Race report: Zuidwolde 5k XC

I was feeling a bit iffy on Friday, and it was the first really bad
weather day after more than two weeks of uninterrupted sunshine, but I
still couldn't resist giving my legs an airing. So I went out while
it was pouring and did a few easy laps in the park, finishing off with
relaxed strides. The evening was weight training, with just a few
ba-only lunges and donkey calf raises for the legs.

Come race day, the weather had cleared up somewhat. I decided to do a
5-10min warmup at about race pace, with a short pickup at the end. My
calves were still not 100%, but nothing too bad.

The field was a bit sparser than last year, where the top man had done
18min something for the 5k. But that was OK, I don't like crowds :)
The place was Zuidwolde - almost 2 hours away from where I live - so
people there were excited that a greek would come all the way there
just for a run. In fact, I was really tired of running in Amsterdam's
forest and there aren't that many 5k events around (everybody seems to
be running the marathon these days). This felt like the right time to
do a 5k, so I had planned it way in advance.

The route was I guess a typical XC route for Holland. There was a
tiny hill (maybe 10-15m high?) with forest, criss-crossed with paths
going up and down. Most of the paths were wide enough for 2 people,
sometimes for even 3-4. I started at the back, thinking I'd use the
qpeople in front to pace myself. I would stick behind somebody, then
drop them as soon as I was feeling sufficiently confident. In the 3rd
km I was behind one guy going slightly slower than me on
the downs and ups but I restrained myself and only passed him at the
4th kilometer, which I did in 4:09. Here are the splits:
4:30, 4:25, 4:45, 4:37, 4:09... total time 22:25 (race time 22:28).
I came in 6/15... so at least above the middle of the pack. First was
18:56.

Full results here:

http://www.uitslagen.nl/uitslag?id=2008051700743

I guess my next target pace would be 4:20, as I felt quite fresh at
the end. I am not sure that in this particular race I could've
pushed the first few km more, but I could've definitely started
picking up the pace a bit earlier in the last km and/or done a faster
sprint for the last 400m. On the other hand, this was consistent with
the times in my long intervals (1-2k) last month of 4:36/km - and
somewhat better than my winter ~23min time trials.

Training week ending May 4 and Shimano Fiets Challenge race report

Week ending May 4
==================

Getting back into my weight training routine, plus a chance to bike a bit!
There was a great race scheduled for saturday, in pretty much the only hilly place in
Holland. Details at the end of the rerport.

In view of the bike race, I tapered the leg weights for the legs this week.
I eased slowly into my previous weights.


Time Type Volume Time Details
28 Apr ---- OFF ----------------- Travel
29 Apr AM Run 3.2 km @ 50 2.4k + 400m@84 + 2x(20,40,60)[2']
29 Apr PM Weights 15.4 Mg @ 60 3x5@90% Upper body, 3x8@90% Arms, 3x12@90% Legs
30 Apr ---- OFF ----------------- Queensday!
01 May AM Bike 20.0 km @ 60 Easy-medium rolling hills
02 May PM Weights 14.5 Mg @ 60 3x6RM Upper body, 3x10RM Arms
03 May AM Bike 40.0 km @ 219 MTB race, muddy, tough hills and paths
04 May PM Weights 5.8 Mg @ 45 3x5RM Lats, Shoulders

Totals
------
Weights: 35.7 Mg @ 165
Run: 3.2 km @ 50
Bike: 60.0 km @ 289
Total: 504 min

Details
-------

The race was fun. I was however quite unprepared. It was my first bike race, if that can be an excuse. However I was thinking all of the following beforehand.

a) Perhaps I should get mud tires rather than semi-slicks.
b) Perhaps I should get a couple of spare tubes.
c) Perhaps I should get a new portable pump since I lost my last one!

The reason I entered this race was that it was advertised as the 'hardest race in holland' and it took place in the hilly region near Maastricht. The highest point there is 330m. However, the race route goes up and down all the time, so the total climb must be at least 2000m (but there is no profile available, alas).

In any case, I thought I'd be able to handle the hills no problem. The weather was unusually warm and dry - fortunately! While I had gone riding out in Austria and Switzerland many times, it was usually dry there and thus there was not a lot of mud.
This race was short of a mud-fest, but there was much more slipperiness than I had been used to. My tyres were nowhere near up to the task.

After the first hill, everybody started going down at breakneck speed - another thing I was not really used to. I hit a stone and my hopes deflated along with my front tyre.
I walked back (up), only to find a mechanic at the top of the hill, who dutifully replaced my tube and gave me a spare. I did have an emergency re-fill capsule, though I had never used it.. so now I was fully equipped.

I went down the hill again - losing a total of around 30minutes, I guess.
The next section was horribly muddy and full of obstacles. I soon realised how mud can get in the way of things and how it can make your pedals too slippery to push against.
Oh well.

At some later point I missed a turn by following the road-bike signs - this fortunately was a slight shortcut and I ended up in the middle of the MTB pack. But in doing so I missed the middle checkpoint. I thought of re-climbing that section, but only briefly. I was already a bit tired.

Soon, it was 2 hours, my projected finish time and I was still at km 25 or so. The rest of the race was a mix of walking and hanging on to the bars for dear life in downhill sections.

Anyway, I made it. Next time I'll at least remember to bring a water bottle!

