Yesterday took a hell of a toll on me. After I stopped for pizza, I found my Achilles’ tendons were extremely tight. I stretched when I got to my hotel, which helped some, but they’re still very tight, which limits the power I can put down. I found a huge swelling around my right sit bone, and my nether regions are pretty raw in general. Road grit on my right thigh got rubbed on my frame bag, so that’s raw too. I woke at 2 AM with a headache, and tossed and turned after that. The headache has mostly dissipated. I didn’t have an appetite at all, which is a real problem. I forced myself to eat some trail mix, which helped.
I’ve also got a few minor equipment problems I’ll need to deal with.
I’ll admit I’m feeling pretty down right now. I know a race like this will have highs and lows, and after yesterday, it would be a miracle if I weren’t low. And I am.
Stopping for breakfast
My goal in the TABR is to average 210 miles per day. Today I managed 185, considerably short of this goal. But I did ride the longest distance I’ve ever ridden, under very poor conditions, in mountainous terrain (and let me just say: mountains are different than hills). So I feel ok about today’s performance.
I’m going to sleep until I’m done sleeping and then see if I can do it again.
Done for the day in Corvallis
Stopping for pizza. Will probably stop for the night in Corvallis.
Stopped for lunch, first long stop of the day. It’s been rainy and cold all day. I was almost late fir the start and didn’t have time to get all my kit on, so I’m soaked and shivering. Not an auspicious start. I’ve been maintaining a decent speed and feel ok otherwise.
I mentally committed to riding TABR 2021 at the end of January 2020, and started training in earnest in March. I didn’t know what I was doing at first—I was still figuring out structured training, but by June, I had the rudiments of a training plan. I’ve been refining and tinkering with it since then.
That should have been plenty of time for me to get in shape for the TABR, but now I am 11 weeks away, and I feel like it wasn’t.
I can attribute part of this to some health issues (some related to my training, some not) that started cropping up in October and weren’t resolved until February. My fitness didn’t deteriorate during that time, but it didn’t improve either.
But part of it is simply being too complacent with my training plan. I could have pushed myself harder during the four months October—February. And I definitely could have started pushing myself harder immediately after that.
I saw a comment from a coach who has worked with at least a couple of successful TABR racers that one should get one’s cumulative training load up to 120 before the race. I don’t see a good way for me to do that right now.
There are a few concepts to understand here:
- Every workout has a training stress score (TSS) calculated for it. This is a single number that represents both the intensity and duration of the workout. The formula for working it out is complicated.
- Cumulative training load (CTL) is a recency-weighted average of the training stress scores for one’s workouts over the previous six weeks. This is also sometimes called “fitness.”
- Acute training load (ATL) is a recency-weighted average of the training stress scores for one’s workouts over the previous week. This is also sometimes called “fatigue.”
- Training stress balance (TSB) is CTL minus ATL. This is also sometimes called “form.” In order to be making progress, this needs to be a negative number, but if it is beyond -30, that indicates overtraining. This also lets one optimize one’s taper for a race: by easing off training immediately before a race, CTL goes down but TSB goes up. However, I’m not sure how applicable TSB is to a multi-day event like the TABR.
In any case, there’s no way for me to ramp up from where I’m at right now (CTL of 82 as of this writing) to a CTL of 120 without my TSB going deeply negative. In fact, there’s no way for me to ramp up my mileage to where I want it without spending some time in the TSB red zone.
My current training plan has me doing two 60(ish)-minute interval workouts, two 90-minute recovery rides, and a long weekend ride each week; I am using a mesocycle of three weeks, where I ramp up my long-ride distance by 10% for each cycle, and in the third week of each cycle, give myself an easy week with a relatively short weekend ride.
In order to build this plan, I’ve had to estimate the TSS for all my workouts. For the weekday workouts, this is a non-issue. I build the workout in Training Peaks and then I ride it on my stationary bike, so apart from the smart trainer having minor tracking issues, the result is nearly identical to the plan. There are no confounding factors like hills or weather. For the weekend road rides, I alternate hilly rides with flat rides; I worked out an average TSS/mile for both categories based on past rides, and use that when estimating the TSS of upcoming rides. I don’t have a power meter on my bike, so TSS for past rides is calculated based on heart rate (which I know is less accurate).
For my most recent weekend ride, this didn’t work. I had estimated a TSS of 355. Due to steep hills and strong headwinds, it turned out to be 465. It’s three days later and I barely feel recovered from it.
My current plan gets me up to a CTL of 105 before the event, and that is after a recent retooling to give me a more aggressive ramp rate. But now I’m wondering if even that is too aggressive. Especially since I’ll intentionally be hitting TSSs of 465 on some of my weekend rides as I ramp up, and I know how much that took out of me. I am worried that I’ll be both overtrained in terms of my health and undertrained for the event.
