Tuesday, December 30, 2014

Vector pedal stroke analysis metrics available

Fulfilling a long-term promise, the Garmin Vector now does pedalstroke analysis! See DC Rainmaker's excellent blog post on the latest firmware update.

The new metrics include:

  1. pedal force offset: this measures how far outboard or inboard the average force is applied.
  2. pedal power stroke: this measures over which angles peak propulsive force is applied
  3. seated versus standing time: if you're standing, the pedals support full body weight (non-propulsive), so the Vector can determine how much time you're standing versus sitting (and record whether you're standing or sitting).

Various applications for these metrics would be bike fit, technique analysis, and interpretation of performance. For the pedal force offset, that obviously suggests a bike fit application. "Power stroke" suggests both fit and technique. Standing versus seating suggests performance analysis: am I faster seated or standing on short steep climbs? What about long sustained climbs?

More data is almost never a good thing. The key is how to act on it.

On fit, saddle height optimization is an obvious application. Can I see some signature in the metrics of the saddle being too high? Does the power stroke compress as the bottom of the pedal stroke becomes less accessible when the seat is too high? Does the top of the pedal stroke becomes less productive when the saddle is too low? I don't know.

My key objection is that the update is provided only for the 510, 810, and 1000. For me, the 500 is the best unit most of the time: it's the most compact, the lightest, and it's relatively easy to use. In most cases, I analyze the data after the event: I don't need diverse metrics displayed during the ride. The numbers are produced by the Vector, not the head unit, since they involve samples taken at substantially higher than the 1-per-second rate of the head unit. So updating the head unit would be simply a matter of accepting the associated data field types. As far as I'm concerned, they don't even need to be displayable. I just want them to be recorded.

Rainmaker made the argument that the Edge 500 is 5 years old now and you can't expect Garmin to continue to support 5-year-old hardware. But this is misleading: the unit isn't some deprecated unit which has been displaced by a superior product in the same space. But they don't, and the Edge 500 is still manufactured and is still popular. If you walk into a bike shop and see it for sale today, that it's 5 years old is trivia. As far as you're concerned it's a current product.

The key point here is that this isn't about increasing the value of an Edge 500, which already has well-established value. Rather this is about increasing the value of the Vector, a much newer unit, so that it works fully with the Edge 500.

So I wish Garmin would fix this unfortunate decision: after spending $1500 on a Vector, customers committed to the 500 shouldn't miss out on important functionality which differentiates the Vector from cheaper alternatives.

Saturday, December 27, 2014

modern bar design: the Enve SES Aero

Caley Fretz is one of my favorite cycling reviewers. He now works for VeloNews, where he wrote this review on the new Enve SES bar:

The shape of the SES is a reflection of the way that positioning has changed in recent years — as bars have dropped relative to the saddle, and our understanding of aerodynamics has improved, it has become clear that the lowest drag is often found with hands on the hoods, elbows bent at 90 degrees, chin to the stem.

The SES design allows for a narrow, aerodynamic hood position — 37cm on my test bar — while retaining a wider drop area, 42cm. Narrow when you want it, wide when you need it.

Here's the Enve:


I love new ideas!

In completely separate news, VeloPress has recently published Goggles and Dust, a book based on the remarkable photos which are part of the Horton Collection:


Tuesday, December 23, 2014

moving target: restoration of Skyline Boulevard Bridge

Back in the day, back in the early days of recorded time (i.e. since I got a Garmin 500 and signed up for Strava), we used to ride an SF2G Skyline route which didn't descend Crystal Springs to Polhemus bus instead remained on Skyline Boulevard, crossed a bridge at the dam under the 280 Flyover, and then climbed back to Highway 92.   The memories of that nearly forgotten time must be reconstructed from the yellowing pages of the Strava archives lest they be forgotten forever.

Note the rather direct route near Crystal Springs. Now compare and contrast:

You can see the difference near the "Belmont" label: a left, right, right, left while the earlier route takes a relatively straight line.

The alternate route involves a descent of Crystal Springs Road, a right turn onto Polhemus, a long climb there to where it turns into Ralston past the Highway 92 interchange, then passing through a parking lot to descend back to Cañada Road via a twisting bike-pedestrian path which ends with a stiff but short climb to cross over Highway 280. The whole thing adds around 15 minutes for a fast southbound rider.

Why? The bridge on Skyline Boulevard was assigned to be updated for seismic compliance. The promised finish: an incomprehensibly long 2.5 years, until the spring of 2013. Wow. But somehow we'd survive.

It's now fall 2014, and the projected completion date: spring 2017, 2.4 years. So they've made 0.1 years of progress in the past 4 years, an efficiency of 2.5%. Building the entire Grand Coulee Dam took 9 years, so they're looking now at a renovation project on a small dam and bridge which will take most of the time to build one of the great engineering feats of our history, close to double the four years it took to build the Golden Gate Bridge.

In response, some riders have taken to occasionally riding the Highway 280 shoulder along the "flyover", the connecting Golf Course Road and Belmont Road. The 280 shoulder is already legal in two sections further north thanks to the hard work of bike advocates in the 1970's, as at the time it was argued that there was no timely alternative for north-south travel. This is the condition today: the Polhemus detour adds a substantial time and effort to the ride.

Personally I've not objected too strongly -- the goal in riding to work via the hillier and longer Skyline route instead of the flatter and faster Bayway route is for fitness. But for some, in particular those working closer to the mountains than the bay on the peninsula, Skyline is faster and riders have a stronger motivation for reduced transit times. But given the history in delays for the project it makes sense to allow cyclists to use this section of 280. The primary issue is simply one of acrophobia, since the "flyover" is elevated well above ground level. But if a car hits you hard enough to knock you off the edge you were probably a goner anyway.

The following is a history of the projections for the completion of the seismic upgrade project, which was forwarded to the SF2G mailing list. I'm not sure who the original author is, but it's good stuff. I'd love to know why the project is apparently progressing so much slower than originally projected.

Here's there's original dates still on this blog.

 The demolition is on schedule for completion by February 2011, with the replacement bridge opening in 2013.

The internet archive has May 4 2011:
Skyline Boulevard between Crystal Springs Road and Bunker Hill Drive is closed as of Friday October 29, 2010, and is expected to be reopened in the Summer of 2013.  Click the road detour map for detour information.

On May 12 2012
Skyline Boulevard between Crystal Springs Road and Bunker Hill Drive is closed as of Friday October 29, 2010, and is expected to be reopened in the Fall of 2014.  Click the road detour map for detour information.

Then in July 2012
Skyline Boulevard between Crystal Springs Road and Bunker Hill Drive is closed as of Friday October 29, 2010, through Fall of 2014.

As a result of the San Francisco Public Utilities Commission's (SFPUC's) Crystal Springs/San Andreas Water Transmission System Upgrade (CSSA) project, construction of the County's new Crystal Springs Dam Bridge project will be delayed until approximately September 2013.

October 2013:
Skyline Boulevard between Crystal Springs Road and Bunker Hill Drive is closed as of Friday October 29, 2010, through Spring of 2015.

As a result of the San Francisco Public Utilities Commission's (SFPUC's) Crystal Springs/San Andreas Water Transmission System Upgrade (CSSA) project, construction of the County's new Crystal Springs Dam Bridge project will be delayed until approximately Spring 2014.  To facilitate construction of the SFPUC's CSSA project, the scenic vista point parking area located south of the Sawyer Camp Trail terminus will be occupied by SFPUC's contractor for material/equipment staging until approximately Spring 2014.  Subsequently, construction for the County's Crystal Springs Dam Bridge replacement project is expected to take about 1 year to complete with an anticipated reopening of Skyline Boulevard (Highway 35) between Crystal Springs Road and Bunker Hill Drive in the Spring of 2015.

So let's see progress:
dateestimated time to completion
04 May 201127 months
12 May 201230 months
July 201228 months
Oct 201319 months
Dec 201429 months

So in the past 3 years of work, the projected time to completion has actually increased 2 months, from 27 to 29. That's not moving forwards, but rather backwards. Is this a new dose of honesty and we'll all be riding over the bridge by summer solstice 2017? I hardly think so. But the project has to be finished at some point, assuming there's not an earthquake during the interval which sends it all crashing down.

Friday, December 12, 2014

2014 Low-Key Hillclimbs: Mount Hamilton report

Thanksgiving marked the ninth and final climb in the 2014 Low-Key Hillclimbs. And as it has been every year we've done the series, Thanksgiving marked the longest "single" climb we do, Mount Hamilton Road.

Single climb is in quotes because John Summerson has asserted that Hamilton isn't a single climb, but rather three. And indeed the descents are considerable. But with a net elevation difference of 1161 meters and total climbing of 1340 meters, that's an "altitude efficiency" of 86.6%, so it is considerably more up than down. A series of three equal climbs separated by two descents to the original elevation would be an efficiency of only 33.3%.

I began my day in San Jose, having stayed with a friend, and a weather check showed the temperature in the city was 44F but at the summit, at 7 am or so, it was already 62F. So it was unusually warm, a common trend this year. Despite this I wore a wool baselayer in addition to knee warmers, shorts, and a short-sleeve jersey. With a light jacket this was sufficient for the warming morning air at base elevation. Before the start I put the light jacket into Kevin Winterfield's rental car, which was doing support duty,

I gave my speech advising riders to not start further up than they should: I 'd assigned group numbers from 1 to 5 to riders for whom I had availanle scores to do so, with 1 being they should line up near the front, 5 near the back. There's a perceived advantage in starting near the front, but if slower riders start too far up it creates passing situations which result in crash risks. So it's way better if people start in the approximate finish order. It's hardly appropriate for me to call for this then step right up to the front of the pack myself, but I tend to overcompensate a bit, and I was starting back with group 2 people which was not the best place for my goal of avoiding the start line snafus which caused me to be delayed last year when a rider had trouble clipping in in front of me and caused me to burn a match to cross a gap.

