The smell of cut grass. Longer evenings. Going for a walk without hat and gloves. Racing in Belgium. Yes, spring is coming to those of us in the northern hemisphere, and with it comes the build-up to Flanders and (with an extra week’s delay this year) Roubaix.
How, then, do the riders build up? Well, let’s assume that our protagonist of choice has got some early miles under his belt, in the Middle East or in Spain. He’s taken a view on whether to ride Opening Weekend or not. He’s raced Strade Bianche and Tirenno-Adriatico, or Paris-Nice. Somehow, he’s avoided injury, illness and fatigue and now simply needs to put the finishing touches on his form. Where should this be done?
Well, yes. But if we think about the complexity of form, the range of options, the different styles of riders, where…?
What follows is a look at the Belgian semi-classics that fill the next few weeks with so much joy. I enjoy Catalunya and Itzulia as much as the next climby stage race fan, but my heart will be in (and around) Flanders. I’ve gone back through these races since 2015 and pulled together a few stats, and a few words on each race. All of which is designed to put these races into their proper place as valuable prizes in their own right, but also as build-up events for the two cobbled monuments, and to consider which race should be targeted if you want glory in the Oudenaarde or on the Roubaix velodrome.
All stats are shown for years 2015 to 2021, each year separated by /
Group size = number of riders given same time as the winner.
Top 10, 20 = gap between winner and 10th, 20th (min.sec)
Ronde, PR winner = position of winner of corresponding Ronde, PR in this race
Winner at Ronde, PR = position of winner of this race in corresponding Ronde, PR (where raced).
Note that due to Covid, E3, Dwars and Paris-Roubaix were not run and entire “spring” schedule was incredibly disrupted. In 2021 Paris-Roubaix was run in October and form and calendar clashes were rife.
Group size = NA/NA/NA/19/26/1/32
Top 10, 20 = 1.20,1.49/1.19,1.53/2.46,6.50/0.00,0.07/0.00,0.00/0.28,3.16/0.00,0.00
Ronde, PR winner = 1,-/-,-/1,-/-,-/-,-/DNF,NA/-,-
Winner at Ronde, PR = 1,10/DNF,113/1,-/-,-/-,-/5,NA/-,-
The first of the upcoming one-day races in Belgium, and the stats come with a massive caveat – this race changed character hugely in 2018 when it became a one-day race. Before that, the three day event saw some tough Belgian cobbled stuff, as well as a challenging individual time trial. It was also a race lots of people left early. So those “ones” early on relate to Alexander Kristoff, and it used to be really valuable as a deep prep race.
More recently we’re seeing a relatively flat race that typically finishes in a group finish. The 2020 version, won solo by Yves Lampaert was an exception, and that was caused by wind. It seems unlikely that we’ll have a wind-blasted race in 2022, and if not this should be a bunch gallop. The likes of Bennett, Groenewegen, Cavendish and Merlier line up.
As a predictor for the monuments this is pretty useless. Enjoy it as a sprint race and hope for echelons. Also, mourn the passing of a three day race for effectively a second Scheldeprijs.
Group size = 1/1/3/1/1/4/NA/1
Top 10, 20 = 0.38,0.38/0.11,4.48/0.52,0.52/0.20,3.19/1.04,2.49/NA/1.30,2.47
Ronde, PR winner = 4,25/2,-/2,1/1,26/4,11/NA/1,-
Winner at Ronde, PR = 14,DNF/27,-/2,1/1,3/36,8/NA/1,68
Now, this is much more like it. The group size and the top 10 and 20 splits look like a monument. You aso see lots of cross-over in winners. It is noteworthy that the links to the Ronde are so much stronger than the links to Paris-Roubaix (of the last six runnings, we’ve seen the Ronde winner win this twice, finish second twice and finish fourth twice). The winner of Roubaix has skipped this twice, and only finished in the top ten once (when van Avermaet won both races).
It is hardly a surprise that this race sets up perfectly as a warm-up for Flanders. The roads are similar, the hills are similar, and many of the same riders go for both. Also, it sits nicely in the calendar, with just nine days (often with a ride done less seriously in the gap) before the main event. The crossover to Roubaix is less than I would have expected.
When Friday comes around you can expect all the cobbled principles to be on the startline. If you want to see your rider thrive in Roubaix this might be too early to shine, but if you want them to win at Flanders, a top four finish seems almost essential. It is time for the cream to rise to the top.
Group size = 1/4/2/19/28/3/7
Top 10, 20 = 6.54,6.54/0.11,0.17/0.06,0.06/0.00,0.05/0.00,0.00/1.40,3.02/1.25,1.57
Ronde, PR winner = 9,DNF/1,-/-,1/39,1/22,DNF/9,NA/-,4
Winner at Ronde, PR = 52,44/1,11/2,1/6,1/11,56/59,NA/6,7
Well, here’s a thing. Roubaix winners have won this twice and finished fourth on another occasion (albeit the fourth came with a six month gap between the races). Ronde winners do less well here – Sagan was the last winner of both, back in 2016 – than at E3, but for Roubaix it is perhaps the other way around. Part of that is timing and form, with Flanders riders perhaps not going eyeballs out a week before the main event, but part of it is parcours. This is a flatter course than E3 and Flanders and relies more on narrow, uneven roads, and on the riders, to gain separation. Of course it isn’t Roubaix but nothing is.