With my current training plan, I am cramming most of my TSS ramp-up into my weekend ride. In theory, I could change my training plan so that instead of comparatively light weekday rides and a very heavy weekend ride, I would ride at more consistent TSS levels throughout the week. This would avoid blowing myself out on one weekend ride, but would be harder to fit into weekdays, and in any case, I feel like I need to have the experience of long, uninterrupted hours in the saddle to prepare.
When I was a kid, my father told me about a friend of his, Rudy, who had ridden his bike coast-to-coast. I think this planted a seed.
In the late 90s, I decided to do my own cross-country ride, and resolved to ride the Southern Tier. I was preparing for that in a desultory way, but my life kind of turned upside down in 2000, and I forgot about that goal. At the very beginning of 2010, something reminded me of it and I realized that I still wanted to do it. I mentioned this to Gwen, who gave me a look and said “you’re not getting any younger.” That was all I needed to hear. I started preparing immediately, and in September 2010, I did it.
Although riding the Southern Tier wasn’t exactly easy, it also wasn’t quite the challenge that I was looking for. For some time, I’ve wanted to do another cross-country ride, but one that would be more of a test. I had the idea of doing this in 2020, ten years after the first one. I’d heard of the Transcontinental Race before, and it fascinated me, but the logistics would be so daunting that it just seemed off-limits. In January of 2020, I learned of the existence of the Trans Am Bike Race and I knew instantly that I would do it. This was before the pandemic reached the USA, but I felt that my commitments to Flipside would make it unrealistic to attempt it in 2020, so I set 2021 as my goal. Of course, then the pandemic struck and everything was cancelled. But I was still able to start preparing for the TABR, and I did. This has been the thing that organized my time during a period when time has gotten fuzzy.
The race starts June 6. The course is 4200 miles, give or take. I’m aiming to complete it in 20 days. Registration has opened for it and I’ve signed up.
This is my new bike.
- Kinesis RTD. This is a 55.5 cm frame, bigger than the 54 cm I would usually ride, which I chose at the suggestion of a fitter, who was concerned I wouldn’t be able to get enough stack on the smaller size. As I’ve got it set up now, I could probably lower the stem and still be comfortable.
- Just Riding Along “Mahi Mahi” rims (30 mm front/50 mm rear), SON Deluxe front hub, JRA house-brand rear hubs, Sapim X-ray bladed spokes. I felt like this would give me the best balance of aerodynamics and handling in crosswinds.
- mostly Dura Ace.
- Easton EA90. (Bottom bracket is also from Easton.) I got over my pride and decided to use 46/30 chainrings. For the amount of power that I can produce and my style of riding, I don’t need anything bigger (I sometimes see cyclists in bike forums asking about putting bigger chainrings on their bikes and I wonder “can you really push that gear?”). There aren’t a lot of road cranks that will take rings in those sizes (Shimano’s GRX have a wider chainline, which I could imagine causing shifting problems), which is why I went with Easton cranks. Although I don’t have a power meter on this bike now, Easton’s spindle-based power meter would be a relatively easy retrofit.
- Ultegra 11-30. Having small chainrings meant that I could put a relatively close-spaced cassette in back. This still gives a very low low gear, and a top gear that’s higher than the 52×13 that was typical when I started cycling. If I feel like I can pull it off, I’ll use an 11-28 instead.
- Redshift; can toggle between a conventional position and a TT position on the fly. I got this used, along with a set of Redshift aero bars. After only one ride, I’m not convinced that it suits my purposes, but it is a well-made piece of equipment.
- Fabric Race Line Shallow, which seems to work for me.
- Old stock from Zipp. Available cheaply; bought as a placeholder to confirm my positioning; I will probably replace it with a shock-absorbing stem from Redshift
- Old stock from 3T, also available cheaply.
- Speedplays zeros. I would like to use SPDs, but they don’t offer a lot of float, and that has been causing me knee trouble lately, so at the last minute I made this change.
- Continental GP5000 tubeless, 28 mm. I waffled on whether to go tubeless on this rig and ultimately decided that I would. I’d read horror stories about how hard road tubeless tires can be to mount and inflate, and about how tires that are the tiniest bit out of spec with the rims can blow off. In this case, I was able to mount the tires by hand with little trouble, and even inflate them with just my track pump, no air blaster required. Getting the valves set up took a little trial and error, and I misjudged the ideal length for the stems.
- Wipperman Connex. Has a reputation for being especially durable, and the master link uses a clever design that does not require a tool to connect or disconnect.
This is the first bike I assembled entirely myself. I had a shop face the brake mounts and bottom-bracket shell, and I bought prebuilt wheels, but the rest I did on my own.
The number of weird, specific bits and pieces I needed came as a surprise. I wound up getting a special socket driver for the bottom bracket and another one for the lockring that holds the chainrings to the cranks. I had bought the levers and brake calipers used, and had to buy barbed fittings and “olives” for the hoses (as it turns out, I should have bought new hoses as well). The rear caliper requires mounting bolts that are sized exactly to the frame, so I had to get those as well, and Shimano’s documentation on this is somewhat lacking, so that took a fair amount of research.
Apart from needing two new tools, getting the bottom bracket and cranks set up was a breeze. Although the preload adjuster didn’t stay locked in place.
