Looking at Eugene Wei’s Invisible Asymptotes

Invisible asymptotes, a recent blog post from Eugene Wei dissecting social networks, product design philosophy, and the celling for Amazon, is tremendous. I spent 7 years in Vietnam growing multiple social networks to millions of users, and thus developed a philosophy about how people engage with each other changes generationally and how new social networks emerge. Wei’s insights are spot-on (it’s true that I say this because his opinions largely match mine, but his are much better articulated and detailed).

It’s an incredibly long article, but well worth reading. Some of my highlights:

Defining invisible asymptote:

For me, in strategic planning, the question in building my forecast was to flush out what I call the invisible asymptote: a ceiling that our growth curve would bump its head against if we continued down our current path. It’s an important concept to understand for many people in a company, whether a CEO, a product person, or, as I was back then, a planner in finance.

On Twitter:

In the case of Twitter, I think the theory is wrong. Given the current configuration of the product, I don’t think any more meaningful user growth is possible, and tweaking the product as it is now won’t unlock any more growth. The longer they don’t acknowledge this, the longer they’ll be stuck in a Red Queen loop of their own making.

On Snapchat and generational communication preferences:

I’ve written before about selfies as a second language. At the root of that phenomenon is the idea that a generation of kids raised with smartphones with a camera front and back have found the most efficient way to communicate is with the camera, not the keyboard. That’s not the case for an older cohort of users who almost never send selfies as a first resort. The very default of Snapchat to the camera screen is such a bold choice that it will probably never be the messaging app of choice for old folks, no matter how Snapchat moves around and re-arranges its other panes.

More than that, I suspect every generation needs spaces of its own, places to try on and leave behind identities at low cost and on short, finite time horizons. That applies to social virtual spaces as much as it does to physical spaces.

On Facebook:

This is one of the diseconomies of scale for social networks that Facebook is first to run into because of its unprecedented size. Imagine you’re in a room with all your family, friends, coworkers, casual acquaintances, and a lot of people you met just once but felt guilty about rejecting a friend request from. It’s hundreds, maybe even thousands of people. What would you say to them? We know people maintain multiple identities for different audiences in their lives. Very few of us have to cultivate an identity for that entire blob of everyone we know. It’s a situation one might encounter in the real world only a few times in life, perhaps at one’s wedding, and later one’s funeral. Online, though? It happens to be the default mode on Facebook’s News Feed.

On Instagram:

There is a purity about Instagram which makes even its ads perfectly native: everything on the service is an audio-visual advertisement. I see people complain about the ad load in Instagram, but if you really look at your feed, it’s always had an ad load of 100%.

I just opened my feed and looked through the first twenty posts, and I’d classify them all as ads: about how great my meal was, for beautiful travel destinations, for the exquisite craft of various photographers and cinematographers, for an actor’s upcoming film, for Rihanna’s new lingerie line or makeup drop, for an elaborate dish a friend cooked, for a current concert tour, for how funny someone is, for someone’s gorgeous new headshot, and for a few sporting events and teams. And yes, a few of them were official Instagram ads.

On understanding the root customer need from data and feedback:

True, it’s often difficult for customers to articulate what they want. But what’s missed is that they’re often much better at pinpointing what they don’t want or like. What you should hear when customers say they want a faster horse is not the literal but instead that they find travel by horse to be too slow. The savvy product person can generalize that to the broader need of traveling more quickly, and that problem can be solved any number of ways that don’t involve cloning Secretariat or shooting their current horse up with steroids.

Intuition “genius” vs data – no one you know is a genius, but a bunch of people you know think they are:

Lastly, though I hesitate to share this, it is possible to avoid invisible asymptotes through sheer genius of product intuition. I balk for the same reason I cringe when I meet CEO’s in the valley who idolize Steve Jobs. In many ways, a product intuition that is consistently accurate across time is, like Steve Jobs, a unicorn. It’s so rare an ability that to lean entirely on it is far more dangerous and high risk than blending it with a whole suite of more accessible strategies.

