Back On The Horse

Ironman 70.3 Santa Rosa
Santa Rosa, CA ★ 2017.05.13

After drifting away from the rather pro forma “I swam; I biked; I ran” race reports, I’ve found myself coming back to these sorts of analytical post-race assessments. While a bit of humor or philosophy is a welcome change, I think that write-ups like these are more useful for other folks, and they are also probably more useful for me to write. I suppose at least some of it is that, for the past several years, my results and performances have been less impressive than I’d hoped, and not being analytical is a way to deflect that. But other than a misfire in Texas a few weeks ago, I feel like this year, both in racing and training, I’m back to training and racing more in line with my own expectations.


Garmin was nice enough to provide me with one of their new ForeRunner 935s for an upcoming Slowtwitch review. It’s the first good GPS watch that I’ve felt comfortable swimming with, so I decided to use it to try to collect some data on things like:

  • How fast I start
  • My stroke rate
  • How my stroke rate changes
  • How my speed changes

You can see the Strava file HERE, my first ever Strava swim file from a race.

Screen Shot 2017-05-14 at 11.44.33 AM.png

This is graph here is from TrainingPeaks. The green line is average pace. I started the watch a bit before the gun went off, so that I wasn’t worried about that. I never really felt great, and in fact, felt better and better as the swim went on. I was very surprised to see that my stroke rate stayed very constant at 70-72spm, which is a bit lower than I’d hope for, but that’s normal in a wetsuit swim – longer strokes and lower stroke rate – versus a non-wetsuit swim. I was also pleased to see that I was able to negative split the swim. And not fade too much.

It should surprise no one that I really don’t have any start speed. This has always been – and continues to be – my weak point on the swim. At this point, I’m probably never going to “develop” this. I had a bit of it when I swam with Simon Whitfield’s group in Victoria, but I lost most of that after my accident and was just too far removed from regular, competitive open water swims with ITU athletes. But the same reason I struggle to really “rev it up” also serves me well in terms of being extremely steady, which is a boon for most parts of long-course racing. Just not swim starts. Such is life. As I said, if you could drop me into the front pack at 500yds in, I bet I could stay there no trouble. The problem is that they are gone by 50yds into the race.

My splits by 500yd (autolap) were:

  • 500yd – 6:09
  • 1000yd – 6:21
  • 1500yd – 6:23
  • 2000yd – 6:01

It was an interesting experiment to have a watch that I felt comfortable using for swimming, and I value the data I’ve gotten. Not sure if/when I’ll do another non-wetsuit swim this season, but if I do, I’d be interested to see how the data compares.

I knew the Santa Rosa swim would be fast with a small – but very fast – group with Andy Potts and Sam Appleton pushing the pace hard, and Jarrod Shoemaker and perhaps some others helping to keep the pace solid for a small group from the get-go. Given that, I came out with the group that I expected to, though further towards the back of that group than I hoped. I was hitting this race on the build, and I didn’t expect to be razor sharp, so I was happy to simply have a good swim, as I wasn’t expecting a breakthrough race. Especially after a disappointing swim in Texas, I was glad to swim well in spite of not feeling great.

It’s always nice to surprise yourself with a breakthrough, but surprises can also go the other way, and if I had to choose, I’d choose to have a performance that was in line with expectations every time. And my swim in Santa Rosa was in line with expectations.


T1 wouldn’t normally merit a mention, but T1 in Santa Rosa was a very long run up a blacktop hill that was especially tough coming out the swim. The hard part with long transitions like this is that you can lose a lot of time. Alternatively, you can also blow yourself out by hammering too fast. Ultimately, the blacktop was the biggest limiter to my speed, as it beat up my feet and really prevented me from running all that fast anyway. If there’s one thing I hope they change for next year (or for this summer’s 140.6), I hope it is full carpeting for the run and for most of transition. This run was just really hard on the feet. Especially given that it was in the 40s (Fahrenheit) at race start.

