TdF analysis - true effects of doping?

[ScienceIsCool]

Hmmm. This has got me thinking about races that have *not* been affected by these factors. I'm guessing that would be one day classics, where there is no such thing as a rest day.

Looking at the Paris Roubaix; even with minor course changes and huge changes in technology and doping, the average speed has been steady since the 1950's. Only two of the top 7 fastest times come in the era of oxygen vector PEDs.

A host of potential hypotheses are:

- Oxygen vector PEDs have no affect on max sustainable output of a 5 hour interval
- One day classics are not susceptible to the recovery effects of oxygen vector PEDs
- Modern equipment, diet, and training techniques have no affect on average racing speed in this type of race environment
- External influences (race radios, sponsor demands, TV coverage, etc) have had a larger affect than PEDs on racing outcomes (TdF speeds)
- corollary to above: riders use PEDs because the overall demands are higher, and cannot sustain one day classic efforts over three weeks and/or all season

I'll check on other one day classics to see if the trend in the data holds.

I have to say that a lot of my preconceptions are being blown away. But the data is what it is. You can't choose your facts to fit your hypothesis...

John Swanson

 

[Captain_Cavman]

One thing to consider is the make up of teams in 1980. Over 75% of the riders were from France or Belgium, only 11 countries were represented and only 4 had more than 4 riders. There were only 130 at the start.

One would have strong grounds for concluding that the TdF field was a lot weaker in 1980 than in recent times.

 

[ScienceIsCool]

Milan San Remo data is similar to the data for Paris Roubaix (source: http://bikeraceinfo.com/classics/Milan-San Remo/milan-san-remo-index.html)

For example, Eddy Merckx managed an average of 44.8 km/hr in 1967. Cancellara managed 41.1 km/hr in 2008. Erik Zabel, 40.9 km/hr in 2000. Saronni 41.2 km/hr in 1983.

The more enterprising can put the data into sets and run a t-test or similar, but the year to year variation is clearly much, much larger than any shift in the mean.

John Swanson

 

[Dr. Maserati]

Milan San Remo data is similar to the data for Paris Roubaix (source: http://bikeraceinfo.com/classics/Milan-San Remo/milan-san-remo-index.html)

For example, Eddy Merckx managed an average of 44.8 km/hr in 1967. Cancellara managed 41.1 km/hr in 2008. Erik Zabel, 40.9 km/hr in 2000. Saronni 41.2 km/hr in 1983.

The more enterprising can put the data into sets and run a t-test or similar, but the year to year variation is clearly much, much larger than any shift in the mean.

John Swanson

When you mentioned Paris Roubaix earlier I was think, "way too many varibales there & MSR might be a better example".
But again in a race like MSR of that length and duration the difference between a headwind and a tailwind can be the difference of half an hour.

Also, in one day events there could be a massive variation in numbers finishing which would have an impact on any figures.

While the 50% figure you posted at the beginning is an amazing (actually shocking) piece of information I think it would be very difficult to extrapolate any more from many of the figures

 

[poupou]

I remember stages that were not "raced" because Hinault as patron has said so. Those kind of stages were common until TV broadcasting flatland stages.
So riders were less tired.
When Comlombian riders became a threat for maillot jaune, Hinault decided to exhaust them before mountain by pushing hard on flat stages.

 

[Indurain]

The shocking result is that between 1980 and 2000 the average speed on a climbing stage went from ~31.5 km/hr to ~33 km/hr with variations as expected due to length and amount of climbing.

I found that a bit shocking! What this indicated is that the change in eras resulted in less than 5% difference! Within statistical variation, this would be unlikely to completely disrupt the sport. And certainly doesn't account for the overall average change in TdF speed.

I would think a 1.5 to 2km'h increase in climbing speed is very significant. If you looked at how much extra power you would need to climb this much faster, I think you would find that quite significant. You are looking at km'h, which I agree may not sound significant. But if you think of the time it takes to climb 1.5 to 2km it is close to 5mins. Also because the stages take so much longer, if you look at them as a percentage of the overall time, I think you will see that they are also quite significant in keeping the avg's higher or lower. One extra mountain stage can make a big difference in the overall average speed.