Occassional thoughts about orienteering
Saturday, January 31, 2009
Today's race - fun with QuickRouteI ran a local event at Longview this afternoon, put my GPS track onto the map using QuickRoute, and spent a few minutes looking at the results.
First, I look at the heart rate histogram.
The graph shows that my h.r. was higher than usual (my normal avg. h.r. for a race is around 168). When my h.r. gets over about 168, I usually have to ease off and recover a bit. My h.r. today averaged 172.
Something different was going on today. I think it was two things. First, it was unusually warm (mid 60s...i.e. nearly 20 C). Second, I bruised my ribs earlier this week and had trouble getting a deep breath. I found myself taking lots of shallow breaths (like a dog on a hot day). I don't know enough about physiology to be sure, but I'm guessing this resulted in an unusually high h.r. that I was able to maintain. Unfortunately, that unusually high h.r. didn't mean I was running unusually fast!
The next thing I look at is the pace histogram.
What should I make of the pace graph? I'm not sure. To my mind, this graph looks good - the peak is toward the left side of the graph. It'd be nice if it was further to the left. You can't make a lot of the pace graph without taking the course and terrain into account.
Then I look at the direction deviation histogram.
Direction deviation is a new feature in the latest version of QuickRoute. It measures the extent to which your direction varied from a straight line connecting two controls (or whatever two points you've marked on the track).
I think the direction deviation data for the entire course tells you something about the course, but not much about your run. The graph shows a fair amount of running that is well off the straight line. I know that. I ran the course. The coures-wide data hasn't told me much.
All three of the histograms are most interesting when you compare them with other races or other runners. I'm not going to do that today.
I like to take a look at the pace track for the entire course. I begin by looking for places where I was moving slowly.
On the map I've marked the places that jump out with letters.
But before looking at those points (which are marked by letters), note that the track doesn't line up especially well in some places. That means I didn't spend much time aligning the route (not true in this case) or that the map or the image of the map is skewed. In this case, the map itself is a bit skewed. It is an old map and was made in several pieces that were then put together.
Getting back to the places I was moving slowly...
A. A fairly short hill (3 meter contour interval) in an open area - why would I be going slowly? Because the yellow area isn't really open, it is ROUGH open and in this case it was full of chest high, grassy vegetation. I walked bits of the hill because it didn't seem like it was worth the effort to push through the vegetation.
B. Like A, a yellow area, but this time downhill. What was going on? Two things. First, the yellow was a bit rough. Second, I took a careful look at the map to sort out where I'd cross the stream. I slowed down a bit as I was looking at the map. If the terrain had been more friendly, I probably wouldn't have slowed as much while I read the map.
C. I fell where the track goes red. I fell forward and it was uncomfortable. The ribs hurt. Not really "hurt" more like unpleasant discomfort. Anyway, I took a while to get moving again.
One of my rules is that when I fall, I get up and take a careful look at the map. I try not to rush off quickly. Over the years I've made a lot of mistakes right after falling.
With QuickRoute you can figure out how long it took to recover and start moving well again. I'd been running for 20:15 when I fell down. In the next 30 seconds I only covered 30 meters. That's a few seconds on the ground and a few seconds looking at the map and starting to move. A minute after falling, I was only 80 meters up the trail and was just leaving the trail to head to the control. It only took about 30 seconds to cover the same distance on a trail on the way from 5 to 6. Falling down cost me 30 seconds.
Actually, the fall probably cost me a bit more. I wasn't running with much confidence for the next 5 minutes.
D. The slow pace doesn't jump out because it isn't red. But, I was running at a yellow-green pace through a flat and open area. My guess - I pushed a bit too hard on the previous leg and was forced to slow down a bit and recover. With QuickRoute you can check that by looking at the h.r. track. Sure enough, about half way between 5 and 6, my h.r. was up around 175. That's too high for me to maintain for long. I had to ease off the pace a bit. If I was race-sharp, I'd have felt that immediately and backed off quickly. I'm not race-sharp. So, I pushed my effort (while my pace was falling). The high h.r. didn't begin to fall until I'd passed the parking lot on the way to 8.
