Tour de France Average Speeds

[Bumeington]

First up, most threads in here make unsubstantiated claims so here is my source: http://www.bikeraceinfo.com/tdf/tdfstats.html

A quick summary:

1970's: Between 34km/h and 38km/h (Presumably depending on course difficulty)
1980's: Between 35 km/h and 38km/h.
1990: 38.621km/h LeMond wins.
1991: 38.747km/h Indurain wins, LeMond 13th.
1992: 39.504 (First 39+) Indurain wins, LeMond abandons.
1993: 38.71
1994: 37.831
1995: 39.514
1996: 40.697 (First 40+, Rjiis admitted doping to win)
1997: 39.188
1998: 41.765 (First 41+, Festina affair)
1999: 40.277 Armstrong positive for EPO (but shhh)
2000: 39.221
2001: 40.016
2002: 39.982
2003: 40.030
2004: 41.016
2005: 41.654
2006: 40.789
2007: 39.23
2008: 40.50

Points for dicussion:
Does anyone know how the course profiles for recent years would have been likely to effective average speeds?
Does the high value for 1992 signify the arrival of EPO in the peleton, why are the next years back to levels that could have been achieved by LeMond, Roche et al?
Fastest speed to date is 1998, does this mean modern testing prevents athletes from doping to this high a level?
Is it possible to ride a tour at 39km/h? Can we believe Contador's 2007 results, or Sastre's 2008 result in light of the fact they are similar speeds to the Armstrong years?
Considering the rider who average exactly 39km/h in last years tour was Robbie McEwen in 122nd place, is everyone superjacked even to finish?

 

[benpounder]

What is with you posting facts when we all know Lance is to blame for everything...

 

[ElChingon]

Is there a factor for no rest days, multiple transfer days and 30 minute feed zone stops? The numbers will never be equal as we all know the racing has changed in many ways, its like comparing apples and lead bricks. Then there's years where we have stages where the racers do a protest ala stage 9 at the giro this year, how is that time factored in? Yea, that affects the numbers doesn't it. Not to mention memorial stages like in '95 for Fabio's death.

 

[slcbiker]

It doesn't seem to me that we can attribute changes in overall speed to doping. Not to say that doping hasn't been happening, but there's a ton of other potential variables.
Stage differences, weather, etc. -- how do you account for that?

 

[TRDean]

I agree...

that you can't make exacting comparisons in the races, however, there is a definite trend present. The greater N the more confidence you can have regarding the trend. Knowing when EPO hit the scene...it is definitely suspicious. Now don't go jumping all over me...I am just saying that the numbers are interesting. Of course training methods would be more advanced, and over time you will see athletic performance getting better..just look at the 1 mile runner times over the last 30 years or so. Interesting none the less. Thanks for the speed averages!

 

[Bumeington]

Is there a factor for no rest days, multiple transfer days and 30 minute feed zone stops? The numbers will never be equal as we all know the racing has changed in many ways, its like comparing apples and lead bricks. Then there's years where we have stages where the racers do a protest ala stage 9 at the giro this year, how is that time factored in? Yea, that affects the numbers doesn't it. Not to mention memorial stages like in '95 for Fabio's death.

Obviously these things are a factor, and I did not mean to advance any opinion in my posting, just to state some facts and see what people made of it. If you think it's hopeless to try and extract any information (and you do have valid points) then that's great, but it would be nice if you could provide your input without appearing angry/aggressive.

 

[al_pacino]

Hi first post after a little while lurking

Although as others have posted you cant compare certain variables am I the only one surprised at how little diiference there is between the year Lemond won(38.621km/h) and the year he abandons(39.504km/h). Considering this is used to demonstrate when widespread drug use came in I was expecting a far bigger jump. Infact the progression in speed seems to match advances in just about every athletic sport over that time.

 

[knorensnor]

the only reasons are the different profiles, the height difference, length, better bikes, better surgery, technology...a few riders taking epo in the late 90's don't make a difference of 2km/h. remember in the 70's and even the 80's the stages were much longer.