Just like athletes, there is a prime in one’s professional life. Michael Jordan can’t play in the NBA right now at aged 55  – just because someone had great product insight ten years ago, this doesn’t mean they have it today:

The reason I recommend a healthy mix of intuition informed by data and feedback is that most product people I know have a product view that is slower moving than the world itself. If they’ve achieved any measure of success, it’s often because their view of some consumer need was the right one at the right time. Product-market fit as tautology. Selection bias in looking at these people might confuse some measure of luck with some enduring product intuition.

However, just as a VC might have gotten lucky once with some investment and be seen as a genius for life (and the returns to a single buff of a VC brand name is shockingly durable), just because a given person’s product intuition might hit on the right moment at the right point in history to create a smash hit, it’s rare that a single person’s frame will move in lock step with that of the world. How many creatives are relevant for a lifetime?

Jeff Bezos talks about how every day is day one – every new day, there is a baseline to move forward and redefine what you can do. Continually win customers, don’t defend (lock them in) them from competitors.

You can’t overserve on user experience, Thompson argues; as a product person, I’d argue, in parallel, that it is difficult and likely impossible to understand your customer too deeply. Amazon’s mission to the be the world’s most customer-centric company is inherently a long-term strategy because it is a one with an infinite time scale and no asymptote to its slope. …

By discovering their own limitations early, they are also quicker to discover vectors on which they’re personally unbounded. Product development will always be a multi-dimensional problem, often frustratingly so, but the value of reducing that dimensionality often costs so little that it should be more widely employed.

Game Planning with Best Bike Split for the Tierra Bella Bicycle Tour

I’m a regular listener of the TrainerRoad Ask a Cycling Coach Podcast. In it, the hosts discuss using Best Bike Split quite a bit in preparation for their events, particularly races. Best Bike Split can take your riding profile, including your power and aero profile, event goals, and course map to create segment-based power recommendations.

I was really interested in trying out the service for Tierra Bella (more on that experience here), and the free version will let you do power-based goals. Paid versions let you to optimize for time and speed-based goals and really drive down into the customization.

In my plan, I set an Intensity Factor (IF) of .75 (actual result was .72) as my goal.

Based on this, BBS predicted:

  • Ride time of 7:14 (actual: 8:13)
  • Net Power of 175 (170)
  • Variability Index of 1.05 (1.14)
  • Training Stress Score of 405 (430).

I didn’t know how well I’d adhere to the plan, so my approach was to spend most of the time in aero down in the drops to get some (I had set up the BBS plan to simulate me primarily riding on the hoods) free speed. The plan called for me to be around 160-170 watts most of the time, which I felt I could maintain in the drops. I would have been unable to maintain 200 watts in the drops for longer periods, however, due to being much more compressed on the bike (legs get closer to the stomach, making breathing more difficult).

Even though I wouldn’t be putting in max power (and thus not maximizing air resistance benefits) on the ride, spending a lot of time in aero position would have cumulative effects.

Why was I so slow compared to Best Bike Split?

There are definitely a numbers of factors. I was able to maintain BBS’ power goals on flats and moderate climbs pretty easily, but I was not as good on sustained climbs and descents. BBS would ask for around 100 watts on descents, not much, but as I’m a shaky descender who didn’t know the roads and who also wanted to avoid the hoards of cyclists climbing the other way (I was on small lane roads with below-average road conditions), I focused on safety rather than speed. I am guessing BBS may have an option to tweak this. On sustained climbs, I found that getting up to 90% FTP after miles 50+ was getting tougher and tougher, especially as the weather moved from 43 degrees F (at 7AM) to a peak of 95 (during the Hicks climb) without airflow. I had to be careful about not overexerting myself – I didn’t want to bonk (run out of energy) with 40 miles to go.

Because I didn’t overextend myself on the climbs and didn’t work on the descents, I was pretty confident I had some energy left to use after reaching the last rest stop with about 12 miles left in the 124 mile (200KM) ride. I did these last segments with 15% higher power than BBS’ goals, making up for some of the lost intensity from the previous 100 miles.