Cutting the steerer tube was nerve-racking–do it wrong, and the fork is a total loss. Bleeding the brakes was especially nerve-racking. I’ve tried it before and gotten it wrong. I may still need to re-bleed the rear brake, because the bite-point seems a little late, but they both work.
Routing the cables and hoses was surprisingly fussy and involved a fair amount of trial and error. Although the result is OK, cosmetically it could be better. The frame routes the cables and rear-brake hose through the inside of the downtube, which was a slow and fussy task. I’m guessing that a Di2 setup would be considerably more fussy.
Using Shimano’s documentation to get the derailleurs set up was an exercise in frustration, and I really didn’t feel like I had it right until I took the bike around the block a few times and played with the barrel adjuster. I wound up ignoring Shimano’s documentation on bleeding brakes and went with Park Tools’ instead. As a Japanese translator and technical writer, I’m kind of disappointed with Shimano’s documentation–both the translation and the overall approach.
So far I’ve only gone on a couple of neighborhood shakedown rides and one real ride. I’m still getting my position dialed in.
On my one full-length ride I had the tires over-inflated and wound up letting some air out at the turnaround point; I could probably still let more air out. The frame is reputed to be especially smooth-riding, but I was feeling slightly beat up. I may change the seatpost for one with a little give (something my Felt VR30 has), or perhaps a suspended seatpost. I was buffeted by a stiff crosswind the whole way, which was a good test for those aero wheels (these the first aero wheels I’ve had). They definitely took a little extra wrangling, but not more than I expected. My average speed for the ride wound up being about 1 mph faster than I would have expected on the Felt I’ve mostly been riding lately.
Ride Time: 4:53:17
Stopped Time: 27:16
Distance: 75.08 miles
Average: 15.36 mph
I went for a ride yesterday that I can only describe as a bad ride. It was cold and rainy. When I rolled out, I thought I’d be warm enough. I wasn’t. My left knee starting bothering me after about 20 miles. I had planned on riding 85 miles, but got discouraged and turned around early, so I only put in 75.
But you don’t learn anything from a ride that goes perfectly. When things go wrong, you can learn a lot. So what did I learn?
Because my right leg was doing most of the work, I couldn’t ride as hard, so I couldn’t generate as much body heat as usual, so my kit—which might have been warm enough if I were riding harder—wasn’t warm enough. Normally my average heart rate on a ride like this would end up around 125 bpm, and be higher at the end than the beginning; on this ride it was around 123 bpm early in the ride, and by the time the ride ended, it was down to 119 bpm. It’s possible that I wouldn’t have been warm enough anyhow: I’ve worn exactly the same kit to commute to work in the same conditions, but my commute is only about 30 minutes each way, not a continuous five-hour slog. I’m not sure whether to chalk up the difference to riding faster on the commute, or just an ability to tough out the cold for a short period.
I suspect my knee was jacked up because of a combination of the cold tightening my muscles, and my pedals not giving me as much float as I’d like. My natural tendency is to ride in a very toes-out stance. I always used to ride on Bebop pedals, which had 20° free float. Conceptually, they’re a lot like Speedplay pedals, except they are more robust and the cleats fit regular 2-bolt drilling.
Bebop pedals were produced by what I think was a one-man company. Eventually he sold the works to a big Taiwanese company that continued manufacturing them for a few years, but ceased production in 2017. I’ve still got a set, but the cleats are worn out. I’ve got a watchlist set on eBay for these: in four months, I’ve seen two NOS sets listed; in both cases for ridiculous prices, and in both cases they’re snatched up immediately. If there were a Kickstarter to put them back into production, I’d subscribe.
Rather than switching to Speedplay pedals, I’ve been using SPDs. But SPDs only have 4–5° float, and I can’t position them to accommodate my weird stance on the bike. In the warm months, this hasn’t been a problem, but my past two chilly rides have been hard on my left knee, and I’m pretty sure this is why.
So what I’ve learned is that I’ve got to bite the bullet, get warmer cold-weather gear, and get pedals with more float.
I recently got an Apple Watch 6. I had been using a Fitbit Surge for about four years, so my impressions will be colored by that background.
Apple’s packaging is amazing. Everything is sleek and snug. There’s almost no cellophane, no twist-ties, and almost no plastic. The cardstock is smooth and perfectly printed. The shipping box used cardboard spacers instead of peanuts or airbags. One gets the impression that Apple has a packaging-design lab that is better funded than NASA.
Compared to my old Fitbit—which is chunky even as fitness trackers go—the Apple watch is much less obtrusive, more comfortable. I can pretty much forget that I’m wearing it. It sits flatter on my wrist, and the rounded edges seem less prone to catching on things. I got the 44-mm (larger) version. My wrist isn’t that big, but I think it looks right and feels fine.