 

[Bumeington]

Hi first post after a little while lurking

Although as others have posted you cant compare certain variables am I the only one surprised at how little diiference there is between the year Lemond won(38.621km/h) and the year he abandons(39.504km/h). Considering this is used to demonstrate when widespread drug use came in I was expecting a far bigger jump. Infact the progression in speed seems to match advances in just about every athletic sport over that time.

Since a typical tour is say 3600km long, we can attribute a difference in speed of 1km/h to a 2 hour time gap in the overall classification.

 

[ElChingon]

but it would be nice if you could provide your input without appearing angry/aggressive.

Dude, ?'s are Questions and this is angry. I'm not bashing the thread just adding questions to ponder in the thread. Don't get bent if the thread does not go your way.

 

[BroDeal]

I think average speeds are near worthless. The courses are different. The weather is different. The wind is different. The equipment is slightly different. The style of racing may be different.

I wish we could get a good compilation of climbing speeds for mountains other than the usual Alpe d'Huez. Time trial speeds for the top ten riders over the years would be very interesting if the numbers could be normalized to account for advances in aerodynamics.

 

[elapid]

I posted a paper on TdF physiology in another thread which answers some of these questions.

Since a typical tour is say 3600km long, we can attribute a difference in speed of 1km/h to a 2 hour time gap in the overall classification.

The average speeds and time differences between first and last are:
1903-1909: 26.9 km/h; 84 hr
1910-1926: 25.6 km/h; 54 hr (Pyrenees and Alps introduced)
1927-1939: 29.8 km/h; 12 hr (three week race duration introduced)
1947-1965: 34.7 km/h; 5 hr
1966-1984: 35.8 km/h; 3 hr (first antidoping control 1966)
1985-2002: 38.8 km/h; 4 hr (introduction of modern cycling equipment)

While the authors talk about clip-on pedals, aerodynamic wheels, lighter bikes, etc making a difference, a 3 km/h difference in average speed is a big jump when you look at the average speeds for the other periods. Modern cycling equipment would have obviously contributed, but I am convinced that doping also contributed.

I think average speeds are near worthless. The courses are different. The weather is different. The wind is different. The equipment is slightly different. The style of racing may be different.

I wish we could get a good compilation of climbing speeds for mountains other than the usual Alpe d'Huez. Time trial speeds for the top ten riders over the years would be very interesting if the numbers could be normalized to account for advances in aerodynamics.

Again from the aforementioned paper:

 

[knorensnor]

true, but when you watch the best time on the Alpe d'Huez, from marco pantani, you just know it was realised with dope...it's still a minute or something under Armstrongs time trial on the Alpe...it's sad but I just don't believe that was ridden clean

 

[elapid]

true, but when you watch the best time on the Alpe d'Huez, from marco pantani, you just know it was realised with dope...it's still a minute or something under Armstrongs time trial on the Alpe...it's sad but I just don't believe that was ridden clean

As someone on the TdF physiology thread pointed out, look at Bahamontes estimated 6.1 W/kg in 1958. Also, what surprised me about the climbing speeds was the minimal difference over the years, not accounting for all the variables of course, with 20.0 km/h in 1905 to 22.0 km/h in 1997-2001.

 

[BigBoat]

The speed increased and I think that was in part due to shortening of the route and the fact that all riders now use deep dish bling bling rims on flat stages. Granted in 2005, the riders all had a tailwind for the first weeks flat stages into Germany. It was also flat. LOL

The Sh$t really hits the fan though on long climbs AND >> the TTT and ITTs. Thats where the true power of the autologous blood doped riders will show. Alp D'Huez times have been in the upper 30s and stayed there through the new "clean" era. LOL! Fingon in 1989 rode 41"50 and Lemond 42:08.