If I look back at my other metrics, this is what I think happened. Overall, my per minute intensity was less (.72 vs .75 IF) than projected, but I worked for much longer, leading to a higher total stress (9 vs 8 hrs, 430 vs 405 TSS) score. Since I had descents of essentially zero effort but then tried to make up for these at the end of the ride, this led to more variability in the power, leading to perhaps less efficient use of the power in terms of speed. I lost out on a lot of potential gain by not putting power on the descents, but also taking them relatively slowly. For example, the Henry Coe descent is 10 miles.

There’s one stretch in the ride that I think can serve as a good example of what BBS thought would happen, and what actually did. BBS does consider weather conditions such as wind direction based on historical data, with premium membership including more data.

From the Henry Coe descent on to Bailey / Morgan Hill, Best Bike Split gave me a 10 mile flat segment of 156 watts on the hoods. I actually spent the entire time in aero.

BBS Prediction Wind

Best Bike Split Prediction (#102)

BBS Prediction

In the first image, you can see that Best Bike Split was expecting a tailwind (see lower right corner of image), which explains why I’d be doing 20.78 MPH at only 156 watts. Keep in mind that this prediction was for me on the hoods, sitting more straight up with increased air resistance, not in the drops (aero) position.

Strava Output

In reality, looking at my Strava record above, I was consistently over the goal – my guess is I was around 15W above for the 10 miles and still 3MPH below BBS’s expected speed, again despite being in aero. This was a 5 minute loss on this segment alone.

I was able to integrate the BBS plan with my Wahoo Elemnt Bolt; one problem I noticed is that despite having the power levels pop up on screen, I would not get directional information on the display. Thus, cues would be for new power outputs, not road directions. I’m not sure if the original map for Tierra Bella is missing this information, but my guess is that BBS overrides it. As a result, you have to use the Elemnt’s map display – it took me a bunch of wrong turns on the course before I realized this. The disadvantage of having the map display on all the time is that you have access to fewer data fields. There could be ways to get both power and directional cues, but I will have to look into that more next time.

Overall, I liked working with the BBS plan. It gave me power levels to focus on while giving me the confidence that I wasn’t overexerting myself. I knew I could last the 125 miles.

Best Bike Split costs $20 per month or $120 per year to be a premium subscriber. Since I am not a racer and don’t do many events, it’s not worth subscribing, but I’d be open to paying per event (ex. $5-10 per event with more features than free but less than full premium).

The Tierra Bella Bicycle Tour – 6745 Calories over 200KM and 8825 Feet of Climbing

Last weekend, I completed the Tierra Bella Bicycle Tour, organized by the Almaden Cycle Touring Club (ACTC). I achieved new highs in distance (124 miles versus 92), climbing (8825 feet vs somewhere in the 7K range), and paced things fairly well with the help of Best Bike Split.

profiles-2017.jpgAbove: the official climbing map for my route. I’m not sure why ACTC’s numbers are so different from mine (600+ feet difference in climbing).

screenshot_20180414-182515.png

Watts/KG for the ride (in Net Power) came out to around 2.7. Starting right before 7AM, I crossed the finish line around 9 hours later, having spent an hour total at rest stops and traffic lights.

Screenshot_20180414-182543.png

With an average heart rate of 146 across the 8+ hr ride, this was 83.4% of my max heart rate (175), or on the lower end of Zone 4 / Lactate Threshold.

Overall Impressions of Tierra Bella

The tour is a great value. It’s fully supported in terms of food, with 6 rest stops and a post-ride meal for the 200KM route (technically, there’s 5 rest stops, but one repeats), and the cost is only $75. If you want to add someone for the post-ride meal (for example, my wife who did not ride), that’s another $10.