This is a matter of perspective. The Surge, when it was new, could get about five days of battery life if you didn’t use the GPS. If you did use the GPS, you only got about 7 hours. By the time I gave up on it, though, battery life was barely one day. Apple touts 18-hour battery life for its watch, which would suggest that at its best, it’s no better than my Surge was after years of use (setting aside the fact that Apple’s watch has much more ambitious hardware). In fact, I’ve found that battery life can be much better. I wore the Apple watch for 48 hours continuously, and tracked one 60-minute workout (which drains the battery more quickly, as it runs the heart-rate monitor more, and depending on what you’re doing, also runs the GPS) during that time, and by the end of this, the battery still had 16% charge. I think the reason got this result is because I’ve set the screen so that it is off when I’m not looking at it: the big feature with the previous version of this watch was an always-on display, although the screen was simplified and updated much less frequently when you weren’t actively using it. This is the default mode, and I’m sure Apple’s battery tests are based on this default. So far, I don’t mind waking up the screen.
I need to give Fitbit some credit here—it would automatically recognize that you were working out and would track that automatically. It would attempt to classify your workout based on some characteristic movements, and it wouldn’t always get it right—it would recognize yardwork as exercise, but didn’t have a category for that, and would always classify it as something weird. It did let me reclassify these. The Apple watch only counts exercise when you fire up the “Workout” app and tell it what activity you’re doing. So it’s not possible to “close your rings” passively through normal activities; there’s the additional cognitive load of starting and stopping a workout. the watch does notice when I’m going for a walk (for example) and prompts me to confirm that yes, I’m going for a walk. When I do this, it doesn’t “backdate” to the point when I really started walking. It also notices when I’ve stopped walking, and prompts me to either pause or end the activity. So it’s not doing everything in the background—it does still require intervention—but it is trying to be helpful.
Also, I’ve found that Apple’s estimates of calories burned are very low. I’ve got a smart stationary trainer that measures my power output; this is the gold standard for estimating work performed. Work is simply power × time, and due to a mathematical quirk, calories burned are very close to work performed. I’ve found that Apple’s estimates of calories burned during a stationary-bike workout are about 2/3rds my work performed (although Gwen’s experience is different from mine). Its “active calories” figure is even lower. “Active calories” is apparently Apple’s term for total calories burned minus basal metabolic rate, although my BMR is about 67 kCal/hr, but Apple seems to be calculating it as about 80. So I don’t know what’s going on there.
I also work out with a chest-band heart rate monitor. Apple’s watch seems to lag the chest band by a couple of seconds, but that doesn’t concern me.
Display, hardware, software
The screen is really beautiful, but I have to admit that I can’t fully appreciate it. My eyes are 54 years old, and I’m close to needing reading glasses. I’ve got the type size maxed out; the watch faces seem to use two sizes of text–small and smaller. I can make out the small text; the smaller text I really need to concentrate on, which misses the point of a gadget you can glance at quickly.
Apple gives you a huge number of basic watch-face designs, many with variations, most of which you can customize with complications. I think the most elaborate design allows for 8 complications in addition to the watch face itself. As clever as that seems in the abstract, I find that I am unable to take it all in, or even focus on the one piece of information that I’m interested in (possibly if my eyes were sharper, I could). Four complications seems to be about right.
Text display for notifications seems OK. Clear enough for me to read, although I’m not going to read a book on my wrist.
The hardware is really nice. I covered some of that in Comfort above. I’ve got the woven nylon band, nothing special, and it’s completely unobtrusive.
I’m a little mystified by the two hardware controls: there’s a button and a crown (which also acts as a button). Pressing the button brings up a list of recently used apps; pressing the crown brings up the app browser (either as the impenetrable “grid” or as an alphabetical list) if you’re on the watch face, or surfaces you one level (either to the watch face or up in a menu system) if you’re anywhere else. I’m not convinced both modes are worth having; there’s also a “dock” of favorite apps that you access by swiping up from the bottom. The crown also acts as a scroll wheel; there have been a few times where the crown was the only way to manipulate something that I thought I could manipulate using the touchscreen, which confused me at first.
I haven’t figured out how to hide the Cycle Tracking app. I’m not going to need that for myself, and tracking someone else’s cycle would just be creepy.
Everyone who has eaten pizza is familiar with the wedge cut and tavern cut (aka party cut). But those aren’t the only ways to slice a pizza. For a variety of reasons, the following cuts have never caught on.
The Spiral Cut
Laborious and extremely difficult to do correctly, this is currently only available from Monello’s in Fort Lee NJ when Sal is on shift. Even allowing for its novelty, it is not very popular, because all the crust winds up at one end. Rumor has it that Sal once attempted a double spiral and spent a week in a treatment center as a result.
The Jigsaw Cut
Another laborious cut. The value of this pattern is questionable since no one has ever wanted to assemble a pizza from pieces.
A project begun by Original Ray’s Famous Center for Advanced Studies in Brooklyn, this is an ongoing project to slice a pizza in which each slice is half the width of the previous slice. Begun in 1998, the project is currently raising the funds needed to split a neutron into its component quarks.
