January 2016 Pattern Disco

[Ginx snewx]

When you say long range is a crap shoot I assume you mean in general, not snowfall.

 

Snowfall is much, much harder to predict on the monthly scale. I think Will said it too, people tend to relate probabilities to deterministic forecasts. If we say good chance at above normal snowfall, people want to know X number of inches above normal for their backyard. It doesn't work that way.

 

Temps are fairly evenly distributed, QPF even less so, and snowfall even less so that QPF.

the whole discussion was supposed to be about snowfall, lol I do not know where it changed

[powderfreak]

The discussion was on Jan 25th are probabilities predictive enough to say the next two months we are able to reach normal climo snow. I say no they are not predictive enough, PF says yes they are.

I don't think they are predictive enough to say for absolute certain, but I would bet on those odds.

And it was for areas that average say 70-80"+ that are currently at like 10-24" on the season.

With forecasting you're never "certain", you go with the highest chance solution. So with that I would forecast that we don't make climo snow up here, in western New England, mountains, up here, and I'd even take the bet at ORH if there's only a 15% chance of them getting that much snow past Feb 1st.

Would you be able to seriously forecast that we make average snowfall by seasons end?

[Ginx snewx]

Everyone just angry over your arguing

Well PF already told you. No one is arguing, discussion of opinions is allowed here in the USA. Please refrain from trying to stifle it by presenting odd responses. Thank you in advance

[RUNNAWAYICEBERG]

Good discussion. Unfortunately, i have nothing to add because i suck beyond d7, even inside d7 is a head scratcher for me.

Just passing by upping my post count, go on.

[ORH_wxman]

Certain predictions absolutely have skill.

 

If I'm making predictions of ORH snowfall against a person who is totally ignorant to the climatology of ORH snowfall, I'm going to produce much higher skilled forecasts than them. They'll throw out random numbers like 27" or 15" and I'll know to avoid those numbers because it takes an extremely rare winter to get numbers that low.

 

Likewise, I'm going to beat someone who consistently predicts 120" at ORH.

 

 

Yes, there is plenty of chaos in between, but it does not "reign" when we're talking about extreme numbers. We have a high sample size of snowfall at ORH on a seasonal and monthly scale. You normally beat the extreme predictions. Predicting ORH to finish at climo probably is not "extreme", but it will take a pretty unlikely result. SO my money is on them not reaching climo.

 

I certainly hope I'm wrong, I'd say I have about 20% chance of being wrong. Maybe 25%...hard to put an absolute number on it, but I'm extremely confident it's less than 50.

[Ginx snewx]

I don't think they are predictive enough to say for absolute certain, but I would bet on those odds.

And it was for areas that average say 70-80"+ that are currently at like 10-24" on the season.

With forecasting you're never "certain", you go with the highest chance solution. So with that I would forecast that we don't make climo snow up here, in western New England, mountains, up here, and I'd even take the bet at ORH if there's only a 15% chance of them getting that much snow past Feb 1st.

Would you be able to seriously forecast that we make average snowfall by seasons end?

I wouldn't forecast anything PF, for all you know a Feb 2010  4 footer could hit the Berks or a 2006 or a Valentines day hit you. Honestly I don't know whats happening for snow. however if you are playing the probability game second half Nino winter is probably stout

[OceanStWx]

Certain predictions absolutely have skill.

 

If I'm making predictions of ORH snowfall against a person who is totally ignorant to the climatology of ORH snowfall, I'm going to produce much higher skilled forecasts than them. They'll throw out random numbers like 27" or 15" and I'll know to avoid those numbers because it takes an extremely rare winter to get numbers that low.

 

Likewise, I'm going to beat someone who consistently predicts 120" at ORH.

 

 

Yes, there is plenty of chaos in between, but it does not "reign" when we're talking about extreme numbers. We have a high sample size of snowfall at ORH on a seasonal and monthly scale. You normally beat the extreme predictions. Predicting ORH to finish at climo probably is not "extreme", but it will take a pretty unlikely result. SO my money is on them not reaching climo.