Maybe I can figure out how much time I lost because I didn't recognize my h.r. was too high. I used QR to figure out that it took me 1:17 to cover the distance where my h.r. was too high. Then I checked a similar part of the course where my h.r. was just a little bit lower. At that pace, I'd have covered the same distance about 20 seconds faster. So, failing to ease off right away probably cost me a good 20 seconds.
E. Uphill, in the woods. No wonder I was moving slowly.
For me, the next step in looking at a track in QR is to start looking at individual legs. That'll have to wait until later.
Back to okansas.blogspot.com. posted by Michael | 6:15 PM
Thursday, January 29, 2009
Some notes about the QR analysisA comment in response to my look at some QuickRoute tracks includes:
...This example seems to suggest that the risk taker may lose something in the short term, but in the long term the risk taker is the winner. Is there a way to work this out better?
It spurs a few thoughts:
1. Millinger might have taken some risks later in the course, where he probably lost the race against Troeng. Check out his track around controls 12 and 14:
2. As a strategy, I think risk taking can be, well, risky. I think the "full speed, no mistakes" way of thinking is probably more likely to work more often.
3. The sort of information you can get from GPS tracks gives us a tool to start to work out a lot of things we've never really been able to study before. By looking at different speeds at different parts of a leg - not just split times - we can figure out a lot about how someone is orienteering. With a bit of work, we should be able to discern different styles of orienteering.
Back to okansas.blogspot.com. posted by Michael | 7:54 PM
Monday, January 26, 2009
Some lessons from sketch mapsThe map shows a course from a training in Norway. The orienteers drew simplified sketch maps of the course. Jan K. did some analysis of the various sketch maps.
Below is his sketch map. The blue marks indicate features that many of the sketch maps didn't include that Jan K. suggests should be included.
Here are just a couple of comments from his analysis (roughly translated):
On the way to the 3rd control it was very useful to have the small house on the map. You saw it from the control and it gave you an easy way to the trail.
The tower behind the 5th control was excellent for finding the control - you could see it a little bit away from the 4th control - the rest of the leg was simple. Some of the runners who boomed the 5th control didn't have the tower on the map.
On the way to the 7th was a very distinct knoll in the middle of the leg....Many of the runners who didn't have the hill on the map boomed the control.
Looks like it must have been a fun training session. You can read the complete story and see some more sketch maps.
Back to okansas.blogspot.com. posted by Michael | 8:27 PM
Saturday, January 24, 2009
Looking at some QuickRoute tracksHere are bits of two QuickRoute tracks from a night training race in Sweden:
These two runners started at the same time on courses that have some forking. The two runners finished 30 seconds apart after about an hour of racing. You can see some interesting differences in their QuickRoute tracks.
Note that the color scales are the same. The darkest green is at 3:30/km and the darkest red is at 7:00/km.
Start by looking at the first leg. Both runners start fast. But if you look carefully, one of the runners slows down a little be sooner. I'm guessing that he's taking an extra look at the map.
That little bit of difference might explain why the first runner hits the control cleanly while the other guy - the one who slowed just a bit later - missed his first control.
Look at the speed of the runner as he makes a small mistake at the first control. He's not moving very fast. That's one of the problems with making a mistake. You have to slow down to correct. As he leaves the control and picks up the pace (the line goes from mostly red to mostly yellow) and gets to the top of the hill northeast of the first control.
At this point, the first runner had the short fork and took the first leg cleanly. He's ahead in the forest. If I had the QuickRoute files, instead of images of the map, I could see how far ahead the first runner was. I'm guessing his lead is a bit over 100 meters. At the pace these guys are running that lead is probably around 30 or 40 seconds. Depending on the density of the forest, that's probably not quite enough of a lead to get out of sight (remember - this is night O' and they are probably running with good headlamps).