The Sierpinski Pie
A shameless attempt at one-upsmanship by Ray’s Famous and Original Research Institute also in Brooklyn, this was an effort at slicing a pizza into a Sierpinski gasket. The project was begun in 1999 and abandoned in 2003 due to the exponentiating slice length.
The Voronoi Cut
A pattern developed by a family of bickering geniuses in Beloit WI to ensure that each slice contained exactly one complete piece of pepperoni, it has found no traction in the wider world.
The Planar Cut
First attempted not at a pizzeria but at a silicon-ingot foundry, this cut slices through the thickness of the pizza rather than across its diameter. Has found increasing popularity with the growth in gluten-free and low-carb diets.
I started getting interested in bikes when I was 13. My interest has waxed and waned over the years, but never really went away. I’ve thought of myself as a cyclist since I got my first pair of cleats at 17. I’ve always been interested in bikes as objects of technology and craftsmanship, and I’ve always enjoyed riding them. At various points I’ve gotten to be a pretty strong rider, but I’ve never been methodical about it—beyond following Eddy Merckx’s advice to ride lots, and perhaps alternating long days with short days, I’ve never been interested in structured training. Lately I have gotten interested, and like so many things, once you pry the lid open on a subject, you discover it goes much deeper than you ever imagined.
I’ve been trying to digest a lot of the concepts involved in structured training, but I haven’t found any one place that packages them up so far. There is a book (and a paper that preceded it) that, well, “wrote the book” on a lot of this, and I may eventually break down and get it. For now, this is my own cheatsheet. Some of this may be wrong. I’m not going to link back to citations for everything. Caveat lector.
- The unit of measure for power. Training these days is structured around power and heart rate. In the past, cyclists would ride to time, or to distance, or to speed.
- The unit of measure for total work performed during a workout. Note that watts are a measure of instantaneous power, but kilojoules are a measure of work over time (same as watt-hours: there are 3600 joules in 1 watt-hour). Thanks to a statistical fluke, the work performed in kilojoules is treated as equal to the calories burned during a workout. So you burn 1000 kCal in a 1000 kJ workout.
- Functional threshold power (FTP)
- The maximum average effort a cyclist can put out for one hour, measured in watts. The gold standard for this is to ride an hour all-out. Most people don’t do this because it’s awful—instead, most people ride a 20-minute FTP test and deduct a certain percentage from that result—typically 5%, but if you’re not exceptionally fit, it should be 10% or 15%. Even riding 20 minutes on the road is difficult just for logistical reasons—you’d want 7 or more miles of level road with no stops, or a hill climb that’s a few miles long, again without stops—so I suspect most people do FTP tests on stationary bikes. FTP is a training tool: it’s used as a baseline for other workouts. All my interval patterns are based on percentages of FTP. It’s also good for bragging rights (unless your FTP is as low as mine, in which case you prefer not to talk about it). I’ve seen articles on how to maximize your FTP test results, but this seems wrong-headed if you’re using it as a training tool, since all the training you base on it will be off. The most celebrated record in cycling is the hour record, which is pretty much what it sounds like: how far you can go in an hour (on a track bike, in a velodrome). The current men’s record for this is about 55 km, and it’s estimated that riders are putting out about 440 watts at that level (which sounds less impressive when you realize it’s about 3/5 horsepower—less than a chainsaw engine puts out).
- Your power-to-weight ratio. Usually refers to Your FTP divided by your weight, but may be used to describe the power you can maintain over shorter periods. A small, strong rider will have a better power-to-weight ratio than a big rider who’s not so strong, but the big rider may have a higher absolute wattage output. These numbers can be important in different contexts: riding fast over level ground is mostly a matter of raw power because once you get past about 16 mph, almost all the drag you encounter is aerodynamic drag, and that increases with the square of your speed. But riding fast up a hill is more about W/kg, because you’re going slower in the first place (so aerodynamics are less important) and gravitational drag becomes a big part of the total resistance you encounter.
- Critical power
- The maximum power that can be maintained indefinitely, in watts. You can put out higher power, but only for a limited period of time, and there’s an asymptotic curve plotting the relationship between power output and duration, so you can put out very high power for a few seconds, fairly high power for a few minutes, and critical power for hours. Going beyond CP means relying to some extent on anaerobic capacity.
- W’ (w prime)
- The capacity for work above critical power, in kilojoules. You can put out effort in the W’ range for 10–15 minutes. The problem with going into the W’ range is that it costs a lot of energy, and once you’re done, you’re very depleted.
- Normalized power
- This is a weighted average power for a workout. Because all the time you spend above critical power tends to burn you out more quickly, a workout in which your effort fluctuates between high and low levels will be harder than a steady-state workout at the same average power. Normalized power reflects that.
- Normalized power divided by FTP.
- Training Stress Score (TSS)
- This is way of expressing a workout’s duration and intensity. Riding for one hour at FTP will have a TSS of 100. This is used to set weekly training goals, so for example you’ll aim for your all your workouts in a week to add up to a TSS of, say, 350.
- Chronic training load (CTL)
- Average TSS over 42 days.
