mitchnick Posted October 13 Share Posted October 13 1 hour ago, snowman19 said: Signals growing for warmer to much warmer than normal November in the east…possibly much drier than normal too, which would just really be a continuation of the anamalous dry pattern we’ve seen since late August: Warm November has been advertised on long range models with a warm December too. We hashed this out when the Euro seasonal came out that things start cooling in January with February continuing the trend, which is why the tri-monthly average after Dec-Feb cools. Iow, nothing new or to see here. 2 Link to comment Share on other sites More sharing options...
snowman19 Posted October 13 Share Posted October 13 30 minutes ago, mitchnick said: Warm November has been advertised on long range models with a warm December too. We hashed this out when the Euro seasonal came out that things start cooling in January with February continuing the trend, which is why the tri-monthly average after Dec-Feb cools. Iow, nothing new or to see here. As of right now, I am very skeptical of a cold/snowy February. That would be very atypical of a canonical La Niña. 20-21 was a recent exception, but that was also caused by strong -AO/-NAO blocking, due in very large part to the solar minimum, IMO. At least for NYC, if December is warm, pray for a greater than 3” of snow total for the month. As @bluewave showed, there has been an extremely strong correlation over the last 30 years between total December snowfall and the snowfall for the rest of the season when you have a La Niña in place 3 Link to comment Share on other sites More sharing options...
mitchnick Posted October 13 Share Posted October 13 51 minutes ago, snowman19 said: As of right now, I am very skeptical of a cold/snowy February. That would be very atypical of a canonical La Niña. 20-21 was a recent exception, but that was also caused by strong -AO/-NAO blocking, due in very large part to the solar minimum, IMO. At least for NYC, if December is warm, pray for a greater than 3” of snow total for the month. As @bluewave showed, there has been an extremely strong correlation over the last 30 years between total December snowfall and the snowfall for the rest of the season when you have a La Niña in place You can be skeptical, it's your perogative. But if people are going to post model runs, it seems only fair to consider the entire run and not just the parts they agree with. 2 2 Link to comment Share on other sites More sharing options...
SnoSki14 Posted October 13 Share Posted October 13 1 hour ago, snowman19 said: As of right now, I am very skeptical of a cold/snowy February. That would be very atypical of a canonical La Niña. 20-21 was a recent exception, but that was also caused by strong -AO/-NAO blocking, due in very large part to the solar minimum, IMO. At least for NYC, if December is warm, pray for a greater than 3” of snow total for the month. As @bluewave showed, there has been an extremely strong correlation over the last 30 years between total December snowfall and the snowfall for the rest of the season when you have a La Niña in place No way! You skeptical? That's crazy 2 1 Link to comment Share on other sites More sharing options...
GaWx Posted October 13 Share Posted October 13 5 hours ago, snowman19 said: Signals growing for warmer to much warmer than normal November in the east…possibly much drier than normal too, which would just really be a continuation of the anamalous dry pattern we’ve seen since late August: *Fwiw* JB keeps hinting at a cold Nov/Thanksgiving/Dec for the E US. I’m not putting much weight on that for obvious reasons. Also, on Oct 8th he ditched his +2 for Megapolis for Nov-Mar for a NN (+1). And that’s obviously subject to further cooling in later updates. Keep this in mind: Since at the very least 2014-5 (because they’re all circulating on the internet back to then), JB has yet to have a mild FINAL Megapolis winter forecast. With the number of mild winters (5 of the 10) there have been, that’s not a good record. The ones 3+ AN at NYC since 2014-5 for those who don’t know were 2015-6, 2016-7, 2019-20, 2022-3, and 2023-4. A. The 5 mild NYC winters (vs respective 30 yr mean): -JB had -1 for Nov-Mar 2023-4 vs