Past freaks like Lemond and Fingon got 6th and 7th respectively in the first epo Tour without epo and this was probably the best performance ever with a low crit. Lemond was a freak and he'd dominated the Tour the year before. Now in later years he DNFd and I think this was basically just because everybody was on epo AND people like Ferrari had come in and made the races that much harder. If 50 guys in the race have their FTPs (all out 60 minute power) where your 5 minute repeats are there's no way in hell you would hang on with long climbs. You'd be toast in a heartbeat.

 

[lookkg386]

Was watching some old cyling videos (yes video!) from the 80's last night and the one thing that looked so different to me was the state of the roads! Surely the quality of the roads they are riding on must make a huge difference to speed?

 

[BigBoat]

As someone on the TdF physiology thread pointed out, look at Bahamontes estimated 6.1 W/kg in 1958. Also, what surprised me about the climbing speeds was the minimal difference over the years, not accounting for all the variables of course, with 20.0 km/h in 1905 to 22.0 km/h in 1997-2001.

Bro the VAM's are an "estimate" and not his true power. We dont really know what it was because he didnt have a watt measuring devise on his bike in 1958. If he had his FTP over 6 in 1958 he'd have won more races than Eddie Merckx bro.... There's no way he really had that much power. LOL, you could win the Tour on 5.7 or so back then.

6 watts per kilo is a lot, just to give an example : Riccardo Ricco during the TDF 2008 had about this much power (6.1) right before he was busted for epo (blood doping and stimulating rectics to keep his passport normal.)

 

[BroDeal]

Bro the VAM's are an "estimate" and not his true power. We dont really know what it was because he didnt have a watt measuring devise on his bike in 1958.

That it is an estimate does not matter as long as the same method of estimation was done with all the riders. Using a good model and estimating the average over a long climb should produce average results that are comparable between riders. It would be good if the model adjusted for estimated bike weights in different decades, for which good data exists, and rolling resistance, which should also be possible. That you cannot produce results with pinpoint accuracy is not relevant if you can produce ballpark figures that illustrate differences.

 

[benpounder]

I think average speeds are near worthless. The courses are different. The weather is different. The wind is different. The equipment is slightly different. The style of racing may be different.

And yet Lance Armstrong's times up L'Alpe have been one of the strongest testimony that he doped.

How long was it till Greg Lemond lost the title of world's fastest time trial? And what was that speed? And how different was that record setting speed pace from the proven dopers of yesteryear and today?

Nevermind...
we love Greg...
...and hate Lance.

 

[Escarabajo]

Hi first post after a little while lurking

Although as others have posted you cant compare certain variables am I the only one surprised at how little diiference there is between the year Lemond won(38.621km/h) and the year he abandons(39.504km/h). Considering this is used to demonstrate when widespread drug use came in I was expecting a far bigger jump. Infact the progression in speed seems to match advances in just about every athletic sport over that time.

The differences in doping can be seen in the TT and the climbs. So mathematically speaking when you take the downhills and the flat stages it averages out to be close. The higher speeds and distances will skew the average masking the true climb speeds. Now there is also differences in wind speed, wind direction and kilometers of mountain to climb. So it is pretty much worthless to compare these numbers.
Thanks.

 

[WildspokeJoe]

1986 was a significant year.

That was the last year the tour was 2500 miles. After that, it was about 500 miles less. Therefore the speed is going to go up.

Also it was the first year a non European won the race.

Before the late 80s the majority of the rides were from the traditional cycling powers. Now only about 25% of the riders who compete are French. Back then it was closer to half the field.

Also for better or worse with the end of communism the pool of riders increased. The quality of the fields went up.

Also the field got bigger too. It was from 180 riders to over 200.

Time trial equipment got better - tri bars and areo helmets

Technology got better ( CLIPLESS Pedals, race radios, Dura-Ace eight speed carbon fiber bikes and wheels)

And let's not forget about wind tunnel testing

Training got better too. Creating a template to follow each year. (recon of the course)

Nutrition got better. Power gels, energy drinks

Selection got better. Riders used to race everything. Convicts of the road Now they pick their battles more carefully.