Take a look at the pace between controls 2 and 3.
The first runner is moving a bit slower. It is subtle, but it also looks like the first runner's route is a little less straight. My guess is that the second runner has definitely caught sight of the first runner and the lead is shrinking.
Leaving the 3rd control, the second runner makes another mistake. He should be heading almost straight south to his control. Instead, he's heading a bit to the west. He's not moving very fast. My guess is he left control 3 going the same direction as the guy ahead of him, but looking carefully at his map and quickly realizing that he's probably got a different fork. This is the sort of mistake that doesn't cost all that much time, especially because his speed in the 4th control circle is fast (i.e. yellow).
At this point in the race, it looks like the first runner is still in the lead. His approach has been steady and cautious and it has paid off to this point. He's also benefited from the short forks. The second runner is a bit ragged. He's made some small misses and he's probably been thinking about catching the first runner. Thinking about catching the other guy could lead to concentration lapses. On the other hand, he's had the long fork (twice!) and he's not far behind.
It turned out that the second guy finished the entire course 30 seconds ahead (though there is enough forking that it is difficult to really compare results). For the record, the first runner is Mattias Millinger and the second is Mats Troeng.
Back to okansas.blogspot.com. posted by Michael | 5:13 PM
Thursday, January 22, 2009
More indoor orienteering...this time in NorwayMore indoor orienteering.
Back to okansas.blogspot.com. posted by Michael | 8:20 PM
Wednesday, January 21, 2009
OrienteeringPhoto from Worldofo.com.
Back to okansas.blogspot.com. posted by Michael | 9:28 PM
Learning from Norwegian SwimmersMarianne Andersen wrote about a presentation on how the Norwegian Swim Team went from no Olympic participants in 2000 to 2 Olympic medalists in 2008. Here's a rough translation of a bit that caught my eye:
...after the 2000 Olympics they knew that they were far behind the best in the world in the amount of swimming kilometers. With that background (the volume of swimming) and the access to swimming pools that the Norwegians had, it was unrealistic to make up the volume immediately. Dale Oen is far behind, for example, Michael Phelps in the amount of kilometers of swimming in a year. So, they decided that they would be the best at alternative training.
Here's the original article in Norwegian.
Back to okansas.blogspot.com. posted by Michael | 8:36 PM
Monday, January 19, 2009
Some of yesterday's legs with the new version of QuickRouteThe map shows my "direction deviation" from the second leg yesterday. Direction deviation is one of the new features of the latest version of QuickRoute. The shading indicates the difference between the direction I was moving and the location of the 2nd control. Where the track is white, I was heading directly toward the 2nd control. Where the track is a red shade, I was heading about 45 degrees (or more) away from the 2nd.
I think direction deviation might be an interesting measure to play with. On this particular leg it gives you the story of a miss that cost me about a minute. I left the first control well - the track is mostly white. I saw the big reentrant ahead of me through the mist and decided to skirt the top of the reentrant to save a little bit of effort (my legs were a bit shot from the climb to 1). That was not a good idea, of course. Rounding features like the tops of reentrants or broad knolls is tricky - it is easy to get off your intended line without realizing it. I was looking ahead as far as possible, trying to see the very top of the subtle reentrant that the control feature was in. What caught my eye was the very subtle reentrant the is south of where I wanted to go. My direction deviation track is pretty consistent...but it is heading to the wrong reentrant!
As I ran down the wrong reentrant I hesitated. I thought that the reentrant shape seemed wrong. But the hills are quite rounded in this terrain and I wasn't sure it was wrong. I ignored the first alarm bells. Paying attention wouldn't have avoided the miss, but it might have save 30 seconds or so.
Here is another leg with the direction deviation shown.