actual of +4 (DJF N/A). So, he missed by -5 -He had +0.5 for DJF 2022-3 vs actual of +5. So, he missed by -4.5 -He had -2 for his final DJF 2019-20 (vs earlier fcast of +0.5) vs actual of +3.5. So, his final missed by -5.5. So, his earlier fcast was not as bad (miss by -3). -He had -2 for his final DJF 2016-7 (vs earlier fcast of 0) vs actual of +4. So, his final missed by -6. So, his earlier fcast wasn’t as bad (miss by -4). -He had -2 for both his initial and final fcasts for DJF 2015-6 vs actual of +5.5. So, he missed by -7.5. So, for the 5 mild NYC winters: -The avg of the absolute value of his final fcast misses was 5.5 and his avg bias was -5.5. The range was -4.5 to -7.5. -He revised 2 of the 5 significantly colder, which resulted in larger - misses. ————————— B. The 5 non-mild NYC winters: -2021-2 had -0.5 for final Nov-Mar (vs 0 for initial). The actual was +1 (DJF N/A). So, his final missed by -1.5 and his earlier fcast was slightly better (missed by -1.0). -2020-1 had +0.5 for his final DJF (vs +3 for his earlier). The actual was +0.5. So, his final was perfect! Also, his earlier fcast wasn’t so good (missed by +2.5) meaning his change was great! -2018-9 had -2 for his final/initial DJF vs actual of +0.5. So, he missed by -2.5. -2017-8 had 0 for his final DJF (vs +0.5 for initial). The actual was +0.5. So, his final missed by only -0.5. His earlier fcast was perfect. -2014-5 had -2.5 for his final DJF (vs -3 for initial). The actual was -4. So, his final missed by +1.5. This is the only one of the 10 that was actually cold, for which he missed too warm, and for which he revised warmer. This also was his coldest fcast. His earlier fcast was slightly better (miss by +1). So, for the 5 non-mild NYC winters: -The avg of the absolute value of his final fcast misses was 1. The avg bias was -0.5. The range: -2.5 to +1.5. -He revised 4 of the 5 (3 cooler and 1 warmer). ——————————— C. Summary of all of last 10 winters at NYC: -He did well when winter wasn’t mild as his bias for those was only -0.5 and avg absolute value of misses was only 1. His coldest fcast was for the coldest winter. -He did poorly when winter was mild with a bias/avg miss of -5.5 (avg absolute value of misses was 5.5). He didn’t predict a single mild winter. He was able to predict well the one cold winter. He’d be a much better forecaster if he’d improve on forecasting mild winters but that’s hard to do when he doesn’t forecast them. That’s why I route for him to predict mild and stick with it. -None of his 10 final forecasts were mild (warmest +0.5/NN) (5 were NN and 5 were BN) as they were from warmest to coldest: +0.5, +0.5, 0, -0.5, -1, -2, -2, -2, -2, -2.5 That means that IF his final fcast of +1 for 24-5 doesn’t change, it would still be the warmest since at least 2014-5 despite the cooler revision from +2 to +1. -6 of the 10 winters were +1 or warmer. The avg of the winters was +2 vs his avg of final fcasts of -1. So, he was too cold by 3 on avg for the 10 winters! So, a bias correction of +3 would probably be appropriate. -He predicted on avg -1.3 for the 5 warmest winters and -0.9 for the 5 coldest winters -He had revisions for 6 of the 10. Of these 6, 5 resulted in a worse forecast and only one a better forecast. Of these 6, 5 were revised cooler just like the one he just did for 2024-5. Only 2014-5 was revised slightly warmer. @bluewave @donsutherland1 4 Link to comment Share on other sites More sharing options...
40/70 Benchmark Posted October 13 Share Posted October 13 2 hours ago, snowman19 said: As of right now, I am very skeptical of a cold/snowy February. That would be very atypical of a canonical La Niña. 20-21 was a recent exception, but that was also caused by strong -AO/-NAO blocking, due in very large part to the solar minimum, IMO. At least for NYC, if December is warm, pray for a greater than 3” of snow total for the month. As @bluewave showed, there has been an extremely strong correlation over the last 30 years between total December snowfall and the snowfall for the rest of the season when you have a La Niña in place La Nina being central based in and of itself is not a big deal...especially since it isn't strong Link to comment Share on other sites More sharing options...