But while my starting position could easily have been trouble, the pace at the start was unusually sedate. I had no problem following wheels, and was able to take advantage to move ahead when space opened up toward the left of the right lane to which we were dutifully confining ourselves. This proved a good move, because not long after moving up I looked back and our group had separated itself.

But up the road, three had already taken advantage of the relaxed start: Chris Evans and series leader Stafano Profumo of Squadra SF, along with Hanns Detlefsen of Sistes and Misters had established a considerable gap. Back in my group, favorites Ryan Sherlock and Adrien Costa along with dark horse Brian Lucido sort of neutralized each other, none wanting to drag the others up this first of three climbs where "keeping the powder dry" is the prevailing strategy. That strategy was not held back in the last decade, or the decade prior by Tracy Colwell or Tim Clark, each of whom used their dominent strength to shed the field from the front. But this was going to be far more tactical.

By Grant Ranch I was still feeling fine. The lead three had a 3-minute gap at this point, and extended it further on the close-to-two-mile descent before the start of climb #2. The descent has in the past caused me problems, since with a tendency for gravel to accumulate in the corners exercising too much caution is always a temptation. But I had no issue keeping up with the riders around me. Part of this was due to using my carbon Ritchey MT-32 wheel with Veloflex Carbon road tire rather than the Edge 100 rim with Vittoria Crono time trial tire I more often use on climbs. This wasn't a tactical choice: the Crono is flat and my attempt to order a replacment in time failed due to Sports Basement getting the wrong tire.

But the front wheel, with better aerodynamics, traction, and braking even if it had slightly inferior rolling resistance and a significant amount of extra mass, was probably the best choice today, and I was able to descend confidently.

Different years different riders have different tactics and it is obvious that one of the tactics this year was to shred the lead group on climb 2, because as soon as we hit the end of the descent, I found myself too deep in the red for the 2848 vertical feet still to come. I had two options: stay with the group as long as possible and then hope to struggle the rest of the way, or throttle back immediately and try to ride a more sustainable pace. I chose the latter option. I thus found myself with Shahram Moatazedi and Oleksiy Mishchenko. Shahram was clearly strong, and I didn't see any obvious signs Oleksiy wasn't also ready to keep a good pace. Shahram pulled at first, then Oleksiy came through, then I took my turn.

There's always an initial enthusiasm at the bottom of a climb, and this was no exception. A later review of my data, applying a power-speed model, showed even though we let the leaders go this was still what for me is an unsustainable pace. But I still felt okay, and as I moved to the front I thought maybe I'd succed in dropping someone. But no luck: they stayed on my wheel. So I put in a decent time at the front then pulled off to let Shahram have another shift at the front.

At about this time I saw the unmistable figure of Daryl Spano up the road. He's tried to stay with the leaders longer. clearly, but had reached the same conclusion I had eventually. Daryl and I are often closely matched, so if I can ride with or faster than Daryl, I know I'm doing well.

With Daryl as a rabbit, we closed the gap, and Daryl joined our group. I'm not sure where we lost Oleksiy, but it was somewhere along this second climb. That left the three of us. I think I took one more pull along the way, but the climb gains only 788 vertical feet in its 3.2 miles, so we were to the top relatively quickly by Hamilton standards.

The second descent with our small group wasn't an issue, and we were soon at the bridge which marks the start of climb #3. This bridge offers a clear view of the summit observatory, which while it appears tantilizingly close, is still 6.6 miles and 2060 vertical feet away. This final climb is longer than all but 16 of the 53 climbs for which I have stats from Low-Key history, so it's a challenging climb by Low-Key standards all by itself. It demands respect. Optimism due to the view must be tempered.

The entry to the bridge is a corner where Nils Takkenen crashed three years ago: it's tempting to rail it but it can take you down. I apexed it a bit more than I probably needed to, sticking with my two companions.

And then we were onto the climb. I tried to follow Shahram but he went out a bit too fast for what I had left in the tank and I let a gap open. I was able to limit the damage, though, and the gap stabilized. But try as I could I couldn't close it.

I wasn't the only one in difficulty, Daryl was dropped behind me. My eyes were ahead, not behind, though: doing my best to keep a tempo which gave me a chance to close that gap to Shahram.

But I could not. He wasn't gaining on my, but I wasn't gaining on him. We were in stasis.

There's a sweeping right turn where going in you can see riders ahead of you above, then coming out you can see chasing riders below. Going in I saw Shahram, of course, as he'd been hovering ahead of, but I didn't see anyone immediately ahead of him (Brian Schuster was next, and he had a big gap). However, coming out I saw Daryl Spano chasing from behind, not much more behind me than I was behind Shahram. Well, I already knew Daryl never gives up... so this was added motivation: pulled from ahead, pushed from behind.

finishing (Bill Bushnell photo)

But so it remained. Daryl never caught me, but then I never caught Shahram. I hoped, I willed him to crack, just a little. All I needed was a little. But he held his gap to the final rise to the summit. Once I turned there, my last chance to catch him was just ahead, so I spun it up and suppressed the pain and gave everything I had left. Indeed, in those final few hundred meters I closed on him but not enough: he crossed the line 5 seconds ahead. So close. Could I have gone a bit faster a bit earlier? That's hard to say: cycling is such a mental game. But I was wasted at the finish, and I think the better rider won that little battle.

Afterwards I looked more closely at the data, applyin a bike power-speed model to my Garmin numbers. Here's the result, where I assume a coefficient of rolling resistance of 0.4%, a CDA of 0.32 meters2, a drivetrain efficiency of 97%. Note the CDA is a compromise between drafting on the first climb and part of the second, being in a tuck on the descents, and less aerodynamic solo riding on the third climb. The other parameters are estimates. I further estimated my bike mass was 6 kg, my body mass 57 kg, and my clothing and water bottle 2 kg. Here's that result, using 20 second smoothing:

calculated watts versus distance

For reference, here's the altitude versus distance:

altitude profile

On the first climb, which felt so easy, the power is sustained at around 260 watts. In reality is was likely lower, perhaps 250 watts, due to the benefit of drafting (I can emulate this by reducing CDA, for example to 0.28 meters2).

On the descent, power was low, as I was coasting in draft, even though the model shows relatively high numbers. This is because I was in a tuck and so CDA was lower than modeled, and additionally because I was in rider draft much of the time, a substantial benefit because larger riders ahead of me have higher terminal velocities.

At the second climb, power was unsustainable high. This is where the selection was made in our group, and although I gave up hope of following the leaders early, I was still putting in a big effort here. That paid off by putting me in a compatible small group, which gave me draft on much of the relatively short, gradual second climb.

On the third climb is where things unraveled on me, though. Although I was following Shahram, and the gap to him remained stable, it is clear we were both running out of gas here. Had my endurance been better I could have closed the gap simply by sustaining the power I'd maintained earlier in the climb. To some degree I was perhaps paying the price for the relatively hard pace at the bottom of the second climb, but in the end the real solution was more fitness. Practice makes perfect and with my present work situation my ability to train on climbs during the week is extremely limited, while access to climbs on weekends from San Francisco is also tough (it's a decently long ride just getting in and out of the city, and I've not had the time to do that as much as I used to). In any case, I did the best I could.

At the end you can see the little blip at the end where I pushed the pace on the final rise. The model shows this was relatively unimpressive. That's a good thing, because if I'm too strong on the final bit, that means I didn't ride hard enough leading into it. I'd given pretty much everything I had.

In the end, no complaints. I did what I could, and my overall standing in the Low-Key Hillclimbs ended up being tenth, which is very respectable.

Big changes are happening in my work life, and it remains to be seen what my plans are for running and cycling in 2015...

Thursday, December 11, 2014

VeloViewer mileage summary and Strava OLH times

This blog has been neglected due to other activities, in particular the final weeks of the 2014 Low-Key Hillclimbs, but additionally due to a transition in my work life. I've been at my present job starting October 2010 through today, but that will be changing the beginning of 2015.

This period roughly coincides with my time on Strave. For aggregated those data, VeloViewer is a great tool. They just implemented an annual £10 fee to access full data, and it's paid off in this one plot.

2014 started off very mediocrely. I focused on trail running until April, and that hurt my total mileage, but it ended with a very successful debut at the 50 km distance, the Woodside Ramble. Then there was a downtime associated with preparing for and actually moving to a new place in San Francisco. Starting in September, however, I recovered, with 4 very solid weeks end-of-September-to-start-of-October in Switzerland. After that I did fairly well, participating in Low-Key Hillclimbs, and getting in a decent number of SF2G rides. You can see the slope has been fairly good, even if total mileage is well below the level of 2012.

Strava just recently introduced a segment time plot, which is really nice. Here's mine for the "Low-Key Old La Honda" segment. Old La Honda times, mostly riding the Ritchey Breakaway on the Wednesday Noon Ride, have been quite stable this year and last, quite a bit slower than 2011, but what can you do... 2010 was a slow year since I wasn't able to train as much as I'd like, with the work transition.