 

I certainly hope I'm wrong, I'd say I have about 20% chance of being wrong. Maybe 25%...hard to put an absolute number on it, but I'm extremely confident it's less than 50.

 

It's the narrowing of the goal posts.

 

ORH has gone from NFL width to arena league.

[Ginx snewx]

It's the narrowing of the goal posts.

 

ORH has gone from NFL width to arena league.

Think Gostowski would make that? speaking of probabilities.....

[OceanStWx]

Think Gostowski would make that? speaking of probabilities.....

 

Given that he hadn't missed in the better part of a decade, I'll stick with those odds. Never going to catch the extremes.

[Ginx snewx]

Given that he hadn't missed in the better part of a decade, I'll stick with those odds. Never going to catch the extremes.

predictable yet not, exactly why I don't gamble for real. I did play Wiz's Yahoo pool but all my winnings go to charity. I definitely didn't use probabilities in that! Snowfall is way unpredictable LR. I will leave it at that

[powderfreak]

I wouldn't forecast anything PF, for all you know a Feb 2010 4 footer could hit the Berks or a 2006 or a Valentines day hit you. Honestly I don't know whats happening for snow. however if you are playing the probability game second half Nino winter is probably stout

Yes those events certainly could happen. But even then we could still be below normal snowfall for most of those areas of 80"+ averages. I'm not trying to be a Debbie here (though it's coming out like that haha), I'm just going along with the probability discussion and do believe it holds value for future expectations.

I'm in Will's camp that I hope I'm wrong but I'm just playing the odds. And if I'm going to do that, I have to acknowledge what you said that it should be an exciting second half regardless (El Nino backend winters)...can't cherry pick the probabilities I wish to acknowledge. The probabilities just tell me not to expect to fully recover from the first half, but we can always hope.

[ORH_wxman]

Nobody is arguing that the numbers are certain. That isn't what probabilities are about.

I'm hoping we catch the lower probability result that we get to climo snowfall. Its not impossible. We will follow it and hope it happens because that is what we do. Just like we watch a baseball game and hope a guy who has a 70% chance of making an out ends up getting on base with 2 outs in the 9th.

[CoastalWx]

Nobody is arguing that the numbers are certain. That isn't what probabilities are about.

I'm hoping we catch the lower probability result that we get to climo snowfall. Its not impossible. We will follow it and hope it happens because that is what we do. Just like we watch a baseball game and hope a guy who has a 70% chance of making an out ends up getting on base with 2 outs in the 9th.

 

There are always the Bellhorn's of the world.

[40/70 Benchmark]

There are always the Bellhorn's of the world.

You love him.

What a Bellhorn fetish.

[Damage In Tolland]

There are always the Bellhorn's of the world.

There is always a chance of an Aaron Boone type result

[40/70 Benchmark]

Yes those events certainly could happen. But even then we could still be below normal snowfall for most of those areas of 80"+ averages. I'm not trying to be a Debbie here (though it's coming out like that haha), I'm just going along with the probability discussion and do believe it holds value for future expectations.

I'm in Will's camp that I hope I'm wrong but I'm just playing the odds. And if I'm going to do that, I have to acknowledge what you said that it should be an exciting second half regardless (El Nino backend winters)...can't cherry pick the probabilities I wish to acknowledge. The probabilities just tell me not to expect to fully recover from the first half, but we can always hope.

Does it really matter what the aggregate percentage is for the season if you have a fun second half?

Put the calculator does in just enjoy the hobby.

[40/70 Benchmark]

There is always a chance of an Aaron Boone type result

Yes.

This winter may be down 3-0, but the series is not over.

[Damage In Tolland]

Yes.

This winter may be down 3-0, but the series is not over.