qg_omega Posted October 13 Share Posted October 13 9 minutes ago, GaWx said: *Fwiw* JB keeps hinting at a cold Nov/Thanksgiving/Dec for the E US. I’m not putting much weight on that for obvious reasons. Also, on Oct 8th he ditched his +2 for Megapolis for Nov-Mar for a NN (+1). And that’s obviously subject to further cooling in later updates. Keep this in mind: Since at the very least 2014-5 (because they’re all circulating on the internet back to then), JB has yet to have a mild FINAL Megapolis winter forecast. With the number of mild winters (5 of the 10) there have been, that’s not a good record. The ones 3+ AN at NYC since 2014-5 for those who don’t know were 2015-6, 2016-7, 2019-20, 2022-3, and 2023-4. A. The 5 mild NYC winters (vs respective 30 yr mean): -JB had -1 for Nov-Mar 2023-4 vs actual of +4 (DJF N/A). So, he missed by -5 -He had +0.5 for DJF 2022-3 vs actual of +5. So, he missed by -4.5 -He had -2 for his final DJF 2019-20 (vs earlier fcast of +0.5) vs actual of +3.5. So, his final missed by -5.5. So, his earlier fcast was not as bad (miss by -3). -He had -2 for his final DJF 2016-7 (vs earlier fcast of 0) vs actual of +4. So, his final missed by -6. So, his earlier fcast wasn’t as bad (miss by -4). -He had -2 for both his initial and final fcasts for DJF 2015-6 vs actual of +5.5. So, he missed by -7.5. So, for the 5 mild NYC winters: -The avg of the absolute value of his final fcast misses was 5.5 and his avg bias was -5.5. The range was -4.5 to -7.5. -He revised 2 of the 5 significantly colder, which resulted in larger - misses. ————————— B. The 5 non-mild NYC winters: -2021-2 had -0.5 for final Nov-Mar (vs 0 for initial). The actual was +1 (DJF N/A). So, his final missed by -1.5 and his earlier fcast was slightly better (missed by -1.0). -2020-1 had +0.5 for his final DJF (vs +3 for his earlier). The actual was +0.5. So, his final was perfect! Also, his earlier fcast wasn’t so good (missed by +2.5) meaning his change was great! -2018-9 had -2 for his final/initial DJF vs actual of +0.5. So, he missed by -2.5. -2017-8 had 0 for his final DJF (vs +0.5 for initial). The actual was +0.5. So, his final missed by only -0.5. His earlier fcast was perfect. -2014-5 had -2.5 for his final DJF (vs -3 for initial). The actual was -4. So, his final missed by +1.5. This is the only one of the 10 that was actually cold, for which he missed too warm, and for which he revised warmer. This also was his coldest fcast. His earlier fcast was slightly better (miss by +1). So, for the 5 non-mild NYC winters: -The avg of the absolute value of his final fcast misses was 1. The avg bias was -0.5. The range: -2.5 to +1.5. -He revised 4 of the 5 (3 cooler and 1 warmer). ——————————— C. Summary of all of last 10 winters at NYC: -He did well when winter wasn’t mild as his bias for those was only -0.5 and avg absolute value of misses was only 1. His coldest fcast was for the coldest winter. -He did poorly when winter was mild with a bias/avg miss of -5.5 (avg absolute value of misses was 5.5). He didn’t predict a single mild winter. He was able to predict well the one cold winter. He’d be a much better forecaster if he’d improve on forecasting mild winters. That’s why I route for him to predict mild and stick with it. -None of his 10 final forecasts were mild (warmest +0.5/NN) (5 were NN and 5 were BN) as they were from warmest to coldest: +0.5, +0.5, 0, -0.5, -1, -2, -2, -2, -2, -2.5 -6 of the 10 winters were +1 or warmer. The avg of the winters was +2 vs his avg of final fcasts of -1. So, he was too cold by 3 on avg. -He predicted on avg -1.3 for the 5 warmest winters and -0.9 for the 5 coldest winters@bluewave @donsutherland1 Kudos on the detail but Joe B is one of the worst on the internet. Severe cold bias for the clicks 3 1 1 2 Link to comment Share on other sites More sharing options...
George001 Posted October 14 Author Share Posted October 14 4 hours ago, snowman19 said: As of right now, I am very skeptical of a cold/snowy February. That would be very atypical of a canonical La Niña. 20-21 was a recent exception, but that was also caused by strong -AO/-NAO blocking, due in very large part to the solar minimum, IMO. At least for NYC, if December is warm, pray for a greater than 3” of snow total for the month. As @bluewave showed, there has been an extremely strong correlation over the last 30 years between total December snowfall and the snowfall for the rest of the season when you have a La Niña in place I’m skeptical of a cold Feb too but I’m more optimistic about December and January. I agree about December being a sign of things to come, usually in La Niña winters winter shows its hand early. 2 Link to comment Share on other sites More sharing options...