The stages are also faster because of TV coverage

Someone said it earlier the roads got better too.

Accommodations got better too. (A better night's sleep)

The mentality of the athlete changed as well. In the 80s and earlier riders used to race to gain fitness. Now they use the off season to build up their base.

Also riders began to get their pay based on their UCI rank. So everyone fights harder for stages wins and recognition.

And the drug got better too but I think it's obvious that drugs weren't the only reason average speeds went up.

If we're basing doping on speed, then isn't obvious who doped?

Greg LeMond held the record for the fastest time trial for OVER 15 years.

 

[lookkg386]

1986 was a significant year.

That was the last year the tour was 2500 miles. After that, it was about 500 miles less. Therefore the speed is going to go up.

Also it was the first year a non European won the race.

Before the late 80s the majority of the rides were from the traditional cycling powers. Now only about 25% of the riders who compete are French. Back then it was closer to half the field.

Also for better or worse with the end of communism the pool of riders increased. The quality of the fields went up.

Also the field got bigger too. It was from 180 riders to over 200.

Time trial equipment got better - tri bars and areo helmets

Technology got better ( CLIPLESS Pedals, race radios, Dura-Ace eight speed carbon fiber bikes and wheels)

And let's not forget about wind tunnel testing

Training got better too. Creating a template to follow each year. (recon of the course)

Nutrition got better. Power gels, energy drinks

Selection got better. Riders used to race everything. Convicts of the road Now they pick their battles more carefully.

The stages are also faster because of TV coverage

Someone said it earlier the roads got better too.

Accommodations got better too. (A better night's sleep)

The mentality of the athlete changed as well. In the 80s and earlier riders used to race to gain fitness. Now they use the off season to build up their base.

Also riders began to get their pay based on their UCI rank. So everyone fights harder for stages wins and recognition.

And the drug got better too but I think it's obvious that drugs weren't the only reason average speeds went up.

If we're basing doping on speed, then isn't obvious who doped?

Greg LeMond held the record for the fastest time trial for OVER 15 years.

Amen brother, all good points, and the fact is human achievements in all sports have been constantly improving, to say simply it's due to doping is a bit broad of a statement.

 

[nobody]

Greg LeMond held the record for the fastest time trial for OVER 15 years.

NO.
LeMond's record is done because it was a one-way downhill ITT. All riders that day set up a fastest time trial.
In the same conditions 99% of todays riders will break that record easily!

That is like hour world record, with EPO and blood doping all riders can beat Merckx record!

The 1984 Carl Lewis will have still a good ranking if he were competing todays.
Where would be a Hinault, Lemond or Fignon?

 

[Eldrack]

The 1984 Carl Lewis will have still a good ranking if he were competing todays. Where would be a Hinault, Lemond or Fignon?

Well if they where racing in this era they'd be on just as much gear as everyone else. Hinault and Lemond would destroy the field in a lot of races being able to win classics or Grand Tours depending on what they focussed their seasons on. Don't rate Fignon quite as much but he'd be a consistent contender in the GTs.

The riders of any era where on what drugs they could find at the time. Take Coppi and stick him on EPO/blood doping/HGH and he would destroy Pantani's record on Alpe d'Huez.

The effect of doping on average speeds is best seen on the climbs, and in the top speeds that the peloton reaches on the flats in the final 10ks, and the average speed of breakaways. The overall increase in average speed isn't affected a huge amount by these things (since a huge number of kilometers are raced on the flat at tempo type speeds), here the shorter distances and increased television coverage are probably bigger factors.

 

[Eldrack]

NO.
LeMond's record is done because it was a one-way downhill ITT. All riders that day set up a fastest time trial.
In the same conditions 99% of todays riders will break that record easily!

Check out the final TT course in the 2008 Giro. Very similar profile, i.e short and downhill. Much better equipment, same last day of the race type thing. Average speed 2.3kph slower than LeMond.