- Acute training load (ATL)
- Average TSS over 7 days.
- Training stress balance (TSB)
- CTL minus ATL. I’ve also seen this referred to as “form.” The idea is that you want to be fresh for big events, so you taper off your training in the days beforehand. This metric is used to optimize your tapering.
- Heartrate (HR)
- Everyone has a maximum heartrate that’s estimated based on age, and heart-rate zones are all calculated based on your max HR. I’ve seen 220 minus age (for men) and 226 minus age (for women) as the most common way to compute this. Going by this, my max HR is 166 (as of this writing). I’ve also seen 208-(0.7 × age), which for me winds up being a few bpm higher (the results from these two equations match for men when they’re 40, for women when they’re 60). I’ve pushed myself to 165 bpm during workouts and held it steady there for a minute or two, so I’ll call these formulas close enough for me. Resting heart rate can improve dramatically with exercise. Four-time Tour de France winner Miguel Indurain was reputed to have a RHR of 26 bpm in his prime.
- Can refer to heart-rate zones or power zones. Anyone who has used a fitbit with heart-rate monitoring has seen the yellow/orange/red zones. And it’s my understanding that the theory behind Orange Theory fitness boutiques is that they keep you in the orange zone. It turns out there are numerous different zone systems, with different numbers of zones and different breakpoints between them, and if you keep track of your data with more than one app/service, keeping their zones in alignment is a chore. Zones are another training tool, where you aim to keep your heart rate or power in a certain zone for a certain period of time during your workout.
- Lactate threshold
- A limit on sustained effort. Any effort beyond LT will basically be a sprint that can only be sustained for about 2½ minutes. This is really interesting to me. When I was younger, we thought of lactic acid as the waste product of exercise, and that lactic-acid accumulation occurred because your body was working too hard to clear it. Now we understand that lactate is part of the energy-transport system (also, we talk about lactate now, not lactic acid), and LT occurs because your body is producing it faster than it can process it. The body needs to deliver oxygen to the muscles in order to clear the lactate from them. We used to talk about the “anaerobic threshold,” that is, a level of intensity where you couldn’t get enough oxygen to your muscles to prevent lactic-acid accumulation, and to some extent, we still do, but nowadays we mostly talk about the lactate threshold to refer to the same idea. I get the impression this is because what we’re describing is the actual measured blood-lactate concentration. Apparently there are numerous ways to define LT, although the one I’ve seen used the most is where the lactate concentration is 4 mmol/l. This is also sometimes called LT2, to distinguish it from LT1, which is the point at which your body starts accummulating lactate faster than it can clear it (“onset of blood lactate accumulation” or OBLA). The effort in watts where LT1 and LT2 are reached can be measured, and obviously, the higher the wattage the better.
- The amount of oxygen you can breathe in per minute, in units of ml/min (sometimes expressed relative to body weight, ml/min/kg). The gold standard for testing this is to exercise to exhaustion while wearing a mask that measures the volume of air you breathe, but it can be approximated based on max and resting HR, or heart rate after specified exercises. There’s obviously a relationship between VO2max and LT, and LT is generally estimated to occur at 50–60% VO2max for untrained athletes, and 70–80% for highly trained athletes.
- A workout where you alternate periods of high intensity with low intensity.
- High-intensity interval training. A style of intervals where you ride close to VO2max with relatively long “on” intervals.
- Sprint intervals
- A style of intervals where you ride beyond VO2max with short “on” intervals.
- Polarized training
- An approach to training based on long periods at low intensity and short periods at very high intensity. The balance can be in the range 85/15 to 95/5. The idea is that you spend all your time in Zones 1/2 or 5, and none in Zones 3/4 (assuming you use a 5-zone system)—or that you simplify your zones to “easy/no-mans-land/hard.”
- Threshold training
- The opposite of polarized training: spend all your time in Zones 3/4. There seem to be different definitions for this (just as there are different definitions of zones), but the general idea seems to be steady-state training at 75–90% of FTP. Current scholarship seems to be that polarized training gets results faster, but threshold training has a lot of adherents.
- Sweetspot training
- Same as threshold training
- Moderate intensity continuous training (MICT)
- Same as threshold training
- High-volume training
- Long miles at low intensity. Also called “base mileage.” My understanding is that you’re supposed to have a lot of base mileage before you attempt higher-intensity training.
- Pedal speed, measured in rpm. Most experienced road cyclists aim to keep their cadence around 80–90 rpm. There’s evidence that lower cadences are more energy-efficient, but it’s easier to produce more power at higher cadences.
- Bike computer
- Also called a “head unit.” Bike computers used to just compute riding time, speed, and distance, and if you were really fancy, you got one with a barometric altimeter. Those still exist, but these days, fancier ones are tantamount to smartphones (many of them run the Android OS), and connect to a constellation of other devices.
- Heartrate monitor (HRM)
- There are two types: the traditional chest strap, which monitors the electrical activity in your chest muscles, and the newer optical type, often built into a watch or armband, which visually detects the changes in your veins as blood flows through (word of the day for this: photoplethysmograph). Chest straps are usually cheaper and have longer battery life. Some of the optical ones can also work as pulse oximeters.
