What's past is prologue--William Shakespeare, The Tempest

Saturday will see the 101st running of Milan-Sanremo--a handy thing for those who love stats, because it means there have been exactly 100 previous editions to play with, neatly divisible into groups of 10. So, what are the trends? What would the average race winner look like, and who among the favorites fits the profile? Races are, of course, tricksy and unpredictable things, but we've already tried to assess form, previous results, and Magic 8 Ball predictions--hey, why not stats?

A couple of notes:
La Primavera was held in April for the first 20 editions, then switched to March in 1928. The age of winner is given as the difference between the month of race and month of rider's birth, irrespective of the exact day either fell upon. Also, references are almost equally split on whether Ugo Agostoni was born in March or July. I went with March since Google gave it a slight advantage, but the difference in caculating averages is minimal.

For favorites, I'm considering (alphabetically): Daniele Bennati, Edvald Boasson Hagen, Tom Boonen, Fabian Cancellara, Mark Cavendish, Tyler Farrar, Oscar Freire, Philippe Gilbert, Thor Hushovd, Alessandro Petacchi, and Filippo Pozzato--the riders given the best odds by the bookies as of this morning (this post is all about the numbers, so that seemed appropriate).

Okay, let's look at the trends--below the fold, that is.

Continue reading this post »

BotA= Best of the Aughts. I got a little tired writing that out again. 

What's the best season of the Aughts? Who rode it? And if you are screaming Boonen 05! Boonen 05!  BOONEN 05!!!, perhaps you should cool your jets a bit and weigh all the evidence before voting at the bottom of this entry. It's not clear at all which season is #1.

What I have below are nine campaigns, nine different seasons where riders did two numerical-type things:

1) The scored over 2000 points in the Cycling Quotient rankings and,

2) They scored 10 or more Virtual Musette points.

Why these two filters? Simple. Without going into detail (and you can ask in the comments if you dare), CQ scores over 2000 means the riders were awesomely productive. They won or podiumed in a lot of races. There were only 17 such seasons in the Aughts turned in by only 10 riders.

Virtual Musette scores over 10 points in a year define dominance. To do that a rider needs to win a lot. There are only 16 such seasons in the Aughts and only nine riders who rode them. 

Together only seven riders did the Double of both CQ  2000+ and VM 10+ doing it a total of 9 times. I took out one season-Valverde's 2006-because I felt his other two seasons in which he passed my filters were better. (Al is the only rider to have three such 200+/10+ seasons.)

Then I added one more rider who had a high VM score (the highest of the decade) but not a 2000 CQ score. You'll see why I added this rider in my description of that season. So that makes 9 seasons.

 But why nine? Because in Chinese numerology nine is most auspicious, the number of the Emperor. Nine is for longevity. It is also the number of the chapter in the Yellow Emperor's Book of Medicine, the Spiritual Axis, translated usually as "The Beginning and the End" that has guided my acupunk career from day one. And since China is deciding to resume its place as the center of our world after a short 400 year hiatus, we might as well bow down to our Middle Kingdom overlords... 

All righty then! The seasons are described on the flip. Go read them then vote and in the comments place your ranking of all nine.

Continue reading this post »

Continue reading this post »

H/T to Vlaanderen90 for the addition of Heras and Kirchen!

Question:

Who's the best rider of the decade? Lance Armstrong or Eric Zabel? The guy who beat the Tour like a drum or the guy who rode and won almost everything under the sun? Who had a better decade-Carlos Sastre or Levi Leipheimer? In picking your all-decade sprinter team, who do you pick first-Robbie McEwen or Thor Hushovd? Who do you rate higher for the decade-Fabian Cancellara or Oscar Freire? Jan Ullrich or George Hincapie?

Answer:

 Depends on how you are rating them. Honest!

Ranking riders is not that easy. Besides the fact that different riders specialize in different types of races and terrains, one can develop very different types of measuring sticks, some of which favor sprinters, others which favor guys who ride a lot, or who win races, or who podium a lot, or how hit they are, or climbers or doping tainted, or...any number of things. Then there's the time period you are ranking the riders. Some riders reach a certain level and chug along for close to a decade. Others have a much narrower but much higher peak, So which is better? To me the answer is both, depending on what you are looking for. The fact is one can rationalize most anything and ranking riders is just one more of those things. 