I never felt at my best on Old La Honda this year. But since my present work plans take me to Berkeley, I don't know when I'll have an opportunity to do the Nooner again. Maybe I need to plan a weekend trip via Caltrain to give it a go when I have some cycling fitness next.

Tuesday, November 25, 2014

Low-Key hits the dirt, part 4: Marin Headlands

Love-hate: my relationship with a mountain bike and Low-Key Hillclimbs.

Low-Key Hillclimbs is a series of 9 climbs each year, on Saturdays from the beginning of October through Thanksgiving, with the final of the 9 on Thanksgiving itself. When we started the series, part of my vision was that we'd use it as a showcase for the variety of climbs in the Bay Area. Back then, in the mid-1990's, racers typically used a low gear of 39-25 or even 39-23. This tended to make the spicier climbs the Bay Area had on offer a bit more of a painful experience than many would prefer. Thus it was usually same-old-same-old: climb Old La Honda, climb Kings, climb Highway 9, West Alpine, Tunitas. But other truly differentiating options went neglected.

So the goal was to broaden people's scope: cover the 2-dimensional space of duration and steepness with a combination of short and long, gradual and steep. In addition to making it more interesting for riders, it would make the competition more interesting, as the riders who excel on steep-long might not be the ones who are best at short climbs.

But the differentiation was less than might be expected. Scores were remarkably consistent from climb to climb. There was a bit of motion, but even with a climb like West 84, which is split into a flat portion and a climbing portion, when we ran split times there was relatively little difference in the rankings on the two.

So to spice things up, we needed to add something more truly different.

In 1998, a flat time trial was on the series. I don't take credit for this one: it was a year I was relatively uninvolved. But the route, Willow Springs, was rained out. So it never got used.

My contribution was the dirt. Dirt adds an entirely different element to the game. You'd think it would be similar: power/mass is the key metric in both climbing and against rolling resistance. But while a dirt climb won't turn the ranking on its head, it does have a much greater influence than switching between paved climbs.

Our first dirt climb was West Alpine Road in 2007. This was our second year back after the break from 1999-2005. It was a mass-start, timed event. We'd applied for a permit to Windy Hill Open Space Preserve. This permit had originally been accepted, but then was subsequently declined when it was realized we were timing riders. Timed cycling events are not allowed on Open Space lands, I was told. I pointed out that given the course, not even world champion mountain bikers could exceed the 15 mph trail speed limit. However, this did no good. To be honest I see their point. Even at a relatively slow speed, if one is in a competitive mode, a rider can behave uncourteously or even dangerously. But as a result we redirected the climb to West Alpine, which is a public road, one of the rare public unpaved roads in the Low-Key domain.

There was one tricky bit in that the road has been washed out since the late 1990's, and thus cyclists are required to use a bypass through open space land, a bypass which would normally be subject to permitting requirements. However, since that bypass is sanctioned for through-traffic on what is still technically a public roadway, and since the law does not forbid cyclists from competing on public roadways (assuming they follow vehicle code), it would have been hard for them to impose their permitting requirements on us.

That climb was a success, with a much-diminished but happy group of 30 riders. I rode Cara's cyclocross bike, too small for me, but good enough for the 18:15 effort. On the singletrack, which had some steep bits, I got off and ran in typical cyclocross fashion but without cyclocross skills. I ended up with 102.1 points, twelfth out of 28 male riders. This was well below my usual standing, as the average of my scores counting toward the overall was 118.8.

Despite the big success in 2007, our fun in the dirt took a hold until 2012. What changed in this time was the proliferation of GPS timing, in particular on smart phones, so that everyone had relatively easy access to a GPS timing device for at least a day. The other thing that changed was Strava established a precedent that simply posting times recorded by riders did not constitute an "event". Nobody demands Strava get a permit every day for reporting times riders took to climb such excellent open space trails as Kennedy Fire Road in Los Gatos. The traditional Thanksgiving ride there, on the other hand, where riders were actively organized and given a time to start together, was shut down.

So we switched to using GPS data for dirt climbs. In 2012 we hit that traditional Thanksgiving climb, used by us the Saturday before, Kennedy Fire Road. I ran this one instead of riding it, foolishly thinking the top was earlier than the actual top and not finishing the climb. I still haven't ridden Kennedy Fire Road. I definitely need to do that.

Kennedy was another success. Some people were deterred by the GPS timing aspect. It lacked the spirit of Low-Key, they said. But not really -- the spirit of Low-Key is to expose riders to different types of climbs, to establish fast times for different climbs, and while the GPS timing took away some of the group aspect, reducing rider groups to whatever riders informally organized among themselves, we certainly accomplished these. Including runners we had 45 finishers at Kennedy. That was solid. By virtue of my mess-up, however, I scored nothing, and indeed this cost me the overall ranking for top runner that year, a year I was focused on running in preparation for the California International Marathon which I ran in December.

With the good experience of Kennedy, I wanted to make a dirt climb an annual component of the series. So for 2013 we added a climb which Tim Clark had recommended: Montara Mountain. This was the most challenging dirt climb we'd done, with steep gravelly rock sections which made getting a road bike up it a real challenge. I used my mountain bike for this one, a 1990's Gary Fisher Genesis-geometry aluminum hardtail which I'd bought used in 1999 or so to ride the trails in Austin. I'd never gotten around to doing that, and only built it up later, getting a 2000 Marzocchi front fork, the first year they offered it with a lock-out dial. I figured lock-out was important to reduce losses from the suspension on smooth dirt climbs. I'm still not sure how true this is, but it seemed a good idea at the time.

It's a nice bike, but it's heavy, several pounds heavier than a modern carbon mountain bike with carbon fiber wheels. And it's a lot heavier than a 'cross bike. But I liked the fat tires for Montara, wanting as much traction as I could get. Even so, I didn't clean it, and unclipped during one of the especially steep bits. My 3x8 gearing is fairly low, but now as low as the gearing most people have with 10-speed or 11-speed cassettes.

I ended up with 98.6 points. I wasn't very fit in 2013 to start with, having only just recovered from a crash in June for which the subsequent physical therapy and associated exercises took a much larger priority on my time than riding, and for much of the time running wasn't an option. But even so, my score was low that climb: on climbs contributing to my final score I averaged 109.8 points. So this was 11 points, close to 10% lower.

The turnout: 51 total runners (6) and cyclists (45), despite a relatively remote start at the coast north of Half Moon Bay. The weather was gorgeous, though, and it was so-much-fun. While there was no organized start time, a bunch of riders started near the usual Low-Key time of 10:00. I started at 10:00 when registration normally closes, others started around 10:10 when we traditionally roll. So there was a decent crowd at the top at one point. It felt a bit like a party, even more than the usual road climbs.

Now it's 2014, and I wanted to add in one of my favorites, the Marin Headlands. I was a bit worried about using the Headlands, because it's national parkland, and higher-profile than either Kennedy or Montara. But the combination of Miwok and Marincello in good conditions is about as well balanced a climb between mountain bike, cross bike, and road bike I know. The trails have some steep bits, a few ruts, but I've ridden them both on my mountain bike and on each of my Ritchey Breakaway. I've ridden nearby Coastal Trail to the Conzelman Road / McCullough Road junction on both of these as well as on my Fuji SL1. So the trails are quite versatile.

For a change, I was planning on using the Winter Allaban for the Low-Key. It made sense: gain the advantage of the fattish tires of the rando bike with its low gears. Sure, the bike is much heavier than my Fuji SL/1, but I didn't want to do a lot of work to get the Fuji ready, and the fatter tires should provide better traction. But it was worst than I thought. On my pre-ride the Sunday before the Low-Key my rear tire, a Challenge Strada Bianca 30 mm (measured @ 32 mm on my 26 mm rims) was slipping occasionally. This wasn't enough to cause any major issues, except on one particularly steep bit on Bobcat returning from Marincello, but skidding is almost certainly a big energy waster, and I figured even if I wasn't skidding overtly small skids I might not recognize as such would cause energy loss I didn't want.

Pre-ride of Headlands route

Kirby Cove side trip, during Headlands route pre-ride

I could have put Cara's cyclocross wheels on the Winter, but I decided instead to go with traction maximus: the mountain bike.

The pre-ride gained less significance, however, as in the days before the climb we got around 2 inches of rain ending just hours before I rode the trail. The trail ended up absorbing most of this, but the surface was left tacky and damp. This likely improved the traction relative to what I'd experienced on the drier surface.

I got to the trail head having ridden there from home. I'd discovered, to my horror, that my front brake was lightly rubbing. The Gary Fisher has J-brakes, and the adjustment on those is very sensitive, since they don't have as much throw as other brake designs. If I change on thing on the Fisher, it will be the brakes. I don't like worrying about them.

I didn't have my multitool, but fortunately Glen Kinion was there, and he led me to his parked car where he had a giant bag of tools. Paul McKenzie was parked nearby, as was Jennie Phillips. It was the Sisters and Misters out for their ride, which wasn't due to start for close to an hour. I adjusted my brakes with Glen's screwdriver, chatted just a bit, but then headed out for my lap.

I left my water bottles, vest, and remaining 2 dates in a bag at the junction between Miwok and Bobcat, then set off for my lap. I was carrying the bare minimum -- only a single gel to consume after Miwok, before descending Old Springs.

