 You don't need to be an 'investor' to invest in Singletrack: 6 days left: 95% of target - Find out more
Two working days to go till shut down and woke after dreaming of work stuff. All was fine this afternoon when confirming today's pour (thursday) looks like it's gonna piss down now!!!!
Can't be many other careers where you can be so far off the mark on a regular basis and still keep your job!
They're not ALL men you know! Sexist neanderthal knuckle-dagger that you are!!
#inbeforemike
Yeah I hear ya. Weather bod's here are predicting a measly 33c on Christmas day. Hardly worth taking an esky for the Christmas beach beverages.
Currently listening to the rain hammering off our roof, its a light shower apparently!
Some guy scraped BBC weather forecasts for a year and compared to actual. Even the day before predictions were barely 50% right on things like will rain. Turned out to be more likely right if you simply assume the weather will be the same as the day before.
Citation needed.
I've heard that said before, if you predict tomorrow will be the same as today you'll be 75% right, forecasters aren't much better at 80%. I think there are 2 issues, first they are trying to be far more accurate, not just it will rain but it will be heavy rain between 10 and 12 followed by showers. Second thing is they usually over do it, something linked in part back to Mickael Fish's hurricane and partly to do with making the weather sexier like giving a mildly flowy bit of weather a name like it's a hurricane in Miami. Either way their detail accuracy is pretty poor around here.
I'm guessing its because its quite hard to do
It changes constantly, so a prediction now based on what data you have will be different to a prediction in another half hour with the new Data.
It's because they are the best we have at forecasting and without them we would simply be guessing.
Also wrightyson do you know how to actually read a weather forecast? They have a percentage probability attached to rain. So it might only be a 50% chance but still get a rain symbol.
Also the timing of events is usually the biggest variable. So for example it might say rain overnight, in which case the daytime gets a fine symbol, however the rain is only a few hours early so it appears in the evening and everyone sees it then they complain.
Anyone who understands the weather even a bit knows how difficult it is to predict and how accurate or otherwise their forecasts are.
What would you prefer? You are asking for them to be fired? How would that help? You think we should just keep firing and hiring people every time there is a bad forecast? How would that help? You really think there's a random person in a job centre somewhere who's great at weather forecasting and is just waiting for the opportunity? I must admit I do get cross with people for criticising scientists for the current state of science. Who do you think would be better?
This is the best humanity can currently manage, so deal with it.
I wouldn't touch the BBC weather forecast with a barge pole, Meteo group took over the forecasts from The Met Office this year and since then, they have been complete garbage.
www.metoffice.gov.uk
Combined with https://www.wunderground.com us usually in the ballpark of reality.
#inbeforemike
😀
Can we add #inbeforethegingerselfie as a Bike forum variant of this?
I use a screen on my phone with a rolling 60hr meteogram widget for work and another for home. They shift a bit a eg Saturday morning gets closer, they scroll closer, but it's usually a pretty good indication. You can click through to the site if you want to look at 10 day forecast, but obviously those are going shift more.
For immediate "Will I get wetter if I leave now or in an hour?" questions, I use Rain Alarm.
I'd recommend both.
The problem with sites like the BBC is that they try to be too definitive. If the computer models show showers across the country, they forecast rain where the showers are and dry in between - but where the showers actually occur is more random. The % chance of rain they give would be more useful, except they bias it on the pessimistic side so it get ignored.
I use windy.com and look at the general trend on each of the four models - if they agree, it's likely they'll be right, if not, the conditions are less stable and harder to forecast so don't depend on it.
Couldn't possibly be because forecasting the weather is hard, could it? BBC forecast for my area accurate(ish) from about 4 days out. Only thing they got wrong yesterday was wind gusting to 25kt, not 40, and I think that's impressive given it was made 4 days before.
I was told that to maintain the same confidence beyond 48 hours requires roughly doubling the computer power / time thrown at the problem for each extra hour.
There's no such thing as bad weather, only bad preparation of excuses not to go out in it.
Nedrapier +1 - I use both of those plus a couple of others which are based on different models. If they all say the same sort of thing then it's plausible, if they differ then you know there's a fair bit of uncertainty.
The meteoblue android app has a page to view all the different model forecasts overlaid onto each other which is a handy way of seeing when they diverge - eg at the mo they're quite consistent to sat PM then start diverging a fair bit, though even the detail of lunchtime today differs from clear skies to p'ing rain.
https://singletrackmag.com/forum/topic/how-do-the-met-office-get-away-with-it/
so have they been alright for the 9 months since the last thread? thats not too bad if so...
Simple answer: Because he is trying to predict the future.
How good are you at that?