Re: LeMond / Shermer / freakonomics

Dans le message de news:[email hidden],
Robert Chung <[email hidden]> a réfléchi, et puis a déclaré :

Quoted message said:

ronaldo_jeremiah said:

What the Numbers Can Reveal
a book review by Michael Shermer

In a quarter-century of serious cycling I have heard all the rumors
about performance enhancing drugs, from stimulants in the 1970s, to
steroids in the 1980s, to blood boosters like Erythropoietin (EPO) in
the 1990s. A friend who knows my penchant for exposing fraud
suggested I track the winners' speeds from the Tour de France to
note the increase after 1991, the year he says EPO was introduced.
Since my friend won the Tour three times, I figured it was worth
checking.

http://anonymous.coward.free.fr/rbr/tdf.png

I actually had instant recall of this graph, especially where the answer is
obvious.

I wonder how today's stars could cope with 500-600 km stages ?
--
Bonne route,

Sandy
Verneuil-sur-Seine FR

Sandy said:

I actually had instant recall of this graph

Quoted message said:

I wonder how today's stars could cope with 500-600 km stages ?

Hmmm. People who have instant recall of my graphs have deeper things to
worry about.

Sandy said:

I actually had instant recall of this graph, especially where the answer
is obvious.

Whent he whole story is known, it would be interesting to plot events in
blood manipulation against the data, especially for the period 1989-1996.

What is the 'obvious answer' you refer to?

-RJ

"Robert Chung" <[email hidden]> wrote in message
news:[email hidden]...

Quoted message said:

ronaldo_jeremiah said:

What the Numbers Can Reveal
a book review by Michael Shermer

In a quarter-century of serious cycling I have heard all the rumors
about performance enhancing drugs, from stimulants in the 1970s, to
steroids in the 1980s, to blood boosters like Erythropoietin (EPO) in
the 1990s. A friend who knows my penchant for exposing fraud suggested
I track the winners' speeds from the Tour de France to note the
increase after 1991, the year he says EPO was introduced. Since my
friend won the Tour three times, I figured it was worth checking.

http://anonymous.coward.free.fr/rbr/tdf.png

I think that team tactics are what caused the large increase in average
speed and that the shorter stages amplified it.

In the 70's Eddy Merckx could afford to buy the best team and he ground
everyone else into the dust. In the 80's La Vie Claire was the team to beat
and it took bad luck for them to lose. Frankly, there's no other Tour team
around like Postal/Discovery and it shouldn't come as a giant surprise that
they provide the winner.

LeMond's famous 1989 win led to a HUGE increase in the top salaries and with
the advent of Big Mig the teams started to make more as well. That meant
that there was a LOT more to lose if you lost your place on the team.

The question is: Has there been a significant increase in average speed in
the last 10 years? Well, according to Robert's charts, there was a much
larger increase in average speed between the 1947 and the 1956 Tours than
there were between the 1989 Tour of LeMond's and the fastest Tour on record
(2003).

Are there DRUGS in the peloton? I'm convinced there are. I'm NOT convinced
that they make much difference nor that they are all that wide spread since
other methods such as altitude tents, altitude training and maybe those
cherry flavored tootsie pops seem to work.

Tom Kunich said:

Quoted message said:

http://anonymous.coward.free.fr/rbr/tdf.png

there was a much larger increase in average speed between
the 1947 and the 1956 Tours than [snip]

The cool thing about the graph is that it shows how much cherry-picking
you had to do to make that comparison.

Tom Kunich said:

The question is: Has there been a significant increase in average speed in
the last 10 years? Well, according to Robert's charts, there was a much
larger increase in average speed between the 1947 and the 1956 Tours than
there were between the 1989 Tour of LeMond's and the fastest Tour on record
(2003).

Check for the effects of phasing out rationing and rebuilding of roads
in Europe as well as population re-establishment after WW2...

RC -

What is the correlation between year and average speed if you partial
out distance?

-RJ

ronaldo_jeremiah said:

RC -

What is the correlation between year and average speed if you partial
out distance?

http://anonymous.coward.free.fr/tdf.csv

Hint: partials are easier to calculate (it's 0.6ish, btw) than they are to
interpret.

Ooops. Sorry. That should be
http://anonymous.coward.free.fr/rbr/tdf.csv

Indeed - what is your interpretation?

-RJ

"Stewart Fleming" <[email hidden]> wrote in message
news:[email hidden]...

Quoted message said:

Tom Kunich said:

The question is: Has there been a significant increase in average speed
in the last 10 years? Well, according to Robert's charts, there was a
much larger increase in average speed between the 1947 and the 1956 Tours
than there were between the 1989 Tour of LeMond's and the fastest Tour on
record (2003).

Check for the effects of phasing out rationing and rebuilding of roads in
Europe as well as population re-establishment after WW2...

What difference does that make? The important factor is that you can pick
several 10 year periods and see a larger difference in speed than Lemond and
Armstrong.

Dans le message de news:[email hidden],
Robert Chung <[email hidden]> a réfléchi, et puis a déclaré :

Quoted message said:

ronaldo_jeremiah said:

Indeed - what is your interpretation?

Rather than look at partial correlations, look at the residuals of
speed on distance and think about the things that influence that
pattern.

I wonder about the climbing stats, and whether they would influence the
results.
--
Sandy
Verneuil-sur-Seine FR

*******

La vie, c'est comme une bicyclette,
il faut avancer pour ne pas perdre l'équilibre.
-- Einstein, A.

Quoted message said:

http://www.ucolick.org/~bjw/misc/rbr/tdf.year_speedresid.names.png

Quoted message said:

Or, the time variation could be mostly down to changes in the way
the Societe de TdF designs the course, with some additional effects
of evolution in national/trade teams and team strategies.
That seems like at least as plausible from the data as le dopage.
Course changes, for example, could explain the time variation: from
the data it's just as easy to call the 1970-1990 era abnormally slow
as it is to call the post-1990 era unnaturally fast.

Yeah. That was the same conclusion I'd come to, and it seems pretty robust
no matter how you specify the model. That's why I wonder about
characteristics of the course over time.

Nice graphs. Easier to see a pattern from graphs than from partial
correlations.

Robert Chung said:

Sandy said:

I actually had instant recall of this graph

Quoted message said:

I wonder how today's stars could cope with 500-600 km stages ?

Hmmm. People who have instant recall of my graphs have deeper things

to

Quoted message said:

worry about.

Do you have a graph of average speed against total climbing height I
would imagine that would fit fairly well.