Miwok went fairly well, I thought. My goal had been my 10 minute Strava PR, and I beat that cleanly on my lap timer, although the Strava segment I later realized includes a portion of the trail prior to the Low-Key start which had been marked by chalk by Reid the day before based on the GPS coordinates I'd posted. We don't use the Strava segment for Low-Key, but rather I define my own "start line" and "finish line" and time how long it takes riders to go between them. I still beat my Strava time, however, by only one second, but that had been by a good margin my best time, so I'd take it.

Old Springs descent was next, untimed. It felt good to be on a mountain bike here, although having swapped the heavier mountain bike saddle for a road saddle, I couldn't get my butt behind the seat as I normally would. Old Springs has a series of steps and I like being well back for these. But still it was better on the mountain bike than on a road bike. Along the way I encountered Gary and Holly riding the other way, from home near Mt Tam to the start to meet their Sisters and Misters teammates.

Marincello was next. This is easier than Miwok -- a fairly steady grade to a false summit, a gradual sweeping left turn, a final steep bit, then a short gradual slope to the top.

Soon after I started I realized I'd forgotten to take the gel. Whoops. That was clearly nonoptimal but I'd need to live with it.

I started running on fumes when I hit the sweeping left false summit. This was fine because it would mean I'd be sure to come close to emptying the tank on this climb. I kept the intensity up on the sweeping left, then gave it everything for the final steep bit, surviving the final few meters between the end of that and the top.

I PRed Marincello as well, by a healthier 8-second margin. PRs on both climbs? How could I be unhappy with that?

After recovering a bit, finally eating that gel, I descended Bobcat back to the junction with Miwok and found Holly and Gary there. It was 2:05 pm, after the scheduled Sisters and Misters start time, but they didn't see any sign of the others. Had the group started early? That was impossible, I said, since Gary and Holly had backtracked the course in arriving there, and I'd forward-tracked the course before meeting them, so there was simply no way. Some hikers arrived from the direction of the lot where Paul and the others had parked. I asked if they'd seen mountain bikers and they said no. So the three of us decided to start together, Holly for time, Gary just to get the ride in, me for the company and for a second loop.

So after I ate my remaining two dates, put my water bottles back on my bike, and returned my vest to my pocket I was ready to go and we were off.

We were overtaken along the way by the late-arriving Sisters and Misters group. I was clearly sub-optimal for this second lap, having given everything on the first lap. But it was fun giving a hard effort here without any pressure to get a particular time.

I was passed by Greg and Paul on Miwok. After the others rejoined, we descended Old Springs together, me following Paul for a PR on the descent despite going very relaxed at the bottom. The rain really cut down on the hiker population, and we had virtually no reason to slow more than normal the whole way. We regrouped again at the bottom, some of the others not as comfortable with the Old Springs steps, and then went on to Marincello.

On this, my second time up Marincello, I really started to feel the effort of the previous climbs. Greg, Paul, Sarah, and Amy all dusted me early. I felt as if I would regain ground toward the top but it didn't happen. Still, I had my earlier time, so didn't worry.

After regroup and photos, and spotting Bill Bushnell taking photos from above the junction, we descended Bobcat together. It was great riding with the group, enjoying the remarkable weather and fantastic views.

Afterwards, I rode back home, while the others packed their cars. I got home not long before dusk, a wonderful day of riding.

So mission accomplished, right? Indeed, the day was a big success. We had 23 riders, 1 "runner" (Bill Bushnell, who hiked the course), 24 total. That was the lowest turn-out of all of our dirt experiments, but Peninsula riders, who are the majority of Low-Key regulars, are averse to driving to Marin or riding anywhere close to rainy weather. It's unfortunate: a lot of great riding opportunities are squandered by the fear of the wet, fear which was notably absent among riders I encountered when I was in Basel Switzerland.

But I don't worry so much about total numbers: it's quality, not quantity which counts. And everyone who did the ride seemed to have a great time. David Collet remarkably beat Andrew Touchstone's "untouchable" KOM on Miwok, Sarah Schroer and Amy Cameron each beat the previous QOM on Miwok, and I'm sure I wasn't the only returning rider who beat my PRs. But the vast majority were not returning riders. I'd shown them trails they'd not ridden before, and so I felt like the goal had been met.

But personally I was disappointed with my 109.9 points. I had really wanted more than this. Still, it was easily the best of my three dirt rides in Low-Key history, two on the mountain bike, one on Cara's cross bike. I should embrace Low-Key philosophy, and focus on the experience. But why was I slow?

The most obvious explanation is simply mass. THe mountain bike is a full 5 kg heavier than my road bike, around 8% more total weight. But everyone's mountain bike is heavier than their road bike. Glen was on a modified randonneuring bike with cross tires which was a steel brick, yet he climbed exceptionally well. There's more. One is that my mountain bike has 175 mm cranks, which was the style of the time (large cranks for mountain bikes), in comparison to the 170 mm or even 167.5 mm I use for the road. But crank length is relatively unimportant in tests: it's more a comfort thing. I've not used a power meter on the mountain bike (I didn't have time to install the Garmin Vectors without adequate time to let them settle in, which experience shows is needed to get accurate power). But I suspect my power is off a bit. I think my attention is too diverted to picking good lines through the dirt, while on a road bike it's 100% toward the effort. But this is just conjecture. Another factor is simply that dirt climbs tend to favor riders with more anaerobic power, since the climbs involve variable grades, and the climbs in the Headlands are relatively short. I think this is indeed an issue, as I tend to do less well, relatively, on climbs like the climbs we did in the Berkeley Hills this year than in longer, more sustained climbs.

So it is what it is. The dirt has been a success and we'll be back. For next year? I'm thinking Purissima Creek, a personal favorite. Getting there is easier from the Peninsula, a bit tougher from San Francisco, then Montara was last year. It's a longer climb than those in the Headlands, again with a wide trail allowing plenty of room to give hikers their space, and it's also quite manageable on a road bike (although with some steep bits). Maybe I'll try the Fuji SL/1 on that one.

So, in summary, my score on mountain trails:

  1. 2007, dirt Alpine: 102.1 points
  2. 2012, Kennedy Fire Road; DNF
  3. 2013, Montara: 98.6 points
  4. 2014, Marin Headlands: 109.9 points

I know what my goal would be next year.

Sunday, November 23, 2014

Low-Key Hillclimbs week 7: the gate

Low-Key Hillclimbs week 7, San Bruno Mountain East, was a great success.

This was the second climb this year for which we applied for, paid for, and received a permit for riding on roads which passed through parks. In this case Radio Road goes from Guadalupe Canyon Parkway (a 4-lane eyesore scarring San Bruno Mountain from the days when the goal was to completely cover the mountain with housing developments, a goal which has at least partially succeeded, but the road is still way overbuilt) up to the radio towers at the top of the mountain. The road is very lightly used, because not only is the park remarkably lightly used (diminished by too much of the mountains taken up by sprawled housing) but the parking is at the bottom of Radio Road. We were bypassing this parking area by going from westbound Guadalupe, through a fire access gate, directly to Radio on the northern side rather than take the route cars would follow, which is to cross Guadalupe, enter Radio on the southern side, then loop around and back under Guadalupe to return to the north side. The New Years race hosted by Pen Velo goes this way, accessing the main entrance from the northern side of Guadalupe, which it climbs from the east.

For our $400 permit ($5/rider for 80 riders) we asked for only one favor: that the fire gate be unlocked, so riders could pass through without dismounting, which is the normally fastest way up to the summit from the western side. I was promised this could be done, and while I was worried when the gate was still locked when I inspected it at 8:50 am (soon before sign-in started at 9:00), it was indeed unlocked for us just before our scheduled 10:00 start.

Since Guadalupe in places doesn't have much shoulder, I didn't want 80 riders heading up the hill together, so we sorted riders by speed (based on Low-Key points, USA Cycling results if I had no previous Low-Keys, Strava data if I could find that. For riders with points, I used an algorithm which split groups at places where the gap between the maximum points I had for riders was relatively large. This was to avoid the problem that riders extremely close in maximum points would be put into separate groups. The first rider in a slower group would still be significantly lower points than the slowest rider in the faster group. The exception to this was the tandems, who started in their own group between groups 3 and 4. I used a separate tandem group because tandems can be a big advantage to drafting solo riders on a route that has any flat sections due to the tandem's generally higher power-to-cross-section ratio.

So the first group set off, close to schedule, at 11:11 AM exactly, secure in the knowledge this would be a rare opportunity to ride to the top without any cyclocross moves, or without the added distance of looping around through the park entrance, the latter providing the additional delays of a sharp right turn and the chance of cars entering the park blocking the way.

But things went a bit wrong. We were for some reason shy on volunteers this week, an unusual state for Low-Key, and due to confusion over the number of groups, our sole volunteer at the turn-off thought the tandem group, which had started between groups "3" and "4", was the last group. It was not. We still had the sizable group 4 yet to come. Yet he shut the gate and rode up the hill.

A big part of the confusion may have been that there was a 4 minute gap between the tandems and group 4, but only a 3 minute gap between the other groups. The volunteer would have expected to see group 4 coming up the road but they were not. They were still a minute out of sight. This additional gap was likely because one of our two starters was riding with group 4, so had to, in addition to checking in riders, get on his own bike and get ready to start himself. This might have added just a fraction of a minute but he was starting groups on integral minutes so it bumped the gap from 3 to 4 minutes.

Stuff happens. With Low-Key every hill is a relatively new experience, so we don't necessarily get practice on every issue which can occur in any given week. The gate issue was new to us. We'd not worked out a protocol for how it was going to be handled. I'd generally assumed we'd deal with it after everyone had finished. But we should have worked this out in detail beforehand.