$75 is about two-thirds to one half the cost of a normal gran fondo-like event, and if you live in the area (Morgan Hill, Gilroy, San Jose), it’s an easy 30 minute drive to Gavilan College and the start line. I came from Mountain View, which is about an hour away, close enough that I didn’t need to sleep at a hotel overnight, saving further on costs.

While Tierra Bella is a fully supported ride, I’m not sure how they handled mechanical issues. Unlike the Sea Otter Classic Gran Fondo, there doesn’t seem to be support from Velofix, a mobile bike shop, or other support staff driving along the route. Last year at Sea Otter, I ran into two flats within the first ten miles.

Because of this, do not expect any frills (finishing medals, beer, celebratory finish line, photographers to sell you photos, etc.), which was perfectly fine with me. The ACTC staff is super nice and was genuinely fun and encouraging to talk to at the rest stops.

I wasn’t that fond of the route overall. A lot of this might have been that I was familiar with a good chunk because it crosses back into Morgan Hill, an area I’ve ridden quite a bit in the past. I like these big touring events mainly to experience new roads, with new sights. It felt like there were three or four main highlights on the tour, between which you had to criss-cross 15 miles each way to get to them – these “commute” miles felt a bit empty and boring, but again perhaps I felt that way because I knew the roads.

Despite all the heavy climbing, none of the descents were particularly good. The initial descent, the Canada loop, was likely the best, while Henry Coe suffers from poor roads (super jittery on my 25 inch wheels at 90 PSI) and small lanes (someone cracked his one week old carbon handlebar on the descent), and Hicks is just too vertical to enjoy (I had to worry about others struggling uphill wile going down and carefully control my speed).

Nonetheless, considering everything, Tierra Bella is a solid event.

Breakdown of the Ride

I was in bed by 8PM the previous night, but didn’t get very good rest (this has been a lifelong problem when I have to get up early for something important). I was up before 5AM and got to the start location around 6:30AM. There wasn’t a mass start, and so I was off right before 7AM.

See a Strava Flyby comparison of others who did the same route here.

Things were going well after the first 20 miles or so, I was gradually moving past other riders at around 160 watts. I don’t remember much from the Canada Loop, and I skipped the first rest stop. Before the Henry Coe climb, however, I got pulled over by police by skipping a stop sign. I couldn’t protest it at all, and just admitted to the failure. Luckily, I was let off with a warning. I also learned that if I had received a ticket, that would have counted as a moving violation on my (car) driving record.

By the time I got to the second rest stop, I knew it was time to eat and went for some wraps and fruit. The Henry Coe climb was a big pain, however. I hadn’t studied the map beforehand, so the whole course was a big surprise. This is my fault. Henry Coe is a 10 mile climb that lasts over an hour at 3W/KG. I just felt like I was going forever, and I became mentally tired after finishing a mini-crest and then seeing another mile of climbing up ahead. I hadn’t planned on taking the third rest stop, but I wanted the break after such a sustained climb.

Descending from Henry Coe was a bit troublesome as the road pavement quality isn’t very good. I didn’t have to worry about cars much, but I had a jittery ride all the way down and had to constantly look for bumps and potholes.

Making my way from the descent, I crossed over west to Bailey Road in Morgan Hill, enduring a 10 mile flat stretch in a headwind.

I was definitely feeling a bit of fatigue by the time I hit the fourth rest stop, and I really went after the food here, getting Oreos, Gatorade, an instant cup of noodles (first event I’d seen this, but hot soup felt really good in my stomach), and other items. In fact, I overdid it, and was starting to feel stomach cramps shortly after leaving on my way up to Hicks. Thus, I relaxed a bit on the power, and hoped I’d feel ok by the time I needed to make the climb. I noticed I was able to make huge gulps of my Gatorade / Cytomax drink mix, so I was a bit concerned I might be getting dehydrated. At the same time, I started to feel twitchiness from my left calf muscle, as if it were cramping. I’ve only experienced a muscle cramp (in any activity) once in my life, so I wasn’t sure, but this was another sign that I needed to drink more, not push pace, and pedal in a way that would not exacerbate the issue. I was unclear if there was a way to pedal without making it worse, but I was determined to keep going.

