Boston analysis

It is really getting funny. The flat earthers, oh I mean no winders, are either getting unbelievably desperate, or they've just been replaced by pure trolls, trying to keep a non-debate going.

I can hardly wait to see what they come up with next. Perhaps proof of alien ray guns slowing down these runners, otherwise they would have run 5 minutes faster!

Heh. This are independant reporting stations, nothing official. You can zoom right in and see them in satellite view. Both of these are small buildings, likely houses, in residential areas surrounded by trees.

I claim no knowledge of how these weather stations work - you got me currious so I zoomed into the one Malmo used for his charts (KBOS). Here's where the wind is measured - I presume at the top of that tower:

That's not going to be very indicative of the wind 20 miles inland - at gound level. I checked most every station I could find along the course - the ones I posted were among the higher wind measurements between Hopkinton and Boston.

Obviously (to me) the marathon was aided - but that graphic based on this weatherstation overstates the magnitude of benefit - I can't find anything inland that substantiates there was near that much wind along the course.

rekrunner wrote:

First I'll re-ask, what is it that you think ARRS is supplying that needs to be critiqued? The data is all publicly available.

Normally you are a methodical and dispassionate poster, but this is getting ridiculous.

People are asking a straightforward question and you appear to be trying to give Ken Young cover, which leads me to believe (perhaps erroneously) that you are either close to Young, or you are Ken Young himself? Carrying someones water for them is one thing, but carrying an empty bucket is quite another.

The questions that need to be asked are obvious: Who, what, when, where, why and how?

1) Which athletes are being analyzed and for which events and over what period?

2) Who is the data being analyzed? What is the methodology?

3) What is the rational behind the choice of the data set and the methods used to analyze the data?

4) How is the conclusion related to, and supported by, the data?

malmo wrote:

By definition, clerical errors due to data entry ARE inconsistent. That's what makes them data entry errors.

True, I just found it strange that they made the same error over and over again. At first, I assumed they had just mistakenly included the 2010 results with the non-aided ones. But actually they did this for 6 of the past 10 year-best lists.

inconsistent wrote:

True, I just found it strange that they made the same error over and over again. At first, I assumed they had just mistakenly included the 2010 results with the non-aided ones. But actually they did this for 6 of the past 10 year-best lists.

Actually, what I found strange were the 3 years 2004-2006 in the middle where they appropriately separated them from the main list, but then went right back to making the same error in 2007.

Also that they correctly separated the 2006 results on the year list, but then included it on the all-time list.

kayry wrote:

I claim no knowledge of how these weather stations work - you got me currious so I zoomed into the one Malmo used for his charts (KBOS). Here's where the wind is measured - I presume at the top of that tower:

http://www.bing.com/maps/?v=2&cp=r1t0h292h0w6&lvl=19.648362001507973&dir=0.3573473691438799&sty=b

That's not going to be very indicative of the wind 20 miles inland - at gound level. I checked most every station I could find along the course - the ones I posted were among the higher wind measurements between Hopkinton and Boston.

Obviously (to me) the marathon was aided - but that graphic based on this weatherstation overstates the magnitude of benefit - I can't find anything inland that substantiates there was near that much wind along the course.

The Wellesley station you linked to, the guy has stuff about it on his blog. It's kind of cool actually.

From what I gather, the readings for temp, humidity and baro pressure should match pretty close to known values (these shouldn't vary by much at all from nearby stations) but for wind readings, there's a lot of latitude granted.

OK, I'm sorry but I just have to ask - are you a comedian? Seriously.

You write: "You look silly concluding that they think it should be a world record, especially when they never said it"

I will admit this ^ is simply genius. I never concluded that they think... If you think I did, please re-read my post (which you've conveniently copied). So, I look silly concluding something that I never concluded because they never said it?!? Pure genius.

Oh, and how about this: "You don't need the algorithm for most questions. You do need the algorithm if you doubt that the times are 1:37 fast for men, 1:42 fast for women, and 13.5s slow on average.:

Ummm...pretty sure that this is exactly what is being doubted. That this is being doubted hasn't really been kept a secret now has it?