 I was about 60s slower than the fastest athletes over T1, which is a lot. Too much really. But I also chose, with the very cold weather, to put my bike shoes on in transition. I was worried about trying to manage getting into the shoes with cold hands. Ultimately, I don’t think this was the right decision. It was not very humid and felt warmer than it was, and I think I could have easily gotten into my shoes in the initial stretch. Given that I only finished about 9sec behind 6th and a little over a minute behind 5th, I’m inclined to see what I could have done better. And certainly transitions was one of them. Especially T1.


The margin of error on the bike has shrunk a lot. When I first was winning races, most people made terrible decisions. Making generally good decisions – like simple wearing *an* aero helmet instead of a road helmet – was enough. Simply having a good position and staying in it was enough. Simply using latex tubes with good tires was enough.

If I rode harder – or as hard – than most people, I’d ride faster.

But not anymore. Now most people have good positions. Now the fact that I’m 6’3″ is problematic because I can never be as compact as someone who is 5’10. Everyone uses aero helmets. Most helmets are pretty good. Everyone wears clothing that fits. Some even have custom-made and tailored gear. Everyone has a pretty good position. And some have great ones. A waxed chain might have given you an advantage at one time; now it’s a requirement. I wax and powder my own chains, but whereas that used to give me a feeling of some free speed, now it’s just because I can’t give that up.

At least in part, I hope that over my career, I have helped to usher in the era that is now causing me issues. I’ve long espoused the importance of bike fit, equipment selection, aerodynamics, and more. And now the bar has been raised. If I derive some satisfaction in getting smoked on the bike, it’s only because I maybe helped sow the seeds of getting my own ass kicked. Bike companies now put time and effort into making sure their athletes make good decisions about equipment and have good fits. Which is a great thing. It does, however, make racing a lot harder…

There was a time when no way I’d put out more power and go slower. But that’s changing. No… It has changed.

The nice thing is that so many athletes are so open about sharing power files. Here’s my Strava file. And you can compare to a lot of others who raced. Sam Appleton, who won the race, and Justin Rossi, an outstanding pro-cyclist and time-trialist turned triathlete, gave me incredibly good benchmarks. I could see that when it came to pure power – on the climbs – I was right there with them. But on the flats, at equivalent effort, they were going about 1kmh faster than I was.

Some of that is bike handling skills. I’ve always struggled to find a good rhythm on the rough roads in Sonoma County. I’ve never ridden as fast on this course (or the old Vineman course) as I thought I should have. There’s just an optimal line, and I never seem to find it. For instance, Justin Rossi – who went 2-1/2 minutes faster than I did on about 10w+ less power (and he’s heavier than I am and is about the same height) – had 250% more time (almost 10min vs 4min) at 0w than I did. He just did a way better job of carrying his speed.

But even on the flats, he pulled away at lower power. Rough math, I figure I was giving up about 15w+/- to Rossi in terms of drag. I believe that this is a mix of three primary ingredients:

  • helmet choice. I really like the MET Drone. But I’m not sure for me it is the best helmet. Lots of guys have tested it and found it to be extremely fast. But in my case, it just doesn’t seem to be the right helmet.
  • clothing. I am not sure that over-tops are optimal aerodynamically. It’s just tough for a sleeved top to fit really well, especially when you’re pulling it on wet, and trying to get it pulled on appropriately. For an Ironman, the sun protection might be worth it if you – like I – prefer to run in a normal tri-suit. But I think normal sleeveless tri-suits or sleeved tri-suits are the right choice. I don’t know that, at least for me, the over-tops are optimal. Without being a one-piece suit, fit issues just become magnified.
  • hand-position. I’ve done some look backs at old photos. My fundamental position is basically the same as it ever was. But I think my bars are flatter – meaning my hands are lower – than they were. And that’s a potentially significant source of drag.

In the past, I might have given up 15w or so to my own “optimal” set up, but that was a fraction of what other guys were giving up relative to their own optimal set-ups. Not anymore. Now other guys are inching closer and closer to their own very best position and equipment. Especially for halves, I think that guys are able to squeeze out extra speed from positions that wouldn’t quite work for 112mi. But they can make it work for 56, especially on a course with some hills that allow them to get up and out of the bars.