What I present for you here though is the way Cycling Quotient ranks riders. Which riders made the largest cumulative mark in the decade? For reference, here is the CQ point scale of races and here is the CQ's breakdown of the UCI race categories. CQ is you probably know, counts every race and fits the results into their rankings. Naturally the big races give out the most points but all races give out some races. There's plusses and minuses to doing it this way but for now let's just let the rankings speak for themselves. See of you agree with them or not: that's most of the fun with these things anyway. Below I have little blurbs on the top riders and a table (Yes! A TABLE!!!!) of the top 33 riders for the decade: every rider that scored more than 6000 CQ points in the Aughts.  Let's start with #1 shall we?

Continue reading this post »

[And yes, this is a VDS matter...]

It was only somewhat surprising that in my chat with Tyler Farrar the other day he didn't know he was fourth in the world in wins. He referred to UCI points, presumably using the World Calendar, which is a nice list of races but kind of skimming off the top. I was referring to CQRanking.com tallies (and editorializing by crossing Jose Rujano off the list), which cover pretty much every race that you can think of, and which count wins as well as points. All are measures of success, but look absolutely nothing alike.

Cycling is forever tinkering with its meta-organization at the top level, looking for ways to measure the performances of riders and teams. The Pro Tour, in its initial form, was one of the most ambitious attempts ever to create a "Major League of Cycling" that fans in places like the US could understand. The concept was to separate, for once, the top level teams from the second tier, pit them against one another all year long over the most famous roads in the sport, and declare various winners. While hobnobbing with the Second Division was not strictly forbidden, and while the Pro Tour did not insist on total control of the startlists in such races, inclusion of non-Pro Tour teams was minimal. This was an exclusive club -- akin to US team sports, where the major and minor leagues never truly mingle, or like football/soccer worldwide, where they mingle occasionally but the dividing lines remain clear.

The effort failed for political reasons that nobody feels like rehashing, but more fundamentally for reasons having to do with the nature of Cycling. Teams build and survive by attracting sponsors, not by selling tickets. So when (inevitably) certain Pro Tour teams wound up having to feign interest in big races that would have mattered a lot more to the Second-Division squads whose space they took, the races and teams objected. Why should the Tour de France host an Italian team that just exhausted itself in the Giro and is sending a bunch of warm bodies to France, when teams like Agritubel or Vacansoleil could bring far more interest to the race and pay a year's worth of bills in the process?

The result is to mostly go back to the way things were. Most of the big races retain some latitude to determine who should be there. Big teams race with their smaller counterparts as often as ever. Pro Tour teams pay extra for a special Seal of UCI Approval and a quasi-guarantee to a slate of races, but the resulting competitions are of questionable value. The Pro Tour calendar is a small subset of the races that matter, and even the broader World Calendar misses out on plenty of important events (Omloop Het Nieuwsblad? Paris-Tours??) while including stuff like the Tour of Poland. In the process, the World Calendar named Astana the #1 team -- a bad joke to anyone following Columbia-HTC (85 wins!) or Saxo Bank. Or anyone who reads the CQRanking, where Astana were 6th in points and 14th in victories, trailing squads like Agritubel (since folded) and the Tabriz Petrochemical Team of Iran.

But if the UCI draws its lines too narrowly, surely CQRanking draws too broadly, using every race under the sun to rank Tabriz ahead of Astana or Rujano ahead of Farrar. Actually, that's just victories, their points system is more just, but IMHO it would be nice to look at points from the "races that matter," and with all due respect to the Vuelta a Venezuela, it just doesn't matter to me.

So how do you figure this? Last year we drew up the Virtual Directeur Sportif calendar, a list of 47 races from around the world, which was slightly expanded from the previous year. We tried to catch all the "races that matter," but inevitably during the year there were races people got excited about which were not on the list. So if you were to create a list of the "races that matter," how would you change our current calendar? In the process of answering this, think about why exactly a race matters. Because the big names are there? Because it's historic? Because the level of competition is usually high?

The purpose of this survey is to assist tinkering with the VDS. We may be able to use winter races somehow, although we will see. Anyway, thanks for your input.

Year's done but we still need to do some final reckoning before we start predicting next year and this post is a first stab at how the teams did. If you want, you can look at the Cycling Quotient team rankings, or the UCI rankings but here I am using the VDS rankings. Why? Because of the three rankings systems our house VDS rankings focus more on the races that we as a group pay attention to than the others.  Don't get me wrong. Those other rankings are good too. But here's where I think those other rankings systems fall short:

- The UCI rankings don't use enough races. For instance of they only use three races from the spring cobbles season: Flanders, Gent-Wevelgem, and Paris-Roubaix. Overall they only rate 24 races. That's just not enough. We as a group are interested in more.