Current speed

Almost a year, a 10km mountain race, and a few injuries later:

5k: 22min
400m: 1:20
100m: 0:15

Getting faster?

After training consistently for two months, I managed to do 3200m in less than 17 minutes, with the first 1600 meter split at 8:18 and the second and 8:04 pretty comfortably and with a good, fluid, forefoot landing technique most of the way.

A Bayesian model for predicting race times

I made a simple Bayesian model for calculating run times based on a set of past performances.
The calculation is done using the formula
u(s; c,a) = c*s^(-a)
where u is a speed, s is a distance for which the speed is calculated and c, a are random variables representing the subject's maximum speed and the degradation of that speed with distance respectively. We start with a prior distribution p(c,a) over these two variables.

Then we calculate a posterior distribution
p(c,a|D) = p(D|c,a) p(c,a) / Z
where D is a set of run time data and Z is a normalisation constant. We can write the first term as
p(D|c,a) = \prod_i p(s_i,u_i | c,a)
assuming the data are independent given c,a, which works OK for this simple model.
Then we assume a laplacian likelihood, i.e. p(s_i, u_i | c, a) = \exp(|u_i - c*s_i^(-a)|).

Finally, we can calculate the most probable pair of c,a, the expected values for c,a, plus confidence intervals for them. This enables us to write MAP, Expected, and Credible Interval values for speed at various distances and thus for running times.

The code (for octave/matlab) is here:
http://www.idiap.ch/~dimitrak/downloads/running_formula.m

Doc

I had some recent shin pains. The doc says it's probably not a fracture, but we are doing some tests to make sure it's really nothing more than generic stress in muscle/tendons/periostal material. No running, will get re-assessed next week.

Did some swimming today, four intervals of freestyle at 400,200,100,50 meters, to check my status and plan some training. I had no shoulder pain during these, so my gym plan, with emphasis on the shoulders must have done the trick. My results were 400m@7:40, 200m@3:16, 100m@1:25, 50m@37.5, which are not much different from the recorded times I have from 18 months ago, so I guess I am remaining at my usual crappy quality!

Running Track -> Shopping Mall

So I went for first mild bike effort yesterday. This was a route I had ran in November and it features a 300m climb. Doing it now on the bike seemed slightly harder since the ground is a bit slippery, but it was easier on the flatter bits. Still, I did not get too tired.

Today I was looking at my maps in order to decide on an easy recovery route and I noticed a stadium drawn in just on the other side of the river. Piqued, I rode over there. There is the new Ice-Hall... and what looks like an empty lot. Going around it, I could see a swimming pool and the rest was just dug ground. Apparently this is because they are going to be making a health spa. Oh well. There is still the open police stadium 2km away.

More frequent runs

I've been doing more frequent runs recently, aiming to keep the amount between 3-5 runs per week, even if it means running very slowly on some days.

At the same time, I've just completed my base weight training, which lasted three months. I finished 10 days of high frequency weight training - and a final week of high intensity weight training. For this last week it was very important to set targets for weights+repetitions and try to follow them. It gave me a nice sense of achievement. I will take the following week off and I'll pick up the effort the week after.

I think I am going to schedule my running workouts in the same way as my weight training work outs, with precise goals for each session. I'll only have time for those in the morning, it seems.

Now, skiing is not so easy to fit into the schedule. Cross-country skating is great - especially as a replacement for endurance-building runs at the aerobic - lactate threshold range. It is also a great exercise for hip muscles. Downhill skiing is another matter. All too often you cannot find an easy enough piste to ski down in. Last weekend I got my knees pretty tired from it. Ski touring would've been my preferred alternative, as it is a great workout plus it helps you to learn ski essentials. but there is not so much snow in the easier hill slopes and I am a beginner.

I wonder if that will leave me any time for swimming + water polo.

Speed versus exertion

I wanted to measure was how much exertion I was putting by running different times. So I ran a few 400m laps at different speeds, with various amounts of rest. This dataset includes all training data from two sessions, for all runs, including recovery runs. However everything done at a speed of at least 3.5 m/s is a proper run.


I added another two data points representing my heart rate while standing and walking at a slow pace, mostly for reference. In order to model the data I chose the function y = f(x) of the form f(x) = exp (ax + b) + c; this seems to have given a relatively good fit, particularly at the high-intensity parts of the work-out.
So, in theory at maximal effort, I should be able to run 400m at 4.5m/s, or push it to 5m/s by going at an exertion level of 110%. I have in the past done 5m/s on a 400m at time trials - and I was knackered at the end. (That does not mean that I reached 110% of your maximum heart rate, but that a maintaining a long-term 5m/s speed would require me to elevate my heart rate to 110% of my maximum) - interestingly, this 110% effort level is in agreement with the number given in some web sites for the VO2 Max % (which is roughly equivalent to maximal heart rate %).

On the other hand, this shows that I really need to somehow be faster with less effort. My 100m running times are abysmal.

This is my new training blog

I decided to create a blog specifically for my training endeavours. Collecting data, statistics and so on.
One of the things that I wanted to know was how fast other amateur runners are doing.. so here is some data from the 10Km 2006 race in Leeds, UK:



My current 10K time is in the 55-60 minute range, so I am clearly near the median - I will be satisfied if I reach the top 10% time I guess.