40/70 Benchmark Posted October 14 Share Posted October 14 2 hours ago, qg_omega said: Kudos on the detail but Joe B is one of the worst on the internet. Severe cold bias for the clicks Agree....shame because he was a great met when his intentions were more pure. 1 Link to comment Share on other sites More sharing options...
snowman19 Posted October 14 Share Posted October 14 11 hours ago, SnoSki14 said: No way! You skeptical? That's crazy It’s almost as unbelievable as you being a useless, clueless troll who contributes nothing!! Link to comment Share on other sites More sharing options...
qg_omega Posted October 14 Share Posted October 14 42 minutes ago, 40/70 Benchmark said: Agree....shame because he was a great met when his intentions were more pure. Grew up on him watching videos on accuweather professional and point counter point, shame 2 Link to comment Share on other sites More sharing options...
snowman19 Posted October 14 Share Posted October 14 @Gawx And just looking at the anomalies, we should definitely see the +QBO continue to strengthen more this month… 1 Link to comment Share on other sites More sharing options...
snowman19 Posted October 14 Share Posted October 14 19 hours ago, donsutherland1 said: Currently, the ECMWF outlook continues to call for a warmer than normal winter across much of North America. Ongoing marine heatwaves north of Indonesia (with some Category 3 pockets) and in the North Pacific (pockets of Category 3 and 4 levels) are very likely contributing to the forecast. Those marine heatwaves, should they persist, could favor MJO Phases 4-6 during the winter and a negative to strongly negative PDO. Those outcomes are generally warm in the CONUS except for parts of the West. Agree. Warmest waters still in MJO 4-6. That’s where the atmosphere is going to want to put the strongest and most persistent convection. And besides the marine heatwave there and in the North PAC around Japan, the heatwave around New Foundland in the Atlantic just refuses to budge 1 Link to comment Share on other sites More sharing options...
PhiEaglesfan712 Posted October 14 Share Posted October 14 I think most reasonable people are in agreement that this upcoming winter will be 2 warm months and one cool one, with January being one of the warm months. Accuweather is calling for a warmer than average winter with below average snowfall. February will be the best month for snow chances, but there could also be some storms in early December: https://www.accuweather.com/en/winter-weather/winter-forecast-for-the-us-in-the-2024-25-season/1699821 1 Link to comment Share on other sites More sharing options...
FPizz Posted October 14 Share Posted October 14 Whomever is the forecaster that can find the cold period(s) inside the warmer winters wins and deserves major props. Otherwise I think we all know that it is going to be warmer the vast majority of the time and it doesn't take much skill anymore to predict that. 2 Link to comment Share on other sites More sharing options...
snowman19 Posted October 14 Share Posted October 14 15 hours ago, GaWx said: *Fwiw* JB keeps hinting at a cold Nov/Thanksgiving/Dec for the E US. I’m not putting much weight on that for obvious reasons. Also, on Oct 8th he ditched his +2 for Megapolis for Nov-Mar for a NN (+1). And that’s obviously subject to further cooling in later updates. Keep this in mind: Since at the very least 2014-5 (because they’re all circulating on the internet back to then), JB has yet to have a mild FINAL Megapolis winter forecast. With the number of mild winters (5 of the 10) there have been, that’s not a good record. The ones 3+ AN at NYC since 2014-5 for those who don’t know were 2015-6, 2016-7, 2019-20, 2022-3, and 2023-4. A. The 5 mild NYC winters (vs respective 30 yr mean): -JB had -1 for Nov-Mar 2023-4 vs actual of +4 (DJF N/A). So, he missed by -5 -He had +0.5 for DJF 2022-3 vs actual of +5. So, he missed by -4.5 -He had -2 for his final DJF 2019-20 (vs earlier fcast of +0.5) vs actual of +3.5. So, his final missed by -5.5. So, his earlier fcast was not as bad (miss by -3). -He had -2 for his final DJF 2016-7 (vs earlier fcast of 0) vs actual of +4. So, his final missed by -6. So, his earlier fcast wasn’t as bad (miss by -4). -He had -2 for both his initial and final fcasts for DJF 2015-6 vs actual of +5.5. So, he missed by -7.5. So, for the 5 mild NYC winters: -The avg of the absolute value of his final fcast misses was 5.5 and his avg bias was -5.5. The range was -4.5 to -7.5. -He revised 2 of the 5 significantly colder, which resulted in larger - misses. ————————— B. The 5 non-mild NYC winters: -2021-2 had -0.5 for final Nov-Mar (vs 0 for initial). The actual was +1 (DJF N/A). So, his final missed by -1.5 and his earlier fcast was slightly better (missed by -1.0). -2020-1 had +0.5 for his final DJF (vs +3 for his earlier). The actual was +0.5. So, his final was perfect! Also, his earlier fcast wasn’t so good (missed by +2.5) meaning his change was great! -2018-9 had -2 for his final/initial DJF vs actual of +0.5. So, he missed by -2.5. -2017-8 had 0 for his final DJF (vs +0.5 for initial). The actual was +0.5. So, his final missed by only -0.5. His earlier fcast was perfect. -2014-5 had -2.5 for his final DJF (vs -3 for initial). The actual was -4. So, his final missed by +1.5. This is the only one of the 10 that was actually cold, for which he missed too warm, and for which he revised warmer. This also was his coldest fcast. His earlier fcast was slightly better (miss by +1). So, for the 5 non-mild NYC winters: -The avg of the absolute value of his final fcast misses was 1. The avg bias was -0.5. The range: -2.5 to +1.5. -He revised 4 of the 5 (3 cooler and 1 warmer). ——————————— C. Summary of all of last 10 winters at NYC: -He did well when winter wasn’t mild as his bias for those was only -0.5 and avg absolute value of misses was only 1. His coldest fcast was for the coldest winter. -He did poorly when winter was mild with a bias/avg miss of -5.5 (avg absolute value of misses was 5.5). He didn’t predict a single mild winter. He was able to predict well the one cold winter. He’d be a much better forecaster if he’d improve on forecasting mild winters but that’s hard to do when he doesn’t forecast them. That’s why I route for him to predict mild and stick with it. -None of his 10 final forecasts were mild (warmest +0.5/NN) (5 were NN and 5 were BN) as they were from warmest to coldest: +0.5, +0.5, 0, -0.5, -1, -2, -2, -2, -2, -2.5 That means that IF his final fcast of +1 for 24-5 doesn’t change, it would still be the warmest since at least 2014-5 despite the cooler revision from +2 to +1. -6 of the 10 winters were +1 or warmer. The avg of the winters was +2 vs his avg of final fcasts of -1. So, he was too cold by 3 on avg for the 10 winters! So, a bias correction of +3 would probably be appropriate. -He predicted on avg -1.3 for the 5 warmest winters and -0.9 for the 5 coldest winters -He had revisions for 6 of the 10. Of these 6, 5 resulted in a worse forecast and only one a better forecast. Of these 6, 5 were revised cooler just like the one he just did for 2024-5. Only 2014-5 was revised slightly warmer. @bluewave @donsutherland1 He is already doing his yearly wash, rinse, repeat ‘cold and snowy east coast from Thanksgiving through New Year’s Day’ hype. I have no doubt whatsoever that he will be predicting below average temps and above average snow for the east coast, December - March by the end of November. It’s the time of year where he peddles likes, follows, retweets and subscription money from his I-95 weenie base 1 Link to comment Share on other sites More sharing options...
GaWx Posted October 14 Share Posted October 14 22 hours ago, donsutherland1 said: Currently, the ECMWF outlook continues to call for a warmer than normal winter across much of North America. Don, I just realized that this Euro winter outlook has 01/09/24 stamped on it. Did you realize that? Is this the latest? What about 01/10/24? Link to comment Share on other sites More sharing options...
donsutherland1 Posted October 14 Share Posted October 14 Below is a link to the Accuweather winter forecast: https://www.accuweather.com/en/winter-weather/winter-forecast-for-the-us-in-the-2024-25-season/1699821 2 Link to comment Share on other sites More sharing options...