I left the 5th control cleanly. The deviation direction track is white/yellow, indicating I was heading fairly straight toward the 6th control. Near the middle of the leg, I got a bit off the straight line. That was planned. I was on an old road (you can see some of it on the map). Note that my track doesn't quite match the trail shown on the map. I think my track is correct and that the trail was mapped a bit off.
Below is the same leg but the track shows my pace. Red and yellow are slow, green is fast.
You can see that I was moving fairly slowly as I left the 5th control. Then my pace picks up, but there are three spots where I slowed. First, the track goes red near the middle of the leg. That's where I stopped to adjust my shoe. When the rain began (which was a leg or two earlier) my foot started to move a bit in my shoe. I stopped to tighten the laces - hence the red bit of my track. The rain caused another problem. I couldn't see the map very clearly through the rain on my glasses (and I need the magnification of the glasses to read the map well). So, I had to come to short stops to really see the map. The two stops show up distinctly in the track. I probably could have (and should have) gotten to the control without having to stop twice to look at the map.
Here is another leg with the track showing my heart rate.
Red shows a race effort (heart rate of 165-170). The green is a heart rate about 120 and lower. The change from red to green was where I turned my ankle. The data show that I didn't move for a bit over a minute. After about a minute, my h.r. had dropped and I was able to get moving again.
Back to okansas.blogspot.com. posted by Michael | 3:56 PM
Icing the ankle and watching basketballWhile I've been icing my ankle/foot, I've also been watching a DVD of the KU-Memphis NCAA Champs game from last year. Maybe some good feelings from watching the game will help heal my ankle? That'd be nice.
Of course, the highlight for a Kansas fan is the 3-point shot that sent the game to overtime:
One of the interesting things about the video clip is how the Kansas players got set up in a defense immediately. Clearly they were thinking about the next play (and clearly they were thinking - slow Memphis enough that they can't get the ball up the court for a good shot and don't foul).
The Kansas coach apparently emphasizes that idea - "think about the next play."
Getting back to orienteering, "think about the next play" is a lot like "orienteer 100 meters in the future."
Here's another parallel between basketball and orienteering from the Lawrence newspaper. The story is about college basketball, but I edited the quote to make it about orienteering...
Some orienteers are too consumed with making excuses for why they boomed a control. They are too busy flapping their jaws about the map, the course setter, etc., to concentrate on the next control.
Others spend too much energy blaming themselves to focus fully on the next control. That disappointment feeds on itself.
Back to okansas.blogspot.com. posted by Michael | 10:05 AM
Sunday, January 18, 2009
Georgia - day 2Running ok until a couple of hundred meters east of 8. That's when I rolled my foot and had to walk/shuffle to the finish.
Back to okansas.blogspot.com. posted by Michael | 6:51 PM
Saturday, January 17, 2009
Today's race in Georgia
Today's quickroute from the M45 course at the first day of the Georgia A-meet.
I couldn't find the trail along the train tracks between 3 and 4. I'm pretty sure it doesn't exist. I found an old culvert at the creek and spotted an overgrown flat area. Maybe that was the trail. In any case, I couldn't find it.
A couple of hundred meters before 8 I turned my ankle on the trail. It shows up clearly on the track. It is annoying to turn an ankle on a trail.
Back to okansas.blogspot.com. posted by Michael | 2:36 PM
Wednesday, January 14, 2009
Route choice testing and WebRouteThis is a leg I ran last night. I tested three route choice options. The idea wasn't to pick the fastest option (I think that is not so difficult). The idea was to predict the time differences between the options.
You can take a look at the three options using WebRoute (click on "show routes" to see the three options). The options are labeled right, middle and left.
Powered by WebRoute - WorldofO.com
Can you figure out which route is fastest? Can you figure out how much slower I ran the other two routes? A hint - the fastest route took me 6:26. I'll post the times for the other routes as a comment in a day or two.