I plan to do some testing on my own before Victoria 70.3 to see if I can’t eke out a bit more speed. On the bright side, the reason this whole dilemma showed up is because I finally rode the sort of power for a half that I expect to. The data from the climbs shows that my powermeter was reading pretty much correctly on the day, and I do my own weighted calibrations with my 20kg+/-5g weight. So the data for the day was good data.

And it’s been a long time since I’ve ridden 323w/330norm for 2hr+. With almost no fade. I rode the first hour at 329w/337norm. And the second hour at 320w/326norm. Hard to ask for much more than that. This was a good ride. And if I had to choose between needing to figure out what equipment I’ll use versus how I’m going to get more fit, it’s a whole lot easier to simply change helmets…

Screen Shot 2017-05-16 at 10.11.16 AM.png

I plan to document my testing and what I find, whether on my own and/or with Jim Martin at ERO, which is pretty close by.


I actually executed reasonably well here. I lost about 10-15s, so obviously could have been better, but after a hard ride, I would say that I was doing the Ironman shuffle when I started running and, as the largest race in North America, T2 was quite a long run. I don’t expect to catch up to the guys with ITU experience in transition skills, but with a couple more halves on the calendar, I need to practice doing a quick and simple transition when I’m not staring a marathon in the face.


To a certain extent, I paid the price for the hard ride. And for exactly having the smoothest preparation leading in. This race was very much a “Plan C” (or D or F) after Texas, and so I didn’t expect to smoke the run. This is also a tougher-than-it-appears run course with a lot of false flats. At one point, I was sure that I was running uphill both ways. At my best, I feel like I’m a 1:15:XX runner. Certainly 1:16:XX. I barely broke 1:18, and I struggled mightily to try to reel in a fading 6th place. But my legs were shot. I never really found what I’d say was a good rhythm, and it was forced all day. To that end, being able to force a 1:17:55 is not bad. Though I’d certainly have taken the same effort for I’m capable of a 1:15:55 on that course with proper run preparation.

Strava’s analysis tool is a nice way to break things down. Of all the features Strava offers, I find this to be among the most useful. Splits by 5km:

  • 5km – 18:05 (where I’d hoped to be)
  • 10km – 18:35 (still right on)
  • 15km – 19:07 (uh-oh, that’s not good)
  • 20km – 18:42 (okay but…)

Screen Shot 2017-05-16 at 10.09.31 AM

I was able to bring it down a bit near the end, but it was never smooth. And there was a pretty strong lull in the middle where I really slowed. The 935 also has optical heart rate, which – once you find out how tightly to wear the watch – works pretty well. It was nice to have HR data for the run, and the data shows that I definitely maxed out what I had. 171bpm avg is great, and a maxHR of 189 reflects what I certainly felt in terms of draining the well on the day.

Wrapping Things Up

All in all, I was pretty happy with the race. I think my best execution on the day, I could have snagged 5th. But of course there are also a lot of things I could have done differently to have finished way worse than 7th. As always, upside is most often much more limited than downside. The race showed what I knew – that I’d done good training leading into Texas and that I was on track for where I should have been to start the year.

In hindsight (20/20, of course…), this should have been my plan all along. Start the year choosing from the multitude of great halves and get my racing legs back after a rocky 2016. Thankfully, though, when Texas didn’t go as planned, I didn’t dwell on it. I didn’t get caught up in sunk costs. And I made the right decision to get right back into racing here. With a clear head and good nights of sleep leading into the race – I was asleep naturally at 8:15pm on Friday night and up before my alarm at 3:15am on Saturday morning, I was able to perform. I stayed the course, and I’m glad I did.

I had fun racing. I wanted to be out there. It was a great day. Some things to work on, but all simple and positive stuff. Looking forward to another chance at it in three weeks.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s