donsutherland1 Posted October 14 Share Posted October 14 9 minutes ago, GaWx said: Don, I just realized that this Euro winter outlook has 01/09/24 stamped on it. Did you realize that? Is this the latest? What about 01/10/24? I fixed it. I accidentally pasted the earlier map. 1 Link to comment Share on other sites More sharing options...
michsnowfreak Posted October 14 Share Posted October 14 20 hours ago, bluewave said: No. I used dense rank sorting by temperature. https://medium.com/@kumarsatwik25/rank-and-dense-rank-in-sql-23ffcf77611e DENSE RANK (): The DENSE_RANK () function is a window function in SQL that assigns a unique rank to each distinct row within a result set, based on a specified ordering. Unlike the RANK () function, DENSE_RANK () does not leave gapsin the rank sequence when there are ties in the ordering. It ensures that rows with the same values in the ORDER BY clause receive the same rank, and the next rank is then incremented without any gaps. Oh ok that explains it. Definitely not a fan of Dense-Ranking. It gives a completely inaccurate assessment. 1 Link to comment Share on other sites More sharing options...
michsnowfreak Posted October 14 Share Posted October 14 18 hours ago, mitchnick said: You can be skeptical, it's your perogative. But if people are going to post model runs, it seems only fair to consider the entire run and not just the parts they agree with. im skeptical of a mild/warm December, at least here. Link to comment Share on other sites More sharing options...
Eduardo Posted October 14 Share Posted October 14 16 hours ago, GaWx said: *Fwiw* JB keeps hinting at a cold Nov/Thanksgiving/Dec for the E US. I’m not putting much weight on that for obvious reasons. Also, on Oct 8th he ditched his +2 for Megapolis for Nov-Mar for a NN (+1). And that’s obviously subject to further cooling in later updates. Keep this in mind: Since at the very least 2014-5 (because they’re all circulating on the internet back to then), JB has yet to have a mild FINAL Megapolis winter forecast. With the number of mild winters (5 of the 10) there have been, that’s not a good record. The ones 3+ AN at NYC since 2014-5 for those who don’t know were 2015-6, 2016-7, 2019-20, 2022-3, and 2023-4. A. The 5 mild NYC winters (vs respective 30 yr mean): -JB had -1 for Nov-Mar 2023-4 vs actual of +4 (DJF N/A). So, he missed by -5 -He had +0.5 for DJF 2022-3 vs actual of +5. So, he missed by -4.5 -He had -2 for his final DJF 2019-20 (vs earlier fcast of +0.5) vs actual of +3.5. So, his final missed by -5.5. So, his earlier fcast was not as bad (miss by -3). -He had -2 for his final DJF 2016-7 (vs earlier fcast of 0) vs actual of +4. So, his final missed by -6. So, his earlier fcast wasn’t as bad (miss by -4). -He had -2 for both his initial and final fcasts for DJF 2015-6 vs actual of +5.5. So, he missed by -7.5. So, for the 5 mild NYC winters: -The avg of the absolute value of his final fcast misses was 5.5 and his avg bias was -5.5. The range was -4.5 to -7.5. -He revised 2 of the 5 significantly colder, which resulted in larger - misses. ————————— B. The 5 non-mild NYC winters: -2021-2 had -0.5 for final Nov-Mar (vs 0 for initial). The actual was +1 (DJF N/A). So, his final missed by -1.5 and his earlier fcast was slightly better (missed by -1.0). -2020-1 had +0.5 for his final DJF (vs +3 for his earlier). The actual was +0.5. So, his final was perfect! Also, his earlier fcast wasn’t so good (missed by +2.5) meaning his change was great! -2018-9 had -2 for his final/initial DJF vs actual of +0.5. So, he missed by -2.5. -2017-8 had 0 for his final DJF (vs +0.5 for initial). The actual was +0.5. So, his final missed by only -0.5. His earlier fcast was perfect. -2014-5 had -2.5 for his final DJF (vs -3 for initial). The actual was -4. So, his final missed by +1.5. This is the only one of the 10 that was actually cold, for which he missed too warm, and for which he revised warmer. This also was his coldest fcast. His earlier fcast was slightly better (miss by +1). So, for the 5 non-mild NYC winters: -The avg of the absolute value of his final fcast misses was 1. The avg bias was -0.5. The range: -2.5 to +1.5. -He revised 4 of the 5 (3 cooler and 1 warmer). ——————————— C. Summary of all of last 10 winters at NYC: -He did well when winter wasn’t mild as his bias for those was only -0.5 and avg absolute value of misses was only 1. His coldest fcast was for the coldest winter. -He did poorly when winter was mild with a bias/avg miss of -5.5 (avg absolute value of misses was 5.5). He didn’t predict