I guess the drinks and food all got absorbed in time, because both the stomach and muscle cramps were mostly gone by the time I arrived at Hicks.

Hicks was an evil climb considering my state. It’s only 1.5+ miles but at a 10 degree incline with significant parts over 15 degrees in the heat. I hadn’t felt too bad in terms of heat before this, but once I started going uphill, the air flow died around me and I could feel heat radiating from the ground. I later found out it was 95 degrees on the climb. I felt like I needed to do 3W/KG (200 watts) just to marginally move up the climb and I was scared of standing up because I didn’t know if I’d blow up. If you had read my mind then, you would have heard a lot of cursing.

At this point, I was 85 miles in and I had serious thoughts whether I could finish it.

But I did, and the rest of the ride downhill and back to Gavilan College in Gilroy was pretty much okay from there.

In thinking about future events, I think 125 miles was a bit much. When I finished, my lungs were a bit tired, which I’m not sure I’ve experienced in the past. I think I would have been fine with 10K feet of climbing and 100 miles total distance. At a certain point in a ride, you just want to go home, and I was definitely less enthused about the ride once I finished the Hicks climb (around the 95 mile mark). However, I realize this may have partly been due to the course map, knowing that I wouldn’t be seeing anything new the rest of the ride, and the high heat.

Parc des Princes (Paris), Camp Nou (Barcelona) and the European Football Stadium Experience

Photos: First two rows of Paris / Paris Saint Germain / Parc Des Prices. Bottom rows of Barcelona / FCB / Camp Nou.

About four years ago, I finally had a chance to watch a football match in Europe, watching a 2-2 tie between Real Madrid against Valencia. The match itself (I remember Real as a bit sluggish) wasn’t that great, but this Cristiano Ronaldo goal was pretty nice:

What I remember from that match is that the Santiago Bernabéu is an old, non-modernized stadium hosting one of the five most valuable sports teams in the world. Surprisingly (or alarmingly), smoking near the seating area was pretty common and the view really wasn’t so good (I paid the relatively reasonable price of $75 or so for nose-bleed seats). My wife Ha remembers nothing from this match, but a couple of months later she got to see Zidane (before he became coach of Real Madrid) there in the equivalent of an old timer’s charity match for less than $25 (I was jealous).

In planning for our Europe trip this past March, I really wanted to watch more football. I’ve seen Manchester United twice in exhibitions (once vs PSG and Zlatan Ibrahimovic in Chicago, the other last summer vs Real Madrid without Ronaldo and Ibrahimovic at Levi’s Stadium in Santa Clara), and to see them in England is still a big to do for me.

Instead, we were fortunate to see Paris Saint Germain vs FC Metz in Paris, and then Barcelona vs Chelsea in the Champions League (think playoffs) round of 16 at Camp Nou.

The Matches

I knew the PSG match would be a blowout – Metz is one of the worst teams in Ligue 1. Unfortunately, Neymar has gotten injured a couple of weeks before the match (months after I bought the tickets direct from PSG) and Cavani was suspended. Kylian Mbappé, the next great young “chosen one” at 19 years old played ok, but despite all this PSG won 5-0.

Ha and I were in row 1 (technically the second row) in the center of the lower deck. The tickets were about $120 each and all considering, a great deal. Since we were close to level with the field, it was a different view from above, but you could truly get a sense of the speed of play and the level of skill, even from Metz. The crowd atmosphere was just ok and the 48,000 seats in the stadium did not look anywhere near sold out.

One thing I learned is that after matches, home players go to the goal side housing their supporter groups and thank them for their support. Supporter groups chant and sing literally non-stop throughout a match.

For the Champions League match, it was a thrilling experience to see Lionel Messi in his prime play in a high stakes elimination match. Messi responded by scoring his fastest goal ever in less than 3 minutes, then followed that up with an assist and second goal.

