Monthly Archives: May 2019

Observational Studies and causation

There’s a problem with observational studies.

Let’s say you tell people to do something – eat less red meat, for example – you are hoping to change their behavior. You end up with some people who totally avoid red meat, some people who reduce the amount of red meat they eat, and some people who just ignore you.

Then you come back 10 or 20 years later and do an observational study and look at how much red meat people eat and how healthy they are, and – lo and behold – you find out that those people who eat less red meat are healthier.

So, you publish your study and a bunch of other people publish their studies.

Unfortunately, there’s a problem; the act of telling people what to do is messing with your results. The people who listened to your advice to give up red meat are fundamentally more interested in their health than those that didn’t listen in a myriad of ways. Those differences are known as “confounders”, and studies use statistical techniques to reduce the impact of confounders on the results, but they can never get rid of all the confounders. Which leaves us with a problem: we don’t know big the residual confounders are in comparison with any real effect we might be seeing.

Residual confounding is why those studies can never show causality; if you look at the studies themselves, they will say there is an association with red meat consumption and increased mortality.

But in the press releases from the research groups or universities, causality is often assumed.



7 Hills of Kirkland Metric Century 2019

Normally, I start a ride report with a description of the ride, how much I like it, etcetera, etcetera, etcetera.

In this case, I’ve done 7 hills so often that it hardly seems worth the effort. If you want that information, there are plenty of examples here

I *will* note that this is the 12th time that I’ve done 7 hills, if you ignore the fact that I skipped a year or two when it was wet.

My typical approach for a ride like 7 hills is what I can the “make sure I finish” approach; start slowly and ride conservatively, and you can be confident that you are going to make it back for your serving of finish line strawberry shortcake, delta any mechanical issues with the bike or the rider.

This year I said, “the hell with that!”

I’ve been playing around with my training this year; I’ve been riding more hills earlier in the season than I have in the past and I’ve been feeling pretty good both aerobically and on climbs. So, the plan is to push a lot harder than I have in the past and see what happens.

That should provide more opportunity for humor and perhaps some pathos as well.

My big preparation for the ride was doing a one-hour recovery ride two days earlier that turned into a 25 minute recovery ride when it started 5% chance of precipationing on me and I headed for home.

Woke up at 5:30 due to my old dude internal clock, got my stuff together, and had a couple servings of SuperStarch. I’m put BioSteel hydration mix in my water bottle (no need for 2 on this ride), and headed for Kirkland about 6:50.

You may ask yourself why I drive to this ride when the route passes within about 3 miles of my house. I’ve ridden to the route and done the ride that way, riding the last few miles at the start of my day and skipping it later. I found it messed with my aesthetic appreciation of the route, so now I drive.

Parked on the waterfront, got out the bike, stuffed my pockets with stuff, and headed to the start. Joyously absent was any thought at all about what I would wear; it was already about 60 degrees and was forecast to get into the low 70s, so no arm warmers, no leg warmers, no jacket, no hat; just the usual minimal stuff.

Market street is the first hill; I climbed it in a little over 4 minutes at 211 watts. It’s just a warmup, as is hill 2, Juanita drive. Hit the light, turned down Holmes point, headed towards Seminary hill.

Seminary is one of the two hardest hills; Winery is steeper but gives you chances to rest, while Seminary is more of a constant annoyance. I rode easy on the first little blip, and then rode hard. The top came 8:42 later, more than a minute faster than last year’s effort and 16 seconds slower than my PR from 2016. But… averaging a fairly significant 270 watts for that time, which is pretty decent for me.

The usual descent and trip over to Norway hill, a nice 426’ hill that I’ve ridden up a lot and I backed off a little at climbed at 234 watts, which was pretty much my target.

Which brings us to Winery. I like to have some rabbits to chase up winery so I was okay when I got passed on the flat part before the climb, I was less okay when the blocked the whole lane on the little bump over the railroad track. I got around them and took off up the steepest first pitch, riding at about 420 watts. It’s a short pitch and I kept that power over it, and then slowed down to recover for the upcoming pitches. 5:26 later I was turning off at the top of the climb and listening to bagpipe music, a heartbreaking 4 seconds from my PR. I’m going to call this one a “virtual PR” because I lost more than that getting by the group at the bottom. 287 watts average was a great effort, but when I went back and looked at last year’s data, I did it in 5:22 but only averaged 250 watts. Not sure what is going on there, though I am in need of some drivetrain maintenance, and it would be good to do that on the bike as well. 

We then headed east and climbed a few more hills, then we came to Novelty Hill.

It would be fair to say that it’s not my favorite part of the ride; too much traffic and not really a very fun hill. This is compounded by the use of the lower part of the climb as an “out and back” route; as you are climbing up the hill there are riders who are ahead of you descending back down at a high rate of speed, not really the most motivating thing to see. Strava somewhat strangely didn’t match the whole climb for me, but I got PRs on various sections so I’m going to call a PR on that section.