a single mild winter. He was able to predict well the one cold winter. He’d be a much better forecaster if he’d improve on forecasting mild winters but that’s hard to do when he doesn’t forecast them. That’s why I route for him to predict mild and stick with it. -None of his 10 final forecasts were mild (warmest +0.5/NN) (5 were NN and 5 were BN) as they were from warmest to coldest: +0.5, +0.5, 0, -0.5, -1, -2, -2, -2, -2, -2.5 That means that IF his final fcast of +1 for 24-5 doesn’t change, it would still be the warmest since at least 2014-5 despite the cooler revision from +2 to +1. -6 of the 10 winters were +1 or warmer. The avg of the winters was +2 vs his avg of final fcasts of -1. So, he was too cold by 3 on avg for the 10 winters! So, a bias correction of +3 would probably be appropriate. -He predicted on avg -1.3 for the 5 warmest winters and -0.9 for the 5 coldest winters -He had revisions for 6 of the 10. Of these 6, 5 resulted in a worse forecast and only one a better forecast. Of these 6, 5 were revised cooler just like the one he just did for 2024-5. Only 2014-5 was revised slightly warmer. @bluewave @donsutherland1 You might be the only person alive with the patience and dedication to work up a verification like this! Like many of you, I also grew up watching JB’s Accuweather vids and loved his enthusiasm. I still think he’s probably an all-around good guy, but I can’t for the life of me understand how a person with such scientific training can allow their brain to become pickled in the nonsense he puts out in almost crusade-like fashion. Fascinating, in a way. 1 1 Link to comment Share on other sites More sharing options...
snowman19 Posted October 14 Share Posted October 14 @40/70 Benchmark Very good call on this being a CP La Niña months ago. You never bought into this being an EP event at any point in time 1 1 Link to comment Share on other sites More sharing options...
michsnowfreak Posted October 14 Share Posted October 14 4 hours ago, FPizz said: Whomever is the forecaster that can find the cold period(s) inside the warmer winters wins and deserves major props. Otherwise I think we all know that it is going to be warmer the vast majority of the time and it doesn't take much skill anymore to predict that. Actually there still are and will be colder than normal winters, so when that happens It will be interesting to see how the models and the pros do with the forecast. This winter could be a good test with a pretty decent consensus forming for colder than normal anomalies developing in Canada as winter progresses....how far into the northern US they will spread is a bigger question mark. Link to comment Share on other sites More sharing options...
GaWx Posted October 14 Share Posted October 14 40 minutes ago, michsnowfreak said: Actually there still are and will be colder than normal winters, so when that happens It will be interesting to see how the models and the pros do with the forecast. This winter could be a good test with a pretty decent consensus forming for colder than normal anomalies developing in Canada as winter progresses....how far into the northern US they will spread is a bigger question mark. The last two Euros at least have had no BN anomalies in Canada or in any of N America on the latest one for that matter for DJF as a whole. The runs are pretty strongly suggesting AN dominating at least the E half of Canada. I haven’t seen the month by month breakdown but the Euro has a much stronger warm signal than the CFS. 1 Link to comment Share on other sites More sharing options...
bluewave Posted October 14 Share Posted October 14 3 hours ago, michsnowfreak said: Oh ok that explains it. Definitely not a fan of Dense-Ranking. It gives a completely inaccurate assessment. It’s a more accurate way of keeping statistical rankings when the distribution becomes more skewed to either the top or bottom of the range. The old way came about when we were in a stable climate era and month after month wasn’t in the top 10 warmest. So ranks aren’t skipped since there can be numerous ties near the top 10 in a rapidly warming climate. Creating gaps due to numerous ties artificially pushes similar temperatures out of the top 10. The NWS out in Upton has been doing it this way also. DENSE_RANK () does not leave gapsin the rank sequence when there are ties in the ordering. It ensures that rows with the same values in the ORDER BY clause receive the same rank, and the next rank is then incremented without any gaps 2 1 Link to comment Share on other sites More sharing options...