Correct. Single digit snowfalls in late Jan. are minuscule, but then again so was Bucky Dent and he launched one

[40/70 Benchmark]

Correct. Single digit snowfalls in late Jan. are minuscule, but then again so was Bucky Dent and he launched one

That one was poor.

[Damage In Tolland]

That one was poor.

Pedro was small and lacked height, but made up for it with arm speed, but Posada's bat was strong enough to drop one in.. Plenty of time for a bloop hit off this winter and still come in with a win

[OceanStWx]

That one was poor.

 

Thoughtful discussion of probabilistic forecasting has him reaching a bit.

[powderfreak]

Does it really matter what the aggregate percentage is for the season if you have a fun second half?

Put the calculator does in just enjoy the hobby.

No it doesn't matter I guess, it's just a conversation about the current winter in a weather forum. Putting our current winter in perspective.

[powderfreak]

Nobody is arguing that the numbers are certain. That isn't what probabilities are about.

I'm hoping we catch the lower probability result that we get to climo snowfall. Its not impossible. We will follow it and hope it happens because that is what we do. Just like we watch a baseball game and hope a guy who has a 70% chance of making an out ends up getting on base with 2 outs in the 9th.

Good post, feel the same.

[40/70 Benchmark]

Thoughtful discussion of probabilistic forecasting has him reaching a bit.

 

Its times like today that I yearn from spring training.

[donsutherland1]

A few thoughts on snowfall and probabilities:

 

1. I agree that seasonal and even monthly snowfall is difficult to forecast reliably (IMO, seasonal forecasting is still closer to a guess than anything else).

 

2. Probabilities can provide insight over the medium-term, though. For example, the statistics from 1950-2015 showed that the frequency of measurable snow during AO-/PNA+ patterns was about 9% higher than climatology in New York City and about 23% higher in Washington, DC.

 

So, assuming the pattern verified and if the 1950-2015 period were representative (and we're not talking about a handful of dates), one could reasonably suggest that there was a higher risk relative to climatology for measurable snowfall, especially in Washington, DC. Nothing was assured.

 

3. In terms of the January 23 +/- 15 day period, nearly 80% of Washington's 10" or greater snowstorms occurred with an AO-/PNA+ pattern (starting date of the storm). So when the guidance was suggesting a big impact there, one could have more confidence let's say 3-5 days out, than if one were dealing with a setup that accounted for almost none of those snowstorms. Again, nothing was assured and there can always be outliers (hence the existence of low probabilities, not zero ones).

 

All said, I believe probabilistic forecasting has value when dealing with the medium-range forecasting horizon. Probabilities, even high ones, should not be treated with a deterministic lens. In the end, even if the proverbial dice are loaded toward a given outcome (e.g., much of winter has passed), that doesn't assure that such an outcome will be realized. The extraordinary snowy period that developed in southern New England during January 2015 provides one such example.

[JC-CT]

Good discussion. Unfortunately, i have nothing to add because i suck beyond d7, even inside d7 is a head scratcher for me.

Just passing by upping my post count, go on.

+1

(my own post count, that is)

[CoastalWx]

This whole debate is silly. What is Steve even arguing?

[OceanStWx]

This whole debate is silly. What is Steve even arguing?

 

I'm pretty sure that Steve is saying long range snow forecasting is bunk. And powderfreak is saying long range snowfall forecasts are mostly bunk, but there is a shred of usefulness.

 

So really we're all arguing the same thing.

 

That's how I read it.

[RUNNAWAYICEBERG]

Ginx, quick question. How do you go about issuing your LR weekly forecasts that you post? do you use probabilities or is it more like a farmers almanac type deal?

[JC-CT]

I'm pretty sure that Steve is saying long range snow forecasting is bunk. And powderfreak is saying long range snowfall forecasts are mostly bunk, but there is a shred of usefulness.

So really we're all arguing the same thing.

That's how I read it.

I read it that Steve is arguing for snow.