Easily the best soccer experience I’ve had, even after paying around $250 per ticket to sit in Lateral 3 (good viewing angles, but fairly high up) on Viagogo, a secondary market for tickets. The face value was around $170, and I think I could have bought tickets for the match direct from the club a week before the event. I didn’t know this would be possible however, so I bought tickets months beforehand just to make sure we’d be able to go.

The European Stadium Experience

Parc des Princes and Camp Nou, like Santiago Bernabéu, are older stadiums. All three remind me (not in a good way) of Candlestick Park in San Francisco, which has now been replaced by the fabulous AT&T Park for the Giants and Levi’s Stadium for the 49ers.

The San Jose Earthquakes MLS team play in Avaya Stadium, which was completed in 2015. Despite housing a mediocre soccer team and holding only 18K people, Avaya is a much better experience than any of the three European stadiums I’ve been to.

How so? Huge high resolution big screens are in all modern American stadiums and arenas, but not in Parc des Princes, Camp Nou, or Santiago Bernabéu. I have good stadium-alternatives food options with food trucks at Avaya but Aramark (popular provider in the US as well) does the food at Camp Nou; our sausage / hot dog was plain bad (not worth finishing). If I remember correctly, there were perhaps 3 different food items one could buy in the whole stadium. 3! The food options in Parc des Princes was about the same.

There is no in-seat food delivery as you might get at Levi’s. Whether you think this is good or bad, there’s no alcohol served either. More on that later.

No ads, no extra information, not even helpful overlays on the big screens, no music, no one hawking snacks in the stand. From a purist perspective, I guess you could say you’re just left alone to enjoy football – in some ways the TV experience for football is the same. That can be a good things, but there’s just a lot of money left that these teams should be and can be making without ruining the spectator experience (see Earthquakes).

These are all just reminders of how American sports, particularly in the last 20 years, has become so competitive from a business perspective, innovating (nowhere near as fast as startups, but clearly fast relative to Europe) to capture more value from the modern sports fan.

Getting back to our hotel in central Barcelona from from Camp Nou was good – a number of different train lines connect at different stations nearby. Worst case, Camp Nou would have been just an hour away walking. Getting back from Parc des Princes was a mess, however. Google and local bus stop information suggested a bus we could take back to our hotel. That bus did not stop at the designated stop. It stopped at another stop (like a mini bus depot) very close by with other buses and just did nothing. I believe it was at least a half hour before any buses left, and all the while, we had no idea what was going on. Bus drivers didn’t know or could not say, or there were no signs that suggested an answer. We would have taken the subway, but the lines just to get into the (one) station were completely packed.

Alcohol and Hooliganism

One of the biggest surprises in European soccer is the lack of alcohol sales at the stadium. I don’t really drink, so it’s not such a big deal, but on the business side, this seems to be a lost opportunity. Then again, I’m not a fan of drunk fans either. I tried to think more about why this is the case, and my guess for it is hooliganism, a term used to describe disorderly, violent or destructive behavior perpetrated by spectators at association football events.

English football in the 1980’s was notorious for this; it was just dangerous to attend games.

During the Barcelona-Chelsea match, I witnessed hints of this after Barcelona went up 2-0, and the Chelsea fans in attendance became upset at a potential missed call. In another section above (and behind us) I had seen a plexiglass-like transparent barrier dividing the top-most section and the one in which I was sitting. I realized that opposing teams fans sat there, but I wasn’t clear why the barrier was needed. I had never seen that at a sporting event before.

After the missed call, I started hearing constant loud bangs on the “glass”. I then started seeing flying objects fly overhead (and likely hit fans). These were coins being thrown by Chelsea (a London-based club). What was astonishing to me is that there was no security to warn fans or to watch the fans doing this.

It was almost everything was perfectly all right and expected. In the US, you see warning messages on the big screens and around you to text message security in case anyone is being disruptive. You feel like you have an outlet in case you feel unsafe. Not the case here.

Then, when you think about not selling alcohol to fans, it does make more sense.