michsnowfreak Posted October 14 Share Posted October 14 2 hours ago, GaWx said: The last two Euros at least have had no BN anomalies in Canada or in any of N America on the latest one for that matter for DJF as a whole. The runs are pretty strongly suggesting AN dominating at least the E half of Canada. I haven’t seen the month by month breakdown but the Euro has a much stronger warm signal than the CFS. To clarify, I was referring to the model consensus, not just the euro. I don't buy the euros broadbrushed temp consensus. If it comes even close to verifying it will be unlike any other nina, with temp anomalies ranging from near normal in southwest Canada to a total range of just +0.5-1.5° in the entire eastern third of the country. 1 Link to comment Share on other sites More sharing options...
michsnowfreak Posted October 14 Share Posted October 14 2 hours ago, bluewave said: It’s a more accurate way of keeping statistical rankings when the distribution becomes more skewed to either the top or bottom of the range. The old way came about when we were in a stable climate era and month after month wasn’t in the top 10 warmest. So ranks aren’t skipped since there can be numerous ties near the top 10 in a rapidly warming climate. Creating gaps due to numerous ties artificially pushes similar temperatures out of the top 10. The NWS out in Upton has been doing it this way also. DENSE_RANK () does not leave gapsin the rank sequence when there are ties in the ordering. It ensures that rows with the same values in the ORDER BY clause receive the same rank, and the next rank is then incremented without any gaps I'm shocked Upton does that. I've never seen any other nws do that. Regardless of who does or does not use it, with all due respect I think it's an absolutely ridiculous way of ranking things. So let's say there's a 100 year sample and you finish with a mean temp of 50.5°. In the other 99 years, 18 years were 50.6° or warmer and 81 years were 50.4° or colder. But instead of saying it's the 19th warmest on record, you use dense ranking and several were ties. So instead you say it's the 14th warmest on record, even though 18 years were warmer. Any years that finish with the same mean temp as a previous year get penalized for doing so and thus cease to exist. . Ridiculous imo. 1 Link to comment Share on other sites More sharing options...
bluewave Posted October 14 Share Posted October 14 24 minutes ago, michsnowfreak said: I'm shocked Upton does that. I've never seen any other nws do that. Regardless of who does or does not use it, with all due respect I think it's an absolutely ridiculous way of ranking things. So let's say there's a 100 year sample and you finish with a mean temp of 50.5°. In the other 99 years, 18 years were 50.6° or warmer and 81 years were 50.4° or colder. But instead of saying it's the 19th warmest on record, you use dense ranking and several were ties. So instead you say it's the 14th warmest on record, even though 18 years were warmer. Any years that finish with the same mean temp as a previous year get penalized for doing so and thus cease to exist. . Ridiculous imo. It makes perfect sense if you understand that the ranking is done by the temperature and not the year. So multiple years can share the same temperature. It makes no sense to leave a gap in the ranks just because a tie occurs. Ties in scores are handled by giving the tied rows the same rank, and the next rank is then incremented without any gaps. 1 1 1 Link to comment Share on other sites More sharing options...
donsutherland1 Posted October 14 Share Posted October 14 4 hours ago, GaWx said: The last two Euros at least have had no BN anomalies in Canada or in any of N America on the latest one for that matter for DJF as a whole. The runs are pretty strongly suggesting AN dominating at least the E half of Canada. I haven’t seen the month by month breakdown but the Euro has a much stronger warm signal than the CFS. Fortunately, some sizable misses are still possible from this timeframe. For example, the October 2023 ECMWF forecast had most of western Canada 1°C-2°C above normal for February 2024, but the month actually wound up colder than normal in parts of that area. I suspect that there will be some BN anomalies in North America. The most likely areas are those areas shown with near normal readings right now. But that doesn't preclude some cold shots into the warmer areas on a periodic basis. I'm still reasonably optimistic that snowfall will be near normal to somewhat above normal from Chicago to Toronto. On a DJF basis, it was too cold with the Plains States. 1 Link to comment Share on other sites More sharing options...
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now