After some flat roads, you end up coming back back over to Novelty for the descent, completing the circle of life. We learned about the circle of life from Disney’s “The Lion King” during the scene where Simba and Nala protected themselves from roving hyenas by building a impregnable perimeter from family-sized boxes of cereal.

Anyway, a couple climbs after that I hit the last rest stop and after a thoroughly pedestrian sandwich (turkey/cheese/green pepper slices on pita bread), headed out for the last climb. My intention was to spin up old redmond road at take it easy, but there was a guy right in front of me so I ended up pushing a bit and coming within my PR by about 10%. Then a couple of fun descents and a mostly-flat trip back to the starting line in which I missed every single traffic light.

Overall, a pretty good effort; I felt strong the whole day which has been an issue for me this spring; I’m not sure if it was the BioSteel or the SuperStarch or my smoked almonds or maybe that small oatmeal gluten free cookie I ate at the first food stop (it was *not* the cookie; that was a mistake).

I was about 15 minutes faster than 2018, averaging 14.7 mph rather than last year’s 14.1, finishing in 4:04:28 and burning 2538 calories for the effort.




Minicamp May 2019

My wife and I have done a few cycling vacations. The ones we’ve done don’t feature particularly long days – maybe 50 miles over the whole day – but they do involve riding for a bunch of days in a row. I’ve noticed that doing something like that helps my fitness; I just feel better overall.

And therefore I decided to conduct an experiment; I would ride 5 days in a row and see what happened. I wanted every ride to be at least 3 hours, but I wasn’t going for century lengths. And I would ride however I felt like that day.

I expected that I’d start out feeling okay and gradually get more tired as the days went by.

Day 1 was a big hills day; I rode a few of the Issaquah Alps. My speed on the first 3 (Squak, Talus, Zoo) was a conservative speed, but after Zoo I took a trip up Pinnacles and decided that I wasn’t up for Belvedeere, much less the trip up The Widowmaker. I crawled up the back side of Summit and headed for home.

Day 2 was supposed to be a ride all the way around Lake Washington, but after doing the south end I opted to take the 520 bridge back across for home. Felt okay but not great.

Day 3 was an evening ride that I lead. I chose the route to be a little hilly but not too hilly. On the ride down to the starting point, that seemed like a really good decision as my legs were hurting, but despite the hurting, they seemed to perform okay when I needed them. I have a 275’ hill on the way home from the ride with a couple of short 13-15% kickers, and those were not fun *at all*.

Day 4 was a ride in the country, specifically a ride out to Fall City. The intent was for it to be moderately hilly. My legs were tired from the night and I wanted to let the day warm up a bit, so I delayed my start until 11 AM. Legs were pretty sore but warmed up quickly. I had planned to ride up Sahalee (0.9 miles, 404’ of up) but that can be a long slog of a climb, so instead, I decided to head up “The Gate” (0.2 miles, 158’). That’s an average of 15%, with a top gradient of perhaps 21%. I didn’t have a lot of pop on it, but I rode up it okay with just a wee bit of paperboying. Worked my way east, the south, rode down Duthie, and then out to Fall City. Where I stopped at the grocer for a Coke Zero. My plan was to take Fall City –> Issaquah back, and take it I did, via the back way. Despite being on the 4th day and 25 miles into the ride, I was able to climb at about 250 watts pretty easy. Hit the top, finished my Coke Zero, did the bonus, and then worked my way to Issaquah and then back home.

Day 5 was the second evening ride for the week. I played with intensity as I spun through Marymoor, and my legs seemed fatigued but okay. The first climb was short but not a lot of fun. And then we hit Sahalee… I started slow, hit the steep spot, and found that my legs felt pretty good, so I rode the rest of ride at a bit more than 300 watts, averaging 280 for the whole climb. That put me close to my PR on the climb, which was a surprise. I did a sprint up a little steep hill on the route and managed somewhere in the mid 900 watts, though my legs *really really really* hurt at the top. I did notice that my aerobic recovery was pretty quick. After playing down the plateau we descended to East Lake Sam and pacelined back and I managed to “win” the fake sprint at the end by pulling out about 30 seconds from the end. My legs felt good, and the climb up to my house was considerably easier than on Tuesday.

Day 6 was designed just to warm up my legs and help them to recover a bit, so a 3.8 mile ride that took less than 20 minutes.



















































Day Distance Elevation Speed KJ
1 33.9 4177 10.9 1682
2 39.6 1575 14.3 1486
3 35.4 1788 14.5 1345
4 41.0 2470 13.6 1587
5 35.1 1903 14.3 1372
6 3.8 180 12.7 119
Total 188.8 12093 7591


The true test is going to be what my form is like after recovering for a few days, but early indications are that the minicamp did what I was hoping; I felt stronger in places where I hoped to feel stronger and my recovery seemed pretty good. I was mostly able to sleep quite well, and – somewhat surprisingly – my hunger didn’t seem to increase that much.