The lead few riders in group 4 saw the gate was closed and did the loop to the left. But most or even all of the riders following these riders took the faster route past the gate. However, crossing the gate, unless you're a skilled cyclocrosser, is a relatively slow process. You've got to stop, get off your bike, pick it up, hoist it over the small barriers to the side of the main gate, then remount your bike and re-accelerate. At least one rider I spoke to took an alternate approach of pushing her bike, diagonally and elevated, through the gaps main gate itself, then stepping through herself. I've done all of these things before (long way around, the side of the gate, through the gate) and the side of the gate option is the fastest. But most of these riders hadn't been here before and didn't necessarily know this.

Bike racing has many traditions, but fairness isn't one of them. There's many many examples every week in pro cycling. Weather changes during a time trial, causing later or earlier starters to face slower conditions, riders get held up by crashes in which they had no fault (generally this rewards riders at the front of the pack, but not always), groups are mis-directed by course marshals, perhaps following the route designated for support vehicles, or gates close at level-crossings for passing trains. All of these things happen: avoiding them is part of the luck with is part of the game. In this case, all I could do was to try to level the playing field as best I could, but I wasn't going to level it for everyone.

For example, consider the riders who looped around. They wouldn't have been tempted to do so had the gate been open, so the challenge of following the fastest route was one they faced that earlier groups had not. I had, however, carefully documented and mapped the route, describing that the gate might not be open and that riders should cross it. So while the riders taking the long-way faced an unfair navigation challenge relative to those who started in earlier groups, it was still a navigational challenge they could have been expected to meet.

Riders going through the gate, however, faced two challenges. One was simply getting through the gate. But the second was congestion. If you approached the gate in a group of four, you needed not only get past the gate yourself, but you needed to additionally wait for the three riders ahead of you to get through. The earlier groups with an open gate didn't face this problem.

But the gate came after a substantial amount of climbing and so any groups, especially in group 4 where riders tend to not have the pack-riding experience the mostly licensed racer population of group 1 have, tend to ride more their own pace on climbs than stick together in a tactical mass. Still, some did have the congestion issue. But congestion is a common issue when things go wrong in bike racing, and if I was going to apply the same compensation to all group 4 riders, as I felt I should since it was a mass-start event for them, I couldn't base it on those facing the most congestion. I had to base it on the conditions faced by those with relatively unimpeded access. To do otherwise would be unfair to group 3, for example.

So I had to establish how much time was lost by group 4 riders. One answer was provided by one of the riders: "I was helped by the stop". She felt she'd been in the read climbing Guadalupe, and the rest gained by crossing the gate gave her the strength needed to produce a stronger effort on Radio Road. Of course I didn't believe this actually benefitted her total time. But the point is important: it's what I have called the "elasticity of rest". When you're delayed by a time t, you get back some fraction of that time t from the effect of recovery on the rest of the effort. This was clearly evident, for example, at the trail race I ran at the same San Bruno park a month ago. Going into the water stop, the runner ahead of me blew through without drinking, while I slowed to a walk and drank a cup of carbohydrate solution. Despite the delay, I very quickly closed the gap that formed, refreshed by the few steps of walking.

This is an extreme example: a few steps of walking versus running is a very short rest, and the shorter the rest, the higher the elasticity or fraction of the lost work which is regained (elasticity is the fraction of work retained in a collision, for example of a ball bouncing off a floor, so I use it here to describe the fraction of lost work which is regained).

So I didn't want to overestimate the time lost in crossing the gate. At first, my reaction was 5 seconds is too short, that 10 seconds would be a good amount. A lot can happen in 10 seconds. If you take 10 seconds to run 100 meters, you lose a world-class track and field race. 3 seconds off the bike, 3 seconds across the gate, 3 seconds on the bike = 9 seconds.

But this neglects the loss of momentum. My next level of estimation was to play out the process of crossing the gate. You need to slow, get off the bike, cross the gate, get back on the bike, then accelerate. If I assign 3 seconds to each of these 5 steps, I get 15 seconds. That seemed more reasonable.

My first check of this was to see how the new adjusted times for the overall climb varied with group number. Here's a plot:

Note the times increase with each group, as expected, with group 1 the fastest, then group 2, then a big increase to group 3, then finally a smaller increase to group 4. From this plot there's no indication of any issue. If anything group 4 is doing better than expected based on the trend established by groups 1-3.

But I met some very sharp resistance, in particular from one rider, when I proposed 15 seconds. "A joke" is how he described it. So I knew I needed to do more: I needed to use a more quantitative approach.

So I defined a Strava segment for the gate crossing. This included the approach to the gate, the crossing of the gate, and the short stretch following the gate. I couldn't isolate the gate because Strava's segment matching is too loose. When I tried to place the end-points too close to the gate I got matches from rider data which didn't even cross it.

My first approach was to take two riders who had been assigned a finish time of within 2 seconds of 18 minutes, one in group 3 (Brian Ward), the other in group 4(Scott Byer). If my adjustment was correct, then these riders should have been of very similar speed. Indeed, Scott Byer, the group 4 rider, took 16 seconds longer on the segment than Brian Ward. To within my target precision of +/- 2.5 seconds, this was in agreement with my 15 second adjustment.

But this was just one isolated example. I then considered my own times on the segment. I'd ridden through the gate when it was open on my way to the summit immediately ahead of group 1 so I could record finishers. Previously I'd ridden the segment ten times, in each case crossing the gate. Some of these times were more rushed than others, on only a few efforts was a really making a maximal effort to get through as quickly as possible. My top three times were 13 seconds, 17 seconds, and 19 seconds slower than my time had been with the gate open. So this also was consistent with a 15-second adjustment.

But I wanted to go further, and look at aggregate statistics. So first I plotted time through the segment versus group number. Group 1 is the fastest, group 2 next, then group 3 and finally group 4 (ignore tandems here). I'd expect that for this segment, the same ordering would apply. So I wouldn't want to adjust group 4 times to match group 1 times, and I wouldn't even want to adjust group 4 times to match group 3 times. I want there to be a steady progression of times from groups 1 to 4: no disproportionate jump from group 3 to 4, but an increase in time nevertheless.

When I did this, I saw there was generally a 5 second increase in times of the fastest riders in each group, except for groups 3 to 4, where the jump (without any correction) was greater. No surprise there: the gate was shut for group 4. br>

This provided a quantitative approach to setting the adjustment. With 5-second precision, I was able to get a close to 5 second jump from groups 3 to group 4 with a 20 second, rather than a 15 second compensation. Here's the plot:


One thing you see here is that the spread in group 4 is larger than the spread in group 3. This is to some degree expected due to the nature of statistical distributions of riders and the fact group 4: group 4 is a "catch-all" group which includes riders of a broad range of fitness and goals. But the fact remains that congestion was a factor here.

I plotted versus group because initially I didn't have a way to plot versus rider time, but when I wrote the Perl code to do that, I got the following plot. Here I show time through the gate segment versus total time. If the compensation was perfect than I'd see no substantial difference in how riders from group 4 line up versus how riders in groups 1-3 line up. Of course, faster riders overall tend to be faster riders through the gate, as well.


As you can see from the plot, there's again more spread in group 4 times. But if anything, with the adjustment, the best group 4 riders now did better through the gate than the best riders from group 3, adjusting for total time. There's several riders who did a relatively long total time but had times through the gate, after the adjustment, competitive in group 3. The red dots in this plot were no group (tandem riders).

This doesn't address the question of recovery. If riders delayed at the gate were able to as a result climb a faster Radio Road, canceling some of the time lost at the gate, this analysis would miss that. One way to check for this would be to compare times on Radio Road to times on Guadalupe Canyon, omitting the gate crossing, for groups 3 and 4. If riders in group 4 benefitted from recovery at the gate then they would do better on Radio Road relative go Guadalupe than the riders in group 3. However, there's confounding factors, like the tendency of riders to try to follow a quick pace set by leaders, and the presence of faster leaders in group 3. So I'll leave it as is. If anything the addition of this sort of analysis would reduce the adjustment of group 4 times, and I'd prefer to give them the benefit of the doubt because the error was ours.

Wednesday, November 5, 2014

San Francisco: E and L both lose

Well, the voting is done, and the results came out largely the same as I had hoped.

In California, 47 passed, as I'd hoped, but so did both 1 and 2. 1 was a water bond, which focused on building dams which don't do anything to either create more water or conserve what we have, the obvious approach the latter in a state which does little to reduce water consumption by either agriculture or individuals. 2 was for a "budget stabilization" account for California. Maybe it will help. 47 helped reduce sentencing on minor theft (<$950) and drug possession. That's a good thing, and I supported it.

Locally, things also were generally good, except I was disappointed that Proposition E, the "sugary drink tax", got "only" 55% of the vote and for new taxes 2/3 is required. This is a bit silly. Consider the case where I have two measures, one which raises revenue via taxes, the other which spends the same amount. The People, being fiscally responsible, vote for either both or neither. 60% vote yes, 40% no. The result? Deficit, since the spending measure passes and the revenue fails. This is a result nobody wanted. So as a result we end up funding everything from bonds, which require only a simple majority, and on those we need to pay interest. It's really a bit of a joke.

