find useful (albeit slightly different buckets for this market)
https://weathertiger.substack.com/p/real-time-2024-atlantic-hurricane
The wide ranges for ACE seems very much reasonable at this point in time.
If anyone else has found something similar feel free to share if you wish.
@SaviorofPlant That weather tiger still has > 165 Ace at 62% .... I
From my notebook: Aanalogs by cutoff:
Top 10 weights (sorted by weight)
----------------------------------------
Rank | Year | Weight | Season ACE
----------------------------------------
1 | 1998 | 0.189 | 181.16
2 | 2023 | 0.186 | 146.79
3 | 2012 | 0.140 | 132.63
4 | 1979 | 0.067 | 92.92
5 | 2007 | 0.057 | 73.89
6 | 2010 | 0.053 | 165.48
7 | 2021 | 0.052 | 145.56
8 | 1999 | 0.051 | 176.53
9 | 1989 | 0.028 | 134.84
10 | 2003 | 0.024 | 176.28
I haven't put enough effort into my model as last year it was a bust.. This year the RMSE is very high early on... (+-50 ACE doesn't inspire any confidence...):
RMSE tuning: 34.14
RMSE validation: 53.24
Difference of RMSE(tuning) - RMSE(validation): 19.10
Last week:
Today (not much change, just a bit tighter CDF):
@parhizj After getting killed on the Francine market I have emotionally decided that the season is going to be much less active than expected. Time will tell if this is profitable.
I do want to note that this type of inactivity is extremely anomalous - even El Nino years usually manage to get a few storms in the first half of September. Like looking at your analogs, 1998 had two Cat 2s active at this point, 2023 had just had two majors and Lee was about to form, and 2012 had four hurricanes in late August into early September. If activity suddenly kicks up then a hyperactive season might still be possible, but I'm increasingly skeptical.
Not likely relevant, but @SaviorofPlant, are you doing (lower, upper] or [lower, upper)? It's not clear from the multi.
17.1 as of this morning, vs an average of 2.5 at this point in the season: