The methodology is also still unfinished. I’m sure I’ll be fiddling with it a lot over the course of the season, but I’m trying to commit to publish these for as many World Cup events as I can. It’s sort of reliant on my having a reliable start list, so if I’m busy or traveling the day before a race, I may not get to it in time, we’ll see.

For the moment all I’m going to say is that I’m publishing predicted finishing order for the entire start list. The “Predicted PR” column is the actual value the model outputs, so it gives some relative sense of certainty. Values that are further apart represent more certainty of the gap in finishing place. But they are not probabilities and do not have any particular units.

Also, in case it wasn’t clear, I’m not terribly interested in participating in Twitter anymore, so anything I produce this season will be going here.

Here’s a plot of this measure for major international distance events going back to the early 90’s.

I want to draw your attention to the separate y-axis scales in each of the panels. This is important to keep in mind, to remember that despite the appearances of how “bouncy” the lines are, the variability in the Top 30 panel is much, much smaller.

The variability of the diversity for podium finishers (Top 3) is quite large, which we’d expect since it will be very responsive to dominant seasons by 1-2 individuals, like, say, Marit Bjoergen. The confirmation a lot of fans today should note is the general steady decline in diversity among the men since 2008 or so. The women’s diversity at this level was also the highest in the 00’s and then fell during the heyday of Bjoergen and Kowalczyk but has been climbing back up lately. A similar trend can be spotted at the Top 10 and (kind of) Top 30 levels, although the changes are happening at much smaller scales (again, keep the y-axis scales in mind).

Here’s the same plot for sprints, going back to 2002 or so. This time I’m using the Top 12, to generally align with the semifinals.

Again, keep in mind the different scales on the y-axis. The Top 30 level is bouncing around much, much less than the others, despite how jagged the graph looks.

Not as much of an obvious overall trend here, except that the 2015 season seems to have marked a fairly dramatic transition from a period of relatively high nationality diversity in sprints to one dominated by Norway & Sweden that we’re only sort of coming out of lately.

I posted a couple of graphs on Twitter showing how extreme the field separation was in this race compared to either just skiathlons or all skiathlons and other long mass start races as shown below:

Someone asked if I could do the same thing but include all events, even shorter races and interval starts. I think the reason I restricted myself to skiathlons and mass starts initially is pretty self explanatory, but just for fun let’s look at all race types. I’m still going to exclude pursuit starts because, well, frankly at this point I hardly even consider them real races (). And this includes all major international events stretching back to the late 70’s. The ones that I have, at least. My data from the 70’s and 80’s is incomplete; I probably only have around 60-70% of the races from that era. Anyway, here you go:

This one is a bit too busy to annotate cleanly on the graph, so here’s a list of the 30 most extreme cases for easier browsing. Be sure to sort on the percent difference from 10th place column first.

Anyhoo, Devon spent some time talking about how rare it is for sprint podium finishers to have come from the later quarterfinals (i.e. the top finishers tend to come from quarterfinals 1 & 2). As you can see below, that is obviously true, although as he also alluded to this is a textbook correlation-causation gotcha.

The best skiers tend to qualify well, so they tend to have better quarterfinal choice options, so they will tend to pick the option that gives them the most rest between the semifinal and the final. One interesting thing, though, is how much more severe the pattern is for the men than the women.

A more sophisticated analysis would attempt to somehow control for skier ability in the above graph, although that gets complicated pretty quickly. Perhaps I’ll revisit that later when I have more time.

(And yes, people who follow the World Cup closely will know that the bar for 7th place finishers in a sprint final in the women’s panels is not a mistake. It would have been really funny if it had been Sophie joining him on this episode rather than Sadie.)

These early season races obviously should be taken with a grain of salt. The fields can be uneven, and many of the athletes are not necessarily in top form, or may be using them more as training as anything else. But this race happens to offer some convenient potential for comparisons because with Johaug in the race your percent back from her is pretty likely a decent preview of your percent back in a genuine World Cup.

The split times don’t show anything too surprising. Johaug pulled away early and the top 4 all maintained their positions fairly consistently through the race. Laukli started somewhat conservatively and moved up dramatically during the first half and then held on during the second lap. (Relative to Johaug, at least.)

Next, we can look at a lap pacing plot.

This tells roughly the same story, but maybe emphasizes a bit more that Laukli really did seem to accelerate through the first lap and then hold that speed quite well for the second lap. Laukli’s Instagram post on the race suggests that she caught a ride from Heidi Weng for a good chunk of the first lap (second lap, I was wrong) (and looked really good going through the 2.5k marker, frankly).

So what does 6.4% behind Johaug in an interval start race translate into, roughly? Well, this wouldn’t be cross-country ski racing if there weren’t a healthy bit of variability. In last years major international interval start races that would have put Laukli anywhere from 8th (in the WSC 10k freestyle, no less!), to 18th, 20th, 26th or 29th. That’s just based on where 6.4% back would have put her last season in the available interval start races. That WSC race seems like a bit of an outlier, so high-teens to low-20’s is probably the safe estimate.

One of the interesting things for American fans though, is how young Laukli is. If we take US women’s results in interval start major international races since 1992, tack on Laukli’s Beitostolen race (the assumption being that Johaug provides a decent yardstick), and plot their percent back versus age we get this:

That’s pretty solidly in the “exciting” quadrant of the graph. Zooming in a bit, we get this:

Again, this is all an enormous amount of analysis for one pre-Thanksgiving result. But what are pre-Thanksgiving races for if not for generating either irrational exuberance or premature panic? Isn’t that their whole purpose?

During most of my lifetime, the US & Canadian men’s 4x10km relay teams have not often been in medal contention, although the Canadians definitely had some strong teams for a period with Devon Kershaw, Alex Harvey, Ivan Babikov and others that were occasional outside contenders for a medal. (For instance, I believe the Canadians were 3rd in one WC sprint and 5th in another in the last 10 years or so.) But frequently both North American men’s relay teams have been out of medal contention fairly early. Naturally, you would expect a friendly competitive spirit to develop between the two neighboring nations as a sort of “race within the race”.

I’ve been doing some long overdue organizational work with the several boxes of paper race results records that I’ve received (primarily from Ruff Patterson & John Estle). In the process I (re)-discovered an entertaining little nugget: a telex (!) sent to the US from Europe summarizing the men’s & women’s relays at the 1985 World University Games in Nevegal, Italy. I haven’t done the research to determine for sure who sent the telex and who specifically the recipient was (presumably either Ruff Patterson, or US Ski Team staff generally) but it tells quite an entertaining story.

Here’s the scene. The US men’s relay team consisted of Joe Galanes, Terry Daley, Josh Thompson (future biathlete) and Todd Boonstra, in that order. The Canadians ran Allain Masson, Wayne Dustin, Owen Spence and Benoit Letourneau. Before we even get to the North American “race within a race” on the men’s side, there was some notable drama in the women’s race. What follows is a lightly edited transcript of the start of the telex:

Team nordic Sovi-jet (sic) women nailed for skating in tag zone, cost ’em gold. No impact on US. For (sic) good legs from men and we’d have bronze except the Canucks blew in and took it with USA fourth.

Hotline: ‘They upheld a protest against skating in the tag zone of a cross-country ski relay today…and it cost the Soviet Union a gold.

Three-by-five-kilometer relay at the World University Games in Nevegal, Italy…But the five-member jury disqualified the Soviets for skating in the exchange zone between laps. The decision was the first time skating protests have been upheld in a major international race…coming on the heels of disallowed protests last month at the Nordic World Championships in Austria and last week at the World Junior Championships in Switzerland, by knocking out the Soviet women, the jury handed the relay gold medal to Czecoslovakia (sic)…with Poland the silver medalist and Finland third.

Exciting! The telex continues in depth on how the men’s relay played out:

In the men’s relay…the Soviets tooch (sic) charge on the final lap and earned the gold with Italy second and Canada holding off the lead from–of all teams Japan early on the second lap while the Soviets were struggling. However, Vladimir Nikitin took over at the start of the final 10-kilometer lap and pulled the Soviets in firts (sic), Todd Boonstra kept the US close on the last lap but it wans’t (sic) enough as the US had to settle for fourth in the 11-team field.

This section of the telex (the “Hotline”) reads like a press-release of sorts. What follows appears to be more of a direct communication between ski team staff describing the race in more detail:

Bat fecal matter (sic), got the okay for Joe to race and he was accredited without problem…but who wud (sic) have figured the Canadians wud (sic) replace Austrians and Finns? We had figured Soviets and ITA as 1-2 and that gave us a shot at the bronze we got four excellent legs…but the Canadians upset everything. Galanes was six secs out of first at about 3.5km, sitting in sisth (sic) place at back of second three-man knot when his pole got caught in netting along track on a lefthand curve. He lost about 20 secs but regained it and was in fourth before running out of gas at end of lap. Still we were close to third, which was what we wanted, Josh and Terry gave us strong middle laps we needed, but the Canadians wudn’t (sic) drop. They went with speed to hang-on at the end and pulled it off. Masson had ’em third behind JPN and SOV, just 25 secs out of first…Dustin took lead and they held it till end of third leg when Soviets regained it, and Nikitin cruised. Boonie cudn’t (sic) catch junior Benoit Letourneau altho (sic) he cut 11 secs off the gap…women never in it with Butts skiing scramble and seventh (last) right from the start. When the Sovi-jets (sic) were dumped, that moved US to sixth.