So no soda tax, but no change in transportation policy either, something which measure L would have endorsed. It looks like L got less than 38% of the vote. That's a resounding reinforcement for Transit First, and should be the nail in the coffin of pro-car policies like free Sunday Parking or low car registration fees, both policies endorsed by our fine Mayor Ed Lee, who was perversely endorsed in the 2011 election by the San Francisco Bike Coalition. The victory for sanity was even greater if you consider what was written on the ballot was a highly misleading representation of the content of the measure: whether the city should "change" its policy on parking and streets. Had this been all I'd seen I'd have voted yes -- I want to see a change, a change back to metered parking on Sundays, a change to bus rapid transit, a change toward pedestrian malls, a change toward a more aggressive expansion of the bike network. Despite this obviously intentional deception, The People have spoken: it's time to stop pandering to the whiners and move forward with a complete streets agenda.

Monday, November 3, 2014

Nov 2014 election: San Francisco ballot measures


San Francisco virtually always has a complicated ballot on election day. This year holds its own in that regard, with 12 ballot measures as well as local offices. Here's some of my picks for the Tue election.

Here's a guide to San Francisco ballot measures.

District 8 Supervisor

Scott Weiner's my choice. He's had an excellent record so far, including solid positions on transportation.

Proposition A

This is a bond to fund transportation costs in San Francisco, including road repair. Well, for one thing we just recently passed a bond to fund road repair. against my support. Funding ongoing repairs with debt is a bad idea because it incurs interest on top of principle costs, and since paving the roads is a steady expense, there's no reason it shouldn't be allocated out of the annual budget. The main reason, though, I'm against this is that Mayor Lee recently ended sunday street parking metering (every other day is metered). This was going to funding MUNI, so in effect the MUNI component of this bond is going for free parking, not MUNI. I vote no until Sunday parking fees are re-instituted. Ed Lee further killed a restoration of the vehicle license fee in San Francisco. There's no reason to make it cheap to own a car in San Francisco, and giving away parking is bad in every way.


Proposition B

This is Scott Weiner's proposal that MUNI funding be proportional to population.

I have mixed feelings about MUNI because I think it is largely a money pit. On the other hand, I support robust public transit. Early in San Francisco's history, pre-1906, public transit was largely a free-enterprise deal, with private companies providing street car service, which was generally excellent. Now transit is provided publicly, and by any reasonable estimate, it's both terrible in quality and high in cost. I personally get very little out of MUNI service, as I avoid it if there's any other reasonable alternative. And in the free market, alternatives have emerged where they've been able to given the highly regulated market of transportation services. For example, Uber and Lyft and ZipCar are all services which have been enabled by the internet. These are small vehicle services: for large vehicles there's the tech shuttles and IT buses which serve commuters. MUNI, in contrast, is amazingly inefficient, with convoluted routes and stop schedules overburdened by NIMBY "every stop is sacred" politics because if you eliminate even one stop, despite the availability of stops one short block both up and down the street, then some grandma with her walker is going to have a hard time getting to the bus. So the service becomes virtually useless.

So it's chicken and egg. Do you throw more money at the problem, since without money it won't improve? Or do you embrace the far more efficient free-enterprise solutions?

With this amendment MUNI's budget will go to $256 million by 2015, up from $245. This is close to $300/person, or $1/day/person in the city.

So I'm on the fence here. Honestly I can't see more money making any difference to MUNI. I'd much rather see money go instead, for example, to the bike share program which is currently crippled by lack of scale. I really need to see a sign MUNI is going to lift itself from the present morass. The contrast to what I saw in Basel recently couldn't be more striking, and it wasn't just a budget issue.

I'll vote no.


Proposition C

This is to continue funding children's services for the next 25 years, since the similar proposition funding the past 25 years is expiring. The definition of "children" for this purpose is being extended from 18 to 24 years old.

With a 25 year record to stand on, you'd think there'd be more facts available in the campaign material about how effective the last proposition was. Of course educating children is important, but I don't see where these funds are going. Despite this, however, I can't find anyone except Starchild making an argument against, and Starchild's main argument is you can't trust government generally and that the extension to age 24 is unjustified. These are valid points, of course, but I'll vote yes. This is the first ballot measured I've considered so far, state or local, where I'll vote yes.


Proposition D

I have no idea why this is on the ballot and wasn't resolved directly by the Board of Supervisors. Do Board of Redevelopment members deserve retirement benefits? I don't know. I'll vote no.


Proposition E

Tax on sugar-sweetened beverages. Yes, yes, yes, and yes.

This bill isn't perfect but few bills ever are. A similar tax has been highly effective in Mexico City, and the campaigning against this one has been misleading and cynical. Proposition E is a resounding yes.


Proposition F

This is a proposed increase in the height limit for development in the Pier 70 region of Bayview-Hunters Point. Obviously with an increasing Bay are population and a drive towards more centralized, less sprawl-oriented housing model, this proposal makes a lot of sense, as well as providing needed economic stimulus to a traditionally undesirable, crime-intensive area of the city. The list of supporters is impressive. Opposition is dominated by the anti-density, pro-parking crowd. I vote yes.


Proposition G

This would provide a surcharge on the transfer taxes paid on certain multi-unit buildings sold within 5 years of prior purchase. It's designed to discourage "flipping", which is perceived to be a negative influence on the housing market. But often "flipping" means buildings in poor condition are improved, then sold for a profit, which improves the general quality of the housing stock in San Francisco, as well as the general quality and attractiveness of the neighborhoods in which the buildings belong. And the amount of these surcharges is extraordinarily high -- over 20% in some cases. What's the economic effect of this? Suppose I have a near-vacant building for 3 years and I want out for whatever reason. Should I sell it to someone willing to make a long-term investment in the building, increasing the benefit to the housing market? No -- I'll sit on it for 2 years until I'm clear of the surcharge period, so it sits underutilized during that period assuming I lack the resources to properly renovate it. Any abrupt tax like this is distortionary, and while I support an incentive towards longer-term ownership, as this provides more stability, I fear that this one goes too far into the zone of stifling renovation. I vote no.


Proposition H

This would require certain fields in Golden Gate Park be kept natural grass without artificial lighting. This sounds good in principle but the reality is that the net benefit may be higher with an artificial field, and artificial lighting vastly increases the time window during fall and winter when the parks can be used. So I'm voting against this one. In any case the argument that grass fields are environmentally friendly, with their heavy requirement for water and fertilizer not to mention gas-powered mowers, is extremely simple-minded.


Proposition I

This would facilitate the city's ability to renovate fields, including the use of artificial turf. It's a direct contrast to H. I vote yes. It is not a bond nor does it directly increase taxes or spending.


Proposition J

This would increase the minimum wage in San Francisco to $15/hour. Obviously this provides a barrier to hiring entry-level workers. At some point the decision is made that I will not hire someone rather than hire them because the cost of hiring is increased. Minimum wages and their impact on hiring and the economy has been a favorite subject for economic debate for many decades. If the price competition is primarily from local businesses, for example a coffee house, and if demand is relatively inelastic, for example people are addicted to coffee and will not switch to making their own, then businesses will not appreciably suffer by paying their employees more and passing on the cost. It can be argued that the free market has resulted in the cost of labor being higher than this minimum wage anyway, and if that's the case, this isn't needed. I'm a bit on the fence on this, so will vote no.


Proposition K

Prop K will make it city policy to increase housing supply, including "affordable housing". This is a complex subject, including the effectiveness of "affordable housing" policies, but I in general support more, denser housing, so I'll support this. Really it doesn't do anything other than state a priority.


Proposition L

How many times can I say no? This is the initiative to reinforce the emphasis on private automobiles for transportation policy, calling it "balance". Prop L would be a disaster. It's the antithesis of the "transit first" policy which at least nominally has set San Francisco apart from other West Coast cities since the 1970's. Every step of the way toward support of pedestrian-cycling-transit friendliness versus parking-and-driving focus has been fought. If you vote for no other reason, if you vote on no other measures, go to the polls a and vote against Proposition L.



I'm friendlier to the San Francisco propositions than I was to the state propositions. I was against all of those except for 47. Here, I'm for C, E, F, I, and K (5) while I'm against A, B, D, G, H, J, and especially L (7). I'm big pro-E and big anti-L. San Francisco Bike Coalition is for A and B: they're for all propositions spending more on transportation. So I disagree with them there. The one I'm closest to supporting which I am against is J. But when I'm not sure I tend to vote no, unless the opinion for is overwhelming.



Sunday, November 2, 2014

California Propositions


Election time, so time to make my choices on ballot measures.

Proposition 1

The arguments against this bill are compelling. First, a focus on surface water storage via the construction of dams doesn't create any more water: it allows more water to be drained from our otherwise overtaxed rivers. Second, the cost of the bond come from the general fund. Costs of water storage should be paid through water use fees. The real issue is water consumption needs to be reduced, and it's simple economics that be incentivized via increased costs. Subsidizing water use doesn't accomplish this at all. So I am a skeptic about this bond justifying it's in excess of $7 billion cost. California needs to reduce it's excessive rate of water use, and this doesn't address that.


Proposition 2

This is supposed to be a "rainy day fund" for the state. Unfortunately, when there's "rainy day funds" there's the incentive to conclude a lot more days are rainy than sunny. A similar "rainy day fund" was passed in 2004, and it resulted in a general defunding of public schools. I don't trust that this one will be any more successful at keeping the state education system sufficiently funded. I'll vote no.