Now, about that algorithm...

What? The notes on the ARRS website says less than the words that Obama reads from his teleprompter.

Again, very straightforward questions:

1) Which athletes are being analyzed and for which events and over what period?

2) Who is the data being analyzed? What is the methodology?

3) What is the rational behind the choice of the data set and the methods used to analyze the data?

4) How is the conclusion related to, and supported by, the data?

Taking out the fact that ARRS, a self-appointed authority on what is aided, uses an elevation drop that is FIVE TIMES more generous than what is permitted by the IAAF.

Now, with that said, do they even understand that this is the biggest stinking pile of self-referential, circular reasoning, shit imaginable? Universities around the world should be suing this as a template for the logical fallacy.

How can can anyone take their opinion seriously? It changes the meaning of the 'giggle test' to the 'falling down on the floor laughing so hard my ass has fallen off test.'

The ARRS should simply stick to what they do best: collect results from events around the world. Create updated all-time and record progression lists for catagories like 'World', "Country', "year", and "Event".

That's what they do best.

Its Getting Comical wrote:

It is really getting funny. The flat earthers, oh I mean no winders, are either getting unbelievably desperate, or they've just been replaced by pure trolls, trying to keep a non-debate going.

I can hardly wait to see what they come up with next. Perhaps proof of alien ray guns slowing down these runners, otherwise they would have run 5 minutes faster!

"No-winders" -- I like it. This deserves its own thread.

The Wellesley station you linked to, the guy has stuff about it on his blog. It's kind of cool actually.

http://geekdoctor.blogspot.com/2011/01/weather-station-lessons-learned.html

From what I gather, the readings for temp, humidity and baro pressure should match pretty close to known values (these shouldn't vary by much at all from nearby stations) but for wind readings, there's a lot of latitude granted.

Interesting stuff - I'll have to go look up the weatherstations around my house that my software has been extracting data from into my workout log.

I followed that blog into quality controls and ended up finding this weather station at about the most wide open space on top of a building at Wellesley college just off the course:

Wind is still probably higher than what is seen at ground level but better than the airport (or someone's backyard) - speed from 1-10 mph from about 7-9 o'clock:

Date Time mph direction° o'clock

4/18/2011 10:01 5 267 7.4

4/18/2011 10:11 10 293 8.1

4/18/2011 10:21 6 286 7.9

4/18/2011 10:31 6 292 8.1

4/18/2011 10:41 6 301 8.4

4/18/2011 10:51 4 311 8.6

4/18/2011 11:01 1 266 7.4

4/18/2011 11:11 5 274 7.6

4/18/2011 11:21 6 270 7.5

4/18/2011 11:31 3 311 8.6

4/18/2011 11:41 3 288 8.0

4/18/2011 11:51 3 308 8.6

Why don't you just email Ken Young and find out? This has been going on for several days and no one thought to do that? Really???

Hey, why didn't I think of that? Because Brett from Tokyo represented that he knew Young and that Young would be posting his methodology on this board, that's why...

The whole tailwind thing is overblown.

Fast course + fast runners = fast times.

Since I'd done a split by average elite male and top 60 non-elites - I thought I'd look at one more split of the boston marathon data - the split for the winners for the last 5 years:

Winner Pace (mpm)

k . 2011 2010 2009 2008 2007

5 4:41 4:48 4:42 5:03 5:16

10 4:43 4:55 4:49 4:44 5:14

15 4:44 4:47 4:55 4:45 5:12

20 4:49 4:55 5:02 4:49 5:25

25 4:41 4:46 5:06 4:46 5:02

30 4:53 4:50 4:47 4:46 5:13

35 4:35 4:45 4:58 5:00 4:59

40 4:35 4:44 4:51 5:04 4:46

42.4:34 4:47 5:21 5:10 4:59

Avg4:42 4:48 4:55 4:53 5:07

I have no idea if this adds anything whatsoever to the wind discussion (and I don't really care since the thread title is "Boston analysis") but damn that run in from 30k to the end (including 2 newton hills) looks awesome.

