It’s coming. I’m having some issues/fun with getting everything uploaded without crashing the server. It’s not the tippity top of my priorities by now, but I’m hoping for some time in May. If I’m lucky, hopefully before the last snow melts out of the Midwest (and, jeezum, before the Larium Glacier melts out up in the Keweenaw).
We have been able to wrangle the data and now have a one-minute long time lapse of the Birkie. It’s posted on Youtube. We might tweak it as we finish posting the data and fully launch the 2013 stats site, but I can’t not post this now, right?
So, I need some help. I have some really fun data. It shows, for every minute during the Birkie, how many skiers were at each quarter kilometer of the course. The example above, for instance, shows the state of the race—the actual position of skiers on the course, by wave—at 9:03 a.m.. The fourth wave is just setting out while the earliest waves are already forming packs. The front of the third wave is catching the front of the second; the front of the second wave is making progress in to the first. Pretty cool, right? (These are all based on four split times, so it’s not a perfect representation of exactly who is where, but it’s a pretty good approximation.)
I want to animate these, probably in to a Youtube or Vimeo and play it at, say, 5–10 fps (so each hour is 6–12 seconds and the whole of the race takes a minute or two). But there are (at least) two relatively major stumbling blocks, and if anyone can help, I’d certainly be grateful:
- I need to export 600 images. One for each minute, 8 a.m. to 6 p.m. I certainly could do this manually, but I would be driven to the drink well before the first wave arrived in Hayward. In other words, I need to automate this. I think it could be done in Excel; you’d need a Macro to find a piece of text, increment it, export the chart and loop back to the beginning. (Why the find-replace? The method I used to get the data was to calculate the position for each racer for each minute. I then ran a separate count for the number of skiers in each 250m block for a particular minute and made a chart. Then I run a Find-Replace to go from, for instance, cell AB to cell AC which increments one minute forward. If anyone has a better idea how to do this in Excel or in a SQL environment, I’m all ears.)
- Take all these images in to some kind of video-making program and compress it in to a “time lapse” thing. I think I know someone who can do this, but if anyone has those skills, and if we get to that point, I’d be very happy for the help.
If you can provide any assistance, it would be fantastic. Drop me an email (ari.ofsevit@gmail) or comment here (you need to log in; otherwise we get lots of spam) to get in touch. Extra bonus points if you’re patient enough to explain to me what you’re doing, instead of just doing it. I’d certainly give you lots of on-site recognition for any help, and probably buy you a beer in Hayward next February.
An obscure Simpsons quote notwithstanding, it’s time for me to brush up on Unix time. Why? Because I have a dataset with close to 20,000 rows of Birkie data, and the split times are in Unix time. It’s exact, yes, but kind of kooky to sort in to real-life time.
In any case, look for data updates in coming days. Data!
Updates: we’re getting some really fun data for the number of people each skier passed and was passed by, split data, and the like. Stay tuned!
A few new things to note:
- New race reports. Check them out. And keep them coming!
- New Birkie Trail Subway-style Map. Yes, I am that much of a nerd.
- A post about the 100k day I skied on the Birkie trail a few years back. The weather should be just about right for something like that this year if the grooming holds.
In the northeast, there is skiing to be had. A surprise foot-and-a-half of snow last week prolonged the season; I’m slated to ski a 50k later this week. Last year, it was the last day of skiing, as the snow melted off the course during the race. It was followed by a week of temperatures in the 70s and 80s (!) which melted every last bit of snow in New England. In the midwest, reports are that there is skiable snow on the trails, with perhaps more to come. Duluth has two feet on the ground, and the UP may not melt until June. (And there’s no 2012-style week-in-the-70s coming there, either.)
This is all well and good, since I still don’t have results from the Birkie and there are still discrepancies. So, keep on skiing, and we’ll have data to pore over once the snow melts and you can sit outside with a mai tai and get a tan (July). And are we going to compare fat bike results with ski results? Will the new Pope be Catholic?
Well, I have a rough approximation here.
Now, it’s time to go skiing.
[update 3/7] I am planning a more stylized version of this map which even further from a geographic view. Plus some other enhancements. I’ll put those on the stylized map page linked above.
Yes, I know that one of the background header images is upside-down. Will fix later.
In the mean time, there are more race reports up. Read them, get the fever, and send your own!
Our race report section is rapidly growing. We have five race reports so far (and want to add more) from a variety of skiers, with some great stories. There’s Shawn, the first ever blind woman to ski the Birkie, and her guide, veteran Birkie Trail skier Jesse. Then there’s Nick, who dropped 60 pounds and quit smoking in pursuit of a Main Street finish. And Rob, a triathlete who skied most of his first Birkie with a full-on snapped ski. Oh, and if you really want, you can read my race report as well.