WebRoute is a new tool from WorldofO.com. It look useful to me. You point to a map on a web page and then people can go in and draw their routes, leaving a short comment.
Back to okansas.blogspot.com. posted by Michael | 7:03 PM
Monday, January 12, 2009
31 Day Challenge updateToday was the 12th day of January and I did my 12th day in a row of O' training.
Today was also the closest I've come to ending the 31 day challenge. My left foot has been sore since yesterday's run at SMP. If it hadn't been for the challenge, I'd have taken a day off today. I'm glad I didn't.
Today was a short O' session - just 20 minutes (actually 22:37) on the sprint map of the KU campus. The map is one of my favorite local maps. You can check it out (with a course from 2006) by downloading the PDF.
Back to okansas.blogspot.com. posted by Michael | 8:31 PM
Sunday, January 11, 2009
Today's score O' at SMP60 minute score course at SMP today. The quickroute shows my h.r. data. I thought I'd have a shot to get all the controls but was running out of time at the end.
Back to okansas.blogspot.com. posted by Michael | 2:37 PM
Saturday, January 10, 2009
Snapshot from San Francisco visitLaurie, Patrick, Mary and Susan taking a break during a hike near San Francisco. I took the photo a month or so ago and ran it through Tiltshiftmaker.com to create the fake tiltshift photo look.
Back to okansas.blogspot.com. posted by Michael | 4:28 PM
Jogging around a sprint courseI jogged around a sprint course this morning.
Back to okansas.blogspot.com. posted by Michael | 2:07 PM
Friday, January 09, 2009
Today's sprint trainingToday's O' session was shuffling around a sprint course at Loose Park.
Back to okansas.blogspot.com. posted by Michael | 6:50 PM
Thursday, January 08, 2009
Night O' on skatesNight skating orienteering - very cool. Check out the photos.
Back to okansas.blogspot.com. posted by Michael | 7:50 PM
Wednesday, January 07, 2009
One week downI'm a week into the 31 day orienteering challenge. So far, so good.
My biggest problem (so far) has been that my feet get a bit sore when I run in O' shoes frequently. I went 4 days in a row of running in O' shoes and my feet felt sore today. If I'm going to make it through all 31 days I'll need to make sure to sprinkle in some runs in trail runners (which is what I did tonight).
I'm worried about getting in an O' session on two days. One day is when we travel to the Georgia meet. I'll go to work then straight to the airport. The other day is when I'll be presenting a report to the City Commission - that could mean a long day. The easiest solution to both of these occasions would be to walk around on the downtown sprint map. But, can I legitimately count walking around on a sprint map as an O' session?
Back to okansas.blogspot.com. posted by Michael | 8:01 PM
Sunday, January 04, 2009
New OK orienteering tops spotted in the terrain
Gene took this snapshot at the New Year's Day trail run at Woodridge. I think this is the biggest single group of OK O' tops yet seen. Long time readers might recall some discussion of the design about a year ago.
Back to okansas.blogspot.com. posted by Michael | 5:58 PM
Saturday, January 03, 2009
Another Lidar view of RockcrusherCourtesy of Eddie, a view of the Lidar data of Rockcrusher. You can compare it to the old O' map and my current basemap.
Back to okansas.blogspot.com. posted by Michael | 9:41 AM
Friday, January 02, 2009
Sprint trainingToday's training was a short course on the current draft of the downtown Lawrence sprint map.
Back to okansas.blogspot.com. posted by Michael | 7:53 PM
Thursday, January 01, 2009
Just like a Norwegian in CanadaI read something that inspired me today...
Together with the other Norwegian orienteers in Vancouver, I'm going to complete 31-days of orienteering in January. Yep, orienteering every day.
I'm going to give it a try. So far, so good. One day, one O' session.
The plan for tomorrow is a short, easy sprint course on the Downtown Lawrence map.
Back to okansas.blogspot.com. posted by Michael | 7:40 PM