- Cycling quotient uses too many races. Take the Jayco Herald Sun Tour. Or the Azerbaijan Tour. Or the Tour of Milad du Nour-all of which can earn a rider CQ points but are races that we just don't follow much if at all. I like that CQ looks at all of these races; it gives one a sense of what's happening in the regional circuits if one is into that sort of thing. But we tend to stay on the World Tour level here.

Of the two, the CQ rankings is closer to VDS than the UCI is and when you get to the table on the jump I'll have a column for them so you can contrast it to VDS. And do I have a Table for you! Last year in the final rankings we had 37 teams scoring at least 10 points in one VDS race or another. Bah! That's just nuthin'!  This year: 62 teams-and that's not counting those fake national teams at the Rund um Koln. It's the Mother Of All Tables that's awaiting you on the jump.To protect your eyes a tad,  I've put the Table at the end of the post because it's so freaking long. Before the post I have a little blurb on each of the main teams-a first cut at evaluating of the main teams before Chris delves deeper in posts over the winter. Follow along!

Continue reading this post »

Some time ago Ursula, Majope (I think?) and I started tinkering with statistics as a way to tell more of the story, much as is possible to do with other sports like baseball or whatnot. While we have not made much headway with all riders, in my opinion we have identified a decent set of sprinters' stats that do tell what a guy is up to. First, is he making the cut in the sprint stages? Once he gets to the head of the race, does he win? Place consistently? etc.

Given the riveting two-man battle for Green at the Tour, I thought I would run some stats for Cavendish vs. Hushovd. These numbers are taken from grand tours raced between 2007 and this morning. I specifically focused on grand tours, for the purpose of evaluating their chances at the Tour, rather than their overall sprinting abilities. Hushovd generally peaks for the Tour above and beyond anything else, so it's only fair. Check it out:

Right now the numbers say that Hushovd should be getting creamed by Cavendish for green, but that's not the case. A few points:

• Hushovd is outperforming his typical Tour de France from the last few years, placing no worse than 8th in any stage, with an average placing of 3.4. Over the last few years, his raw placement average is 20... but this is skewed by a couple outliers, including at least one relegation. Subtract those and his average placing is more like 10th. However you slice it, the God of Thunder is having a very good Tour, even by his standards.
• The simplest reasons why Cavendish trails Thor are his one day out of points (16th, though with the bunch), and Hushovd's intermediate points (not sure how many, but it's not a ton). For Cav, it's all or nothing -- and almost always all, but that one slip-up did hurt. Cavendish too is having an excellent Tour, but that one poor result demonstrates the immense premium placed on getting in position every day.
• One other reason Cavendish is losing, and this has gotten no play: the lack of bodies in the finale. Over the years Hushovd has only made the top five of Tour sprints 43% of the time, barely more than half the number as Cavendish, yet this year both riders are 4 of 5 in that regard. Why? After all, this is the most star-studded sprint field in years, maybe decades: Boonen, Freire, Bennati, Farrar, Van Avermaet, etc. Using the CQRanking.com head-to-head feature, Boonen beats Hushovd 23-16 in sprints; Freire beats him 21-15; Bennati... loses 16-11. In any event, these guys are more often than not putting themselves between Hushovd and the winner, which explains why Hushovd over the years has frequently found himself out of the top five or even top ten. Yet this year Bennati is absent, Freire is in and out of competition, and Boonen is openly moaning about how he can't win, which usually means he doesn't bother with the sprint. This is hurting Cavendish: his advantage is Thor's weakness, the big argy-bargy, but Thor isn't being hurt by dropping five spots; when not winning he's getting second or fourth. Meanwhile, Thor's advantage and Cav's weakness is the odd intermediate sprint or difficult finish, and Thor has made the most of those.

One note, Bennati did take 9th today, so it's possible he will be heard from as the race goes on. I don't know what Freire's plans are, but one would hope he will try to assert himself eventually. It's possible we will see days where Cavendish wins and Hushovd fails to minimize his losses with a minor placing. For Cavendish, however, this is out of his hands, and is making for some nervous days.

Trying some of the crosswind-conga again today. Nothing doing yet.