- Power meter
- A meticulously calibrated strain gauge applied to a part of the bike’s drivetrain (pedals, crankarms, crank spider, rear hub) that transmits its readings to a bike computer. Although they’re coming down in price, power meters are still really expensive, and add at least $200 to the price of whatever part they’re attached to. Having a power meter on the bike has transformed competitive cycling, as racers know how much power they can put out for how long, and don’t make heroic-but-doomed breakaways.
- Smart trainer
- An old-fashioned stationary trainer would usually let you vary resistance manually, but did not calibrate this beyond “easy” to “hard.” A smart trainer uses computer control to apply resistance, and can measure power reliably. Smart trainers can use smartphones or bike computers as their controllers and front ends.
- Golden Cheetah
- Freeware analytic software. I first downloaded it about four years ago, and was so confused by what I was seeing that did not look at it again until last week. I still don’t understand half of what’s going on in there, but I’ve got a little bit of a handle on it.
I almost never use bookmarks in my browser. I can get to the sites that I visit regularly by typing a few letters into the URL field of my browser. For the most part, if a site interests me and I might want to come back to it in the future, I add it to my RSS reader. And that means that there are some sites lurking in my list of feeds that haven’t been updated in years, but are still out there. And they’re still pretty cool, so I thought I would share them.
Most of these are visual, and are easy to dip in and out of; a few require some commitment.
- Object Lessons
- A blog about the design of things.
- Agence Eureka
- Scans of French ephemera
- Between Mirrors
- Morbid artwork
- Scans from old illustrated books, with related scholarship.
- Concept Ships
- Features spaceship artwork; links to related blogs for concept land vehicles, robots, etc.
- North Korean Interiors
- What it says on the tin. Either inspired by or inspiring to Wes Anderson.
- Scales of Perception
- Images that play with your sense of scale.
- Type Hunting
- “Found typography,” mostly on old packaging.
- Typeset in the future
- Deep dives into the typography used in SF movies. One of those “blogs that became a book” sites.
- WTF, Evolution?!
- Another one of those “blogs that became a book” sites.
- Bret Victor
- An amazing technologist. I am guessing he is too busy with work to update this much.
- Very deep dives into Japanese pop culture in the postwar era.
Every weekday as part of my commute, I ride across the old Montopolis Bridge, which is closed to motor traffic but still open to bikes and pedestrians. The city is restoring the bridge, and have built this tunnel to shield passersby from the work going on above.
When I ride through it in the morning, heading east into the sun, with my sunglasses on, I can’t see anything immediately around me and I feel as if I’m floating.
For many years, I subscribed to the Economist, and it was my primary news source. I would spend about an hour a day working my way through it. I let that subscription lapse for a while, and tried resubscribing, but discovered that during my time away, my perspective had changed, and I was put off by what I thought of as their “free-market triumphalism:” This is their schtick. Their straight news reportage is great, but they also have a lot of opinion writing, and they are quite open about their bias that free-market mechanisms can solve just about any problem. I felt I had seen too much evidence that these mechanisms were not working, and that their philosophical position was out of sync with the real world.
I was reminded of this when I ran across an article on California’s cap-and-trade system. Long story short, the system is subject to being gamed. Surprise.
This got me thinking about the rigor with which market-based remedies are applied in general. In political discussions today, we generally see two approaches to solving public problems: firm regulations—meaning hard limits, taxes, etc; and “business friendly”—meaning businesses can do whatever they want.
And this made me wonder whether we’ve ever seen a fair example of market-based remedies being put into practice. In order to see this, we’d need to see a free market in practice.
Free-market types will argue that we have a free market, and that distortions will eventually come to light, and the market will correct for them. This is false—or if it isn’t false, then we have different ideas about what we mean by the phrase “free market.” If “free market” simply means “whatever you can get away with,” then yes, we have the free market that free-market types advocate for. But if it means “level playing field,” then we do not have a free market. Level playing fields do not naturally emerge, and they require constant maintenance to keep them level. We see examples of this over and over again. When you buy a car, if you want to buy a new car, you’re obliged to sign an agreement to use arbitration (in case the dealership sells you a lemon, etc), with the arbitrator being chosen by the dealer. The only way to opt out of arbitration is to opt out of buying a new car at all. In labor markets, we see employers exerting more and more onerous demands of their workers, going so far as to include non-compete clauses in the contracts of minimum-wage employees. If you don’t like those terms, don’t get a job. In competitions between businesses, there is generally an advantage to incumbents, which incumbents seek to maximize. The entire investment approach of the Softbank Vision has been to create monopolies that can subvert free markets —admittedly this has not been working out so well for them, which might support the argument that free markets correct distortions, but the fact that this was a reasonable investment strategy to contemplate in the first place, and that some entire industries have been destroyed along the way, is an argument against.
In all these deals, there is clearly an imbalance between the parties.