So, your chosen riders should be either skipping this or taking it easy (for Flanders) or thinking about showing power and form (for Roubaix). Too early to say what the startlist will look like at this stage, but expect a mix of some classics riders looking for another pipe-opener and some faster guys who’ll be skipping Flanders.
Dwars door Vlaanderen
Group size = 1/34/1/1/5/NA/1
Top 10, 20 = 1.29,4.46/0.00,0.00/1.03,1.03/0.59,2.42/0.19,0.19/NA/0.26,0.26
Ronde, PR winner = -,-/-,-/2,-/9,-/51,DNF/NA/30,73
Winner at Ronde, PR = 106,DNF/-,-/36,82/29,29/-,-/NA/10,OTL
Dwars suffers from sitting too close in the calendar to Flanders, and here we see a race that (with the exception of a group sprint in 2016) looks like a Flanders finish, but simply doesn’t feature guys who are going to thrive in the monuments. The last meaningful crossovers were top tens for Gilbert and Terpstra, who would win Ronde (in both cases as long attackers from, perhaps, guys who expected to ride in service of others).
Expect to see more lieutenants given their chance to ride for themselves, and don’t be surprised if some big names (van der Poel rides, so does Stuyven, so does a very strong QS lineup) are here for the exercise and not pushing for the win.
Group size = 14/16/13/29/34/21/22
Top 10, 20 = 0.00,0.24/0.00,0.03/0.00,0.05/0.00,0.00/0.00,0.00/0.00,0.00/0.00,0.00
Ronde, PR winner = 1,-/DNF,-/-,-/-,-/-,-/-,NA/-,-
Winner at Ronde, PR = 1,10/-,-/-,-/-,DNF/-,-/-,NA/-,-
This race is here because it happens in Belgium during the buildup to Roubaix. Alexander Kristoff’s peculiar 2015 once again throws the numbers, but essentially we shouldn’t expect anything that happens here to make any difference to the monuments. This will be a sprinter’s day, and if your monument guys are here, just hope they avoid injury.
Group size = 1/1/8/1/1/4/3/2
Top 10, 20 = 0.02,0.19/0.20,0.31/0.06,0.16/0.14,0.19/0.12,0.18/0.13,0.25/0.12,0.53
Ronde, PR winner = -,-/-,DNF/15,-/-,-/6,-/2,NA/-,64
Winner at Ronde, PR = -,-/-,-/10,-/-,-/-,-/DNF,NA/-,-
I’ve thrown in the Brabant arrow not because it historically means much, but because of the calendar jiggery-pokery that puts it before Roubaix. Will anyone use it as a stepping-stone? It seems most unlikely – if anything it is a better guide for Flanders – but if your rider needs a few more kms under the belt, maybe, just maybe, we’ll see a warm-up ride in Brabant in advance of the second monument.
Group size = 2/1/1/1/1/2/2
Top 10, 20 = 0.49,2.34/0.49,1.16/0.53,2.29/0.25,1.13/0.17,1.58/0.08,2.41/0.47,2.15
PR Winner = 7/-/2/6/DNF/NA/55
Winner at PR = 10/11/-/3/-/NA/68
I’ve thrown the monuments in here for two reasons. First, to show the group size and top 10 and 20 figures for comparative purposes, and second, to answer a different question – is Flanders the best warm-up for Roubaix?
Well, it isn’t bad. We have three top ten rides in Flanders for Roubaix winners, and three winners of Flanders who’ve gone on to finish in the top 11 at Roubaix. The last two years don’t really work for comparison, so that is from a list of five. This year, of course, there’s two weeks between the races. I suspect that, if you’re looking for a Roubaix winner, this will be the race to look at for form.
Group size = 6/4/5/2/2/NA/3
Top 10, 20 = 0.31,0.31/2.20,6.18/0.12,0.12/2.23,3.07/0.37,3.07/0.37,1.40/NA/1.16,6.21
Ronde winner = 10/11/-/3/-/NA/68
Winner at Ronde = 7/-/2/6/DNF/NA/55
There you have it, the last of the stats. We haven’t seen a solo rider enter the velodrome since Niki Terpstra in 2014 – but that is a peculiarity of recent editions, with the vast majority of races earlier this century won that way. It is helpful to have a quick finish, sure, but massive power and the ability to win solo is definitely more use.
This is not a race that it is easy to find a winner of from a trend perspective, and it’ll be harder than ever this year with the disruption, but let’s see if the extra recovery allows a Flanders winner to double up for the first time since Cancellara in 2013.
A very rapid look at some numbers, and a lot of very different races under one umbrella. It is great to see the numbers confirming two things that we all thought we knew - E3 is the King of prep races, and Roubaix is too peculiar to fit any statistical parallel. Of course things will be altered this year (and I haven’t mentioned Amstel Gold, over the border in the Netherlands but this year, theoretically, a Roubaix prep race. I haven’t mentioned Route Adelie either, but we have to stop somewhere) and the impact of that is unclear. The links between Roubaix and GW are tentative but intriguing, but this year I expect the links between Roubaix and Flanders to be stronger than ever.
So... Belgian semi-classics. Six races. Six glorious days of racing. Six chances to get ready for the monuments. Who’ve you got, and where do you want ‘em?