No, I do not understand the “bat fecal matter” reference either. It continues…

Taylor: ‘We wanted four good legs and we got ’em but there was no way we could anticipate the Canadians being so tough. Dustin is skiing very hot right now and is the classic picture of someone getting better as his confidence grows. Getting Joe last night was a shot-in-the-arm and obviously was the key to us being any kind of a contender, who knows what wud hv (sic) happened if he hadn’t gotten hung-up in the nets but he gutted it out and made good comeback. Terry and Josh really showed something and I’m going to talk with Terry about his thoughts on skiing for us, he’s raw talent and wud (sic) be a terrific addition if he decided he’d put his career on hold. Boonie skied a honey of a race, too, but he’s not the Todd Boonstra we’ve seen in the past, still, absolutely no complaints–except someone forgot to tell the Canadians what we planning (sic)’

The US would have won the bronze if it hadn’t been for those meddling Canadians! After all that, one of the things that actually fascinates me the most is that after mentioning the surprise position of the Japanese team early in the race, the telex fails to mention that they were disqualified! Here’s images of the actual race results:

I can’t find an Article 382.5 in the current FIS rules, so I’m not sure what the Japanese team was disqualified for. But I guess that was overshadowed by the drama with the Soviet women skating and the battle royale between the US & Canadian men.

Therese Johaug once again won a race by a very large margin, this time in the 30km classic mass start in the 2021 Oberstdorf WSCs. Personally, I don’t think that I’m as impressed by huge mass start winning margins as other people seem to be. Obviously, unlike interval start races the visual impact of large gaps in mass starts can make quite an impression. But for me, at least, the relationship between my amazement and the size of the gap is not strictly linear. It’s probably something more like this:

When a skier pulls away in a mass start race, I do become steadily more amazed as the gap grows, up until it reaches a certain point and then my amazement sort of plateaus, and even declines a bit after a while. (Obviously, the specific units on the x-axis of this Very Serious and Rigorous graph would change for races of different lengths.)

Which is all just to say that I found Johaug skiing away from the field extremely amazing, up until the gap started getting up into the 90 second territory. After that, I’m not sure any further increases in the margin mean all that much. But that’s just me! You can obviously take joy and amazement from whatever you like.

A consequence of this is that I don’t think you can meaningfully compare winning margins from mass starts except at the level of breaking it down to basically three categories:

• It was a sprint finish
• They were in contact but not enough for a sprint
• There was a very big gap, it wasn’t close at all

For me, mass start winning margins fall into one of those three categories, and I personally don’t get much out of any further parsing of the time gap. But what would I even be here for if I didn’t give you a list? So here’s a list of the biggest (by percent back) winning margins in major international events in the last 30 years or so.

So perusing this table you can see why Johaug’s 10km freestyle interval start gap is still the most impressive thing I’ve seen recently.

Johaug’s winning margin isn’t the absolute largest on this list, only because of the aforementioned women’s 15km championship events, but if you sort by percent difference, hers is the largest by a fair bit.

This sort of situation is what I consider the ideal use of data in skiing. Not some big fancy analysis, just a simple gut check with a quick graph. The metaphor I like to use is that race result data’s best role in the sport (at the moment) is as a sort of “guardrail for the brain”. Our minds are amazingly talented at taking small bits of information and spinning pretty elaborate stories to explain them. Some simple looks at the data may not provide us with some groundbreaking insight, but it can often prevent us from driving wildly off the road.

So here’s the history of major international classic sprint performances for the US and Canadian skiers:

In line with my previous discussion, I’m not claiming this gives us any big insight. But! I can glance at these graphs and quickly calibrate how I feel about how the US & Canada did today, in a broad sense:

• The US women had a notably off day, maybe not terrible but not great
• The US men did ok, maybe about as well as we might expect
• The Canadian men had a pretty good day (coming from a pretty low baseline)
• The Canadian women had a maybe ok or only slightly disappointing day, again from a pretty low baseline

My one caveat is that this is obviously at a “program level” assessment of the day, rather than at the level of the individual athletes, where their particular histories would potentially lead you to different conclusions about whether their individual results were good or bad today.

The question is how have race speeds changed over time. As you might expect we’re going to limit ourselves to a single race format and distance for consistency, in this case 15/10km interval start events for the men & women respectively:

One women’s race from the 1990’s that is a significant outlier has been removed; I’m quite sure it’s been mis-recorded on the FIS website somehow. The trends are just linear fits to the data. If you really want to squint at the scatterplots you could try to fit a smooth curve and try to spot places where it “jumped” but skiing races are just too variable to cleanly identify that level of change, and the linear trends obviously describe the data quite well.

Just eyeballing the graph, all four events (men & women, freestyle & classic) have improved by around 15%; some a little more, some a little less. It’s easy to brainstorm all the various causes for this improvement:

• skiers train better, eat better, have better technique
• better waxes
• better skis
• better, more reliable grooming (this is my favorite explanation and the most commonly overlooked in my opinion)
• faster courses? (This is kind of an oddball one, but I wonder about it sometimes. Old time ski courses were narrower, with fewer longer loops and more twisty (maybe?), particularly before we had to accommodate skating and mass starts. Those kinds of courses with constant transitions and turning were possibly slower to ski than today’s trails that look like super-highways.)

Every time I make this graph, or something equivalent I have the same somewhat entertaining thought: what sort of courses would we need to have on the WC circuit to force current athletes back to the average speeds of the 1990’s? The 1980’s? It’s kind of amusing to think about what the courses would have to look like to add 10, 20 or even 30 seconds per kilometer to today’s top skiers.