We see the same thing when it comes to solving public problems, such as pollution, only more so. There is a small number of well-funded organizations with a vested interest in one outcome, and a large number of unfunded individuals, and some public-interest organizations, with an interest in the other outcome. One of the two sides is able to tilt the playing field to their advantage.
When we talk about market-based remedies, we’re using the language of the “business-friendly” crowd, and so it is natural to conclude that these remedies are intended to achieve business-friendly outcomes. And it seems that’s usually the case. But what if the means do not necessarily entail the ends? What if market-based remedies were a tool for achieving rigorous regulatory goals? I don’t think this idea has ever received a fair shake. This would require narrow guardrails with sharp edges to keep people on the straight and narrow.
I still think that the Economist’s free-market triumphalism is wrong, but now I wonder whether it is wrong in theory or just wrong in practice.
SRAM just introduced a new version of their top-end road group, Red, with a different approach to gear ratios that they claim gives more range and tighter spacing. This smelled like marketing hokum to me.
Warning extreme nerdery follows.…
Microsoft’s Word’s ubiquity is rivaled only by its badness. Since we’re stuck using it—and often using files created by other people in it—we need to find coping mechanisms.
One especially vexing problem in Word is the way it deals with placed graphics. This post isn’t an exhaustive tutorial on how to work with graphics in Word–it lays out one method that will work in most cases, and explains how to make that work.
Let’s say you receive a file that looks something like this, with a placed photo and some text boxes and arrows laid over it to call out features.
You edit the file, do something seemingly innocuous, and you wind up with something like this
Obviously you can’t let the file go out into the world like this, and because you are a good person, you want to leave things better than you found them. So how do you fix this? Or if you’re required to create files like this, how do you prevent this from happening in the first place?
It’s easy to get into trouble with Word any time you try using its page-layout features. If at all possible, it’s best to treat every document as a continuous river of text, rather than isolated blocks. The problem with images is that Word gives you numerous options for treating images as isolated blocks, and exactly one option for treating them as part of that river. When you mix externally created images and graphics that are created in Word, things get complicated. And these are overlaid on one another, things get even more complicated.
In the image shown above, there’s a photo that was created externally, and three text boxes and arrows that were created within Word. So the first thing to understand is how Word treats these differently: the photo is a picture and the arrows & text boxes are shapes. They have different formatting options available to them. However, interestingly, you can crop a picture in Word using its “picture format” tools, and that turns it into a shape (!).
Most of the trouble you run into with these hybrid images revolve around placement options. Word gives you two sets of parameters for dealing with pictures/shapes in text: positioning and text wrap
If a visual element has the positioning in line with text, then it behaves like a typed character—it can sit on a line with other characters, it moves around with other elements, etc. And I argue that this should be your goal for most or all visual elements you use in Word. You can set them on their own line, use other techniques to marry them to captions, center them, etc.
With all the other positioning options, the element is anchored to a spot on the page—a certain distance from a corner, for example. If you anchor the element, you use the wrapping options to tell Word how to wrap text around (or over, or under) the element. There may be legitimate reasons to do this, but Word is a rotten tool if that’s what you’re trying to do. I often see files where someone has placed an image with fixed positioning that they really just want inline with text—and then they insert a bunch of carriage returns to put the text down below it. This will break as soon as the text above gets a little longer or shorter.
Also, just for fun, if you set the wrap to “in line with text,” Word automatically does the same for the positioning, and vice-versa. This kind of makes sense, but can be confusing.
To simplify your life, treat each graphic as a standalone block, on its own line, flowing with the text.
This gets more complicated when you’re combining a picture with shapes. By default, the picture is placed “inline”. By default, a shape is positioned relative to something—positioning can be relative to the page, margin, paragraph, line, with separate options for horizontal and vertical position. Ain’t nobody got time for that.
So we’re back to inline positioning as the right way.
But with the Orientalist mysticism that you only find in cheesy action movies, when you’re dealing with a hybrid image like this, Word forces you to do things the wrong way before you can do them the right way. Here’s the trick: we need to manually position the picture and the shapes relative to each other. And Word doesn’t let you manually position elements that have inline positioning—again, it does make sense, but is confusing until you understand the principle.
First, make sure that all the visual elements have some kind of positioning that isn’t inline—it doesn’t really matter what.
Second, get all the shapes lined up correctly over the picture that acts as the backdrop. If some of the shapes are getting hidden behind the picture, select the picture and then execute Picture Format > Arrange > Send to Back.
Third, I like to group all the shape elements. This is probably unnecessary. Shift-click on all the elements in turn to select them and then execute Shape Format > Arrange > Group. The image below shows the shape elements grouped together, with a frame around them. You can still separately manipulate the elements in a group—it’s possible to move a grouped element unintentionally; if you need to move the group, you need to grab it by the group’s frame.
Fourth, shift-click to select the grouped shape elements and the background picture, and group those.
Fifth, set the positioning of these grouped elements to “inline with text.” Phew! It’s faster to do than to read.