Proposition 45

This would require the state insurance commissioner to approve changes in insurance rates. It would also forbid credit history or history of insurance coverage from being used to determine eligibility or rates. I certainly agree with that, and I think the argument that this creates "costly new bureaucracy" is likely over-stated, but I fail to see an adequate case presented to the voters that this measure is effective in reducing health care costs. It's really too complex for the amount of time most voters can deliver to it. I vote no by default: the burden of proof is on the initiative to prove its worth. In this case, health care is already highly regulated, with existing programs and reviews, and the implications of voting for this are insufficiently clear.


Proposition 46

This is labeled as a requirement on drug-testing doctors. It seems like a good idea: surgeons operate on your brain, you don't want them dosed out on painkillers or whatever. But this is just a front. Nobody really cares about drug-testing doctors. It's really about raising the pain and suffering cap from $250k to over $1M. That's where the money is. I'm all for reimbursements in the case of negligence but excessive pain and suffering awards have a distortionary and negative effect on the health care system and on health care costs. I vote no.


Proposition 47

This reduces certain crimes like drug possession on "petty theft" (theft of something which has a value of no more than $950) from felonies to misdemeanors. Obviously we put too many people in prison for too long for the wrong reasons. We need to retreat from the "throw away the key" mentality which led to the fiasco of "3 strikes". I vote for this one. Note my Trek 1500 was purchased for $800 but all of my other bikes are beyond the "petty theft" threshold, but this would include phones which are very commonly stolen unless the content of the phone is assigned a monetary value, which seems unlikely.


Proposition 48

This would allow an Indian casino near Mono Lake. I vote no.



I vote no on all state propositions except 47.

Old La Honda: Chasing Mark

After last week's Powertap snafu where I tried pacing myself off the power meter only to later realize it was substantially over-reporting power, I wasn't feeling super-warm-and-fuzzy today about following the same approach in what was essentially a mandatory new attempt at the Wednesday Noon Ride. Using the calibration cycle on the Garmin Edge 500 had appeared to restore the Powertap to the regime where it was able to stay in tune during coasting phases alone, and the numbers I typically saw in the display were more typical of the pre-Switzerland mediocre Dan than the "suddenly blessed with amazing fitness" Dan the Powertap had been assuring me had replaced it.

Chris Evans and Brian Schuster of Squadra SF were at the start, two very solid climbers, especially Chris who'd been putting in an impressive string of mid-16-minute Old La Hondas with his approach of starting each week with a leg-ripping effort which I can only imagine matching for more than a mercifully small number of seconds. So as far as I was concerned, Chris was of no relevance. More relevant was Mark Johnson, who'd been putting up some low-to-mid 18-minute times on a regular basis. I certainly hoped to be able to stick with him. A big stretch was to move into the high-17-minute range, which is my known record on the Ritchey Breakaway. I didn't want to think about that. But I certainly had hopes of an 18:30.

I'd been discussing pacing strategy with Chris, noting the exceptional nature of his starts. He said he'd experimented with more measured starts and they didn't allow him to recapture the time boost (or is it distance boost?) he got from going hard at the beginning and then tapering back to what he can sustain. My assertion, and he agreed with me, is riders with exceptional top end power (or work capacity) can do a harder start than those with more limited reserves above threshold. I agreed from my own experience that there is no recovery from a too-slow start. A large fraction of that time is gone for good.

We hit the base and boom! Chris was up the road, rapidly disappearing around a corner. At the front of the remaining crew, Mark and Bill Peucel were chatting. Bill's been riding very well lately, but I knew I didn't want to be climbing at my chatting pace, let alone his which was a bit slower. So I passed them and set off at a moderately aggressive tempo.

Last week I'd been in the 36/23 most of the climb and I wanted to stay out of that gear, shifting into the 36/21 instead. I set a PR in 2009 riding exclusively in the 36/18, with the Fuji, reaching the top in 16:49. I later broke that (in Dec 2011), but I certainly knew that I wasn't going to be spinning anything close to 2:1 today. But if 36:18 is 16:49, then at the same cadence a 36/21 is 19:37. I certainly expected to be considerably faster than that. But recently, since returning from Switzerland, I've found my desired climbing cadence has been higher. Maybe it's because my legs are tired from a relatively heavy load of riding and cycling and that forces me to seek refuge in higher cadences, which result in reduced muscular loading.

So in my 36/21, I surged past the two leaders. Mark followed. I basically ignored him, keeping the cadence on my gear. However, I started to feel this pace was becoming burdensome, especially during the steeper of the positive grade undulations. Eventually I was forced to relent, downshifting into the 36/23 I used last week. Glances at my power meter, which I was trying to ignore, weren't encouraging.

Mark took the lead here, simplifying my pacing task from this point. It was simple: stay with Mark. So I got onto his wheel and tried my best to not think about anything else. Eventually the top would arrive.

Mark does this climb almost every week, and he's a confident guy, so knows how to pace himself. The pace never dropped so far I felt as if I wasn't being inspired to go my best. Often, following others, I feel like when they move to the front they lose motivation and the effort drops too far. This wasn't the case with Mark. Sitting on his wheel had become a real challenge. I found myself letting little gaps open on the steeper bits, using the more gradual bits to reclose them. All that mattered to me here was his wheel.

Around a corner, a large truck approached, leaving us maybe a 1-meter ribbon of road, slowing but not stopping. Mark slowed a bit as he passed, but I slowed a bit more. So leaving the truck behind he'd opened a significant gap: maybe 5 meters, and he wasn't going to wait around for me to close it back up.

But the reduced pace passing the truck provided recovery, and as the top approached soon after, I accelerated in the low gear to try and pass him by the stop sign. I failed, but I had at least reclosed the gap which had opened passing that truck.

Here's the power data, comparing running average to previous climbs:


And here's smoothed power versus distance:


This was an exceptional week for pacing in that I spent much of the final half of the climb glued to Mark's wheel. Pacing strategy was thus nullified: I follow Mark. If Mark slowed too much, I'd have taken the lead, but I was at my limit just following him. So micro-analyzing my pacing from this point is meaningless. I was on Mark's pace, not mine, taking advantage of a few % power discount for drafting, depending on the wind speed and direction.

The numbers: 272.95 watts, 18:43.38. Meh. I was 274.0 watts on 11 June, albeit with a time 13.27 seconds slower. My other powers this year, last week excepted when my Powertap was high on life, were three out of four within 2.3 watts of this value. My time was best of the year by 8.31 seconds, the previous best on 18 June. I am maybe 1 kg lighter than I was then, explaining the quicker time in conjunction with the time I spent glued to Mark's wheel, but meh. I was hoping for more from my trip to Switzerland.

That said: I did a 20 mile run on Sunday, approximately double my typical "long run" distance these days, running only approximately once per week other than a few blocks of consecutive days in Switzerland. I was definitely still feeling this going into the ride, and had definitely felt on my quick-pace ride into work the day before.

One interesting aspect of the ride was cadence. Here's a plot of cadence on the climb:


I spent much of the time over 80 rpm, which is unusual for me: historically I'm more likely to be in the low-to-mid 70's.

Gear selection can be calculated as cadence / (speed × development), with units appropriately matched. I get good numbers assuming a development of 2.100 meters. Here's the result:


I started with a 36/21, which I'd normally feel comfortable in the whole way up, except half-way I retreated into the 36/23 for much of the time, except for a few times I upshifted on flatter bits. Gear selection stopped being an issue once I let Mark take the lead. I simply shifted into the gear in which I felt most comfortable riding on his wheel. It's different when doing the pace oneself, in which case you're picking a gear which maximizes speed. I think my preference for a low gear, high cadence, was due to fatigue.

So I have some headroom on Old La Honda. How much? I don't know. I need to avoid gaining back any of the mass I lost in Switzerland, for one thing. And I need to keep up the intensity and consistency in my riding, two factors which had been notably missing for much of this year, including the months preceding my trip in September and early October.

Tuesday, October 28, 2014

Low-Key Hillclimbs week 4: Berkeley Hills

As we descended back to the Peets Coffee which marked the start of the week 4 Low-Key Hillclimb route, Rich Hill said to me: "has Low-Key ever done a day with 4800 feet of climbing before?"

I paused. Hamilton? Not quite, only 4400. Diablo? Not close. Portola Valley Hills last year? No -- less than 3000 total feet. What about weeks with optional extra climbs, like the Diabolical Double or the Lomas-Marin Ave combo? No and no.

"I don't think so. No -- I'm pretty sure not," I replied.

I felt good about this, because after 6 intense efforts up climbs with vertical gain from 476 to 764 vertical feet, I was cooked.

Details of the route are here. Paul McKenzie designed a route of absolute brilliance, tying together a combination of classic and obscure climbs in the maze-like Berkeley-Oakland-Hills with a minimum of overhead. Of those 4800 vertical feet, 3825 were against the clock, 975 part of the untimed transitions (still timed, actually, but with a 1 hour + 3 min/km time limit which was plenty for relaxed regroups plus a dawdling recovery pace from one to the next). The goal was to have riders repeatedly pummel themselves with one hard effort after another, not stopping until 6 challenging "short climbs" were done. And while I call the climbs "short", that's "short" in comparison to Old La Honda, the canonical middle-length climb, but still long enough to earn a Tour de France cat 4 or even cat 3 designation.

Pacing in a route like this is an exercise in macro-versus-micro. There's pacing within each climb: go out super-hard at the start then try to hold on, or target a steady effort? I went for more of the latter, in deference to the macro-pacing aspect: the week wasn't just a single climb but six, and while there's recovery between them, that recovery is obviously only partial. Yet I had enough faith in my recovery, given my combined running and cycling volume over the past two months, that I wasn't going to hold back too much on the early climbs to save anything for the later ones. Better to push it hard, then again, then again, then again.

And this worked. I faded a bit on the fourth climb, not having started the ride with enough calories (oral surgery has me on a soft-food diet, and all I had was a half-pack of Sharky chews consistent with that, although I probably should have used some sugar solution in my bottles, which I generally avoid due to the difficulty getting the valve adequately clean). But then after climb 4, we reached Paul's parked car which he'd stocked with chews, gels, and cans of Coke. A 1/3 can of Coke mixed with water + a pack of Clif Blocks was a big boost, and I was able to make a good effort up Wildcat, climb 5.

After Wildcat, though, I felt done. If this was an interval session, at this point I would certainly have headed home, feeling I'd done an excellent day of work. But this wasn't an interval session. It was Low-Key. Stopping simply wasn't an option. We still had South Park...

So I recklessly threw myself at the South Park climb. There was no reason not to. This was it, the last climb of the day. And with Patrick Gordis (who had skipped two of the intermediate climbs and so was no longer officially part of the "event") dangling in front of me I had plenty of motivation. I wanted, waited for him to crack, even ever so slightly, and I'd have him. But although he slackened his effort just a bit when the road leveled out partially before the final kick to the top, allowing me to reduce the gap, I simply could not catch him. But it didn't matter, really. I'd given everything, drained the tank which I'd thought had already been drained. It was a good, solid effort when there'd been nothing left. Now I was truly, legitimately, done.

This feeling of digging deeper than I had thought I could was intoxicating. For me, the day was a success.

Interestingly the ranking on each of the climbs was virtually identical. People would move up a place or down a place, but with the exception of the Quarry-Volcanic climb #3, which ended in muddy dirt which some people handled better than others, taking any one of the climbs versus their sum would have produced very similar standings and very similar scores. So was the day a waste? You see this sort of argument all the time, for example related to Tour de France stages. "Short stages produce the same results, which are dominated by the final climb, so why waste the time, effort, and expense of longer stages?" This misses the point. It's not just the result, but how you get there where the value is. It's about the story. The story of week 4 of this year's Low-Key Hillclimbs was a relatively unique one, rivaled only by last year's Portola Hills route, which comprised much shorter efforts. Certainly no other week in the 2014 series will be the same.

Here's my VAM from the ride:


I've plotted VAM (rate of vertical ascent) versus time for portions where VAM exceeds 300 m/hr. Tunnel is a relatively gradual climb so VAM is relatively low due to power given up to wind and rolling resistance, with the exception of one surge. On this climb, I was riding with Bill Laddish and Robert Easley most of the way up, the last time on the day I'd be able to stick with them. The pace early was sluggish, so I took the lead, a step I knew was a tactical mistake, but I wanted to establish a solid pace to break the group up. This worked, with only Bill and Robert following, until they took over. I was following them until a glance at the altitude profile on the Edge 500 showed we were what appeared to be quite close to the summit, so I ramped up the effort (see the spike). Unfortunately there were still 2 km to go: the Edge had spontaneously zoomed out, as it likes to do, and without having given adequate attention to the axis scales I thought the summit was a lot closer than that. This surge, however, inspired Bill and Robert to keep up the pace, a pace I could not sustain, and I finished alone in 3rd. This was my best result of the day.

Thorndale was next. This one went very differently, as I'd had my jacket on for the descent from Tunnel and when I stopped at the base of Thorndale to remove it, nobody else paused. I was thus completely at the back. This wasn't too bad, though, as the climb was so steep, with essentially no car traffic, that I was able to pass through the group without much delay. You can see my VAM is nice and solid here, in part due to the steepness of the climb, but demonstrating I had no significant issues with bike traffic. With the grade, any advantage from drafting a rider who'd started in better position was negligible, especially since we had what should have been a tailwind at this point (although despite the weather forecast for high winds, I didn't feel much).

After Thorndale, some of the rain which had been forecast arrived, as as we waited at the top, a light cool rain fell. But it wasn't much, and none of us got wet, even those of us without jackets. Once again I put on my jacket, knowing for Quarry-Volcanic we'd be forced to pause at the start to cross the gate.

By the time we got to that gate, the rain had already abated. There was the question, though, about what the condition would be of the dirt section which finishes this climb.

We began on pavement, though. The start was ragged, with riders heading out one or two at a time, me close to last. It was another steep one, albeit without the sustained steepness of Thorndale, and I didn't see a disadvantage in starting relatively late. Soon, however, we reached the end of that. First there was gravel, but that didn't last, and then we were on dirt. The dirt had a thin layer of mud on the top, enough that my rear wheel would spin with any choppiness in my pedal stroke, or if I let my weight get too far forward. Between watching my pedal stroke, keeping my weight back, and picking a line through the mud I slowed considerably here, and two riders I'd passed early re-passed. This was to be my worst result of the day.

As we finished, riders ahead were unclipping to open a gate, through which they passed. This provided access to a lot from which one could admire the view, which with the low-hanging clouds was worthwhile. The foot-assisted passage through the gate clogged up my mud-intolerant Speedplay pedal-cleat combination, however. I missed the Bebops which now live on my Ritchey Breakaway: they have the float, low stack, and relatively light weight of a Speedplay (at least the Speedplay stainless steel spindle version: the Ti-spindled Al-bowtied spindles on my Fuji I was riding here are lighter), but have no issues with modest amounts of mud. As we descended I couldn't clip in, but we paused again at the gate at the bottom during which a combination of water remaining in my single bottle and scraping with a sharp rock cleared things out well enough for me to get my shoe into the pedal once again.

I was feeling a bit depleted here, as I noted, my pre-ride supply of Sharkies now long since gone. But I knew that Paul's aid station was after the next climb, so I figured I'd be okay. Running has taught me that the typical cyclist addiction to constant water + calories tends to be overdone.

Given that this climb wasn't particularly steep, 10% sustained on Fish Ranch, my VAM here is holding up fine, despite my lack of calories. When we emerged from Fish Ranch onto Grizzly Peak the grade leveled out considerably, to more like 7%. My VAM dropped here until a final effort where I got it back up to around 1450. On this climb I benefitted from using the Garmin course navigation feature where it provides a list of upcoming waypoints with the distance to each: it was the first climb of the day where I used this. Since I knew the finish was essentially at the intersection of Grizzly Peak and Lomas Cantadas, I knew at a glance how far that was. However, this would stop working before the next climb. I've had this mode fail several times before. If freezes, failing to progress. I really don't know why -- one of many little bugs in the otherwise very useful Garmin Edge 500 navigation feature.

After a much-needed break at Paul's excellent aid station, the Lomas Cantadas descent was taken slowly due to its steep turns and the wet foilage on the road from the recent rains, extending into the night before. A faster group had taken a slightly longer route than my group did, however, and we merged at the point where our paths converged.

Two more to go...

Our first climb of the day, Tunnel, is a very popular climb although I'd done it only once before. The next three were fairly obscure, such that even the local riders hadn't all done them. But Wildcat Canyon, our next climb, is one I'd done many times before. It starts steeply enough, nearly 8%, but then levels out substantially for the long run to Inspiration Point, where we were finishing. This made it probably the most tactical climb of the day, with only Tunnel close. It was very important to start in good position then to stick with a good group on the steep portion to assure a descent draft on the gradual section. A small gap at the top of the steeper bit could explode if one were to miss the train.

Climbing South Park with Rich Hill (Paul MacKenzie following us took this photo)

I started fairly well. Bill Laddish and Robert Easley, another rider who like Patrick Gordis hadn't done the full route, took off. I thought my position at the bottom had been descent but I missed this move and I wasn't going to get it back. That was fine, though, as I was in a group with Patrick, Bob Gade, and Robert Easley which was working together very well. Well, by "working well" I mean Patrick put in a killer pull up the whole bottom portion of the climb before the other two shot off together while I retreated to that special place we go when we're just trying to suppress unpleasant human body feedback. I ended up pulling Patrick over the first portion of the flatter section before he said "I'll take a pull", came past, and dropped me. From here to the end was just an exercise in perpetual suffering until I was reprieved by the arrival, at long last, of Inspiration Point and immediately prior of Paul's green line.

I was done, simply done. But being done wasn't an option, so I joined the others for a brief respite before South Park.

For South Park the plan was far simpler: start hard, stay hard, and hold it to the finish. I wasn't going to pay any attention to what others around me were doing. After a relatively late start, trying to postpone the inevitable, I supposed, I eventually caught and passed Rich Hill among other riders, then saw Patrick again just up the road.

I waited for Patrick to relent, just a bit, to open a crack I could exploit. But he simply would not relent. Finally the road leveled out a bit, the calm before the terminal storm, and there I saw my chance. Patrick didn't slow: his speed increased with the decreasing grade, but his effort was clearly off a bit here, and I was able to upshift and reduce the gap. But then the grade increased again and Patrick's focus returned. I couldn't reduce the gap any further and then we were done.

From the VAM plot I'm pleased to see I was able to hold a decent effort over this sixth and final climb of the day. My final surge on the last steep portion wasn't much of a surge, but finally a last effort to sustain the unsustainable pace I'd been holding. In the end, given what I had, I seem to have nailed the pacing on the day fairly well, even if tactically my performance may have been mixed.

But it was an excellent day, and an excellent course design by Paul. I'm not sure what we'll do next year but whatever it is will have a very difficult time living up to this one.