Seems to me the strongest argument for the non-winders is that Hall took the race out under 1:02 at the half. The pack was how strong then? More than 15 guys? In the end, 12th place was over 2:12, meaning very few could handle the pace, even with the wind. I was a happy wind-believer at first, just marveling at a great, but aided, race. Now, Canova has me thinking that it was Hall that was the divine wind, pulling (not pushing) Mutai and Mosop to greatness. As Gebremarium stated, “He pushed it, he pushed it all the time. He was like a pacemaker; he helped us a lot.”

Should it be a WR? No. Was it better than 2:03:59? I'm starting to believe so. We'll see if a mental barrier has gone down, and faster times come quickly. Hall's legacy may not be as a champion, but as a king-maker!

King Maker wrote:

Seems to me the strongest argument for the non-winders is that Hall took the race out under 1:02 at the half. The pack was how strong then? More than 15 guys? In the end, 12th place was over 2:12, meaning very few could handle the pace, even with the wind. I was a happy wind-believer at first, just marveling at a great, but aided, race.

I feel off my chair backwards when I read that. There were 15 guys together at the halfway mark at 1:02 so that makes it proof that the wind didn't help them?

Zero Gravity wrote:

The whole tailwind thing is overblown.

Fast course + fast runners = fast times.

Yup, this course has never really been run before by fast runners.

And it's always been a really fast course.

Now why didn't that occur to me before?

Thank you, Zero - you really put this controversy to rest.

Some people find me funny, sometimes when I'm trying to be, and sometimes when I'm not. But I can not make a living as a comedian.When I said "You look silly ...", I didn't mean "you" specifically, but more like a "royal You", because it seems to me like that's the main controversy. Sorry for causing that confusion -- I'll take full blame for that, but will also accept credit for being a genius.My impression though is that the exact magnitude of the advantage (the only reason to ask for the algorithm) is not "exactly what is being doubted" by the ARRS quotes. Most people seem to be jumping to a conclusion that a claim of "statistically valid" means "lobbying the IAAF for world record status". My point is that by including the times in their database, the ARRS is not unilaterally arbitrarily declaring, nor lobbying the IAAF to accept, the Boston times for world record status.If you want to argue that 1:37 should really be 2:37 or 3:37, I will not stop you. I do not have the ARRS algorithm, just a few hints at what it might be, and some definitions about what it is not. Everyone can keep asking here, but it seems to me that it's not falling on deaf ears so much as falling on no ears. Like everyone else, I also do not have my own personal algorithm to cross-check the results, nor, like most others, any other deep insight, to say that 3:00 is a better estimate than 2:00, for any local wind/weather advantage.

A Better Question wrote:

OK, I'm sorry but I just have to ask - are you a comedian? Seriously.

You write: "You look silly concluding that they think it should be a world record, especially when they never said it"

I will admit this ^ is simply genius. I never concluded that they think... If you think I did, please re-read my post (which you've conveniently copied). So, I look silly concluding something that I never concluded because they never said it?!? Pure genius.

Oh, and how about this: "You don't need the algorithm for most questions. You do need the algorithm if you doubt that the times are 1:37 fast for men, 1:42 fast for women, and 13.5s slow on average.:

Ummm...pretty sure that this is exactly what is being doubted. That this is being doubted hasn't really been kept a secret now has it?

Now, about that algorithm...

rekrunner wrote:

If you want to argue that 1:37 should really be 2:37 or 3:37, I will not stop you. I do not have the ARRS algorithm, just a few hints at what it might be, and some definitions about what it is not. Everyone can keep asking here, but it seems to me that it's not falling on deaf ears so much as falling on no ears. Like everyone else, I also do not have my own personal algorithm to cross-check the results, nor, like most others, any other deep insight, to say that 3:00 is a better estimate than 2:00, for any local wind/weather advantage.

My question is common sense. What possible algorithm could possibly be used to proclaim a wind blown marathon statistically valid(sic) (or not?)

What the hell does statistically valid mean?