Anaerobic Stuff - Mr Peabody's WABAC Machine

Time to get into Mr Peabody's WABAC Machine. C'mon Sherman, let's wind the clock back to 2007....


This post is another take on my February 2007 Darth Vader item. Back then I wrote, with considerable assistance from Mr Peabody - er, I mean Dr Andy Coggan, an item about Maximal Accumulated Oxygen Deficit (MAOD).

Andy introduced the concept of using power meter data from a well paced individual pursuit as a means to estimate MAOD (which ordinarily would require lab based testing). He expands on it in the book, Training & Racing with a Power Meter, pp 244-248 (2nd edition).

Just to recap, MAOD is the "gold standard" measure of an athlete's anaerobic capacity. Expressed in litres of O2, it's the difference between the energy produced aerobically and the total energy demand. In an event such as the individual pursuit, a rider's total energy output is typically ~ 70-80% via aerobic means and the balance of course via anaerobic metabolism.

So I thought I'd take the analysis method from that previous post, run it on my recent events and add another twist - the points race.

In recent posts I've mentioned a few track endurance* events I've raced:
- 4km Individual Pursuit (Aussie National Championships - C4 paracycling)
- 2km Team Pursuit (Masters 150+ State champs)
- 20km Points race (Masters 45-49 State champs)

Edit: I've since updated the list to add in the 1-kilometre time trial I raced at the Paracycling nationals the day before the 4km individual pursuit.

It's all part of my comeback to competitive cycling, as these are the events I most enjoy. Well except for the individual pursuit. That's an event impossible to enjoy. But it's fun to do some analysis of pusuiting because it reveals so many things about a rider. Physiologically, technically, aerodynamically and psychologically.

In the weeks and months before my accident in 2007, I rode the same events, the only difference being the individual pursuit was 3km, not 4km and the team pursuit was 3km then vs. 2km this year (different distances for different masters age and paracycling categories).

I've been riding these events for many years but 2007 was my best season up to that point, with a win in the Team Pursuit (in a new state record time), a bronze medal in the National Masters Points race champs and two personal best times in the 3km IP. So for me, relatively speaking, they provide very sound benchmarks for how I've bounced back since then. I'm not going to go into that here though as I've already covered that a number of times.

OK, back to Mr Peabody and the analysis. Here's the chart showing cumulative O2 deficit from my recent races:


Click on the pic to enoxygenate (apologies for the Phil Plaitism).

The picture details the cumulative O2 deficit for four rides - my individual pursuit (red line), the team pursuit (blue line), the 1-km TT, and also a roughly 5-min section from my points race last weekend (the Richie Benaud cream jacket tan line). I'll get to the points race later.

Just to explain the chart - let's take the red line for the individual pursuit. You start the event from a dead stop (your bike is held in a starting gate which releases on count down to zero) and then accelerate over about 15 to 20 seconds to a high cruising pace, which you are then attempting to maintain for the balance of the event. The red line is a measure of how much oxygen "debt" I am incurring as time passes.

I incur this O2 "debt" since my power output in a pursuit is somewhat higher than my sustainable threshold power (which can be produced almost wholly via aerobic metabolism - or in a "pay as you go" sense). Once you ride above threshold, you are tapping into your limited anaerobic work capacity - and it really is limited - meaning that such efforts are by necessity going to be short lived. Harder you go, the less time you'll last. Nothing new about that.

Not only that but once you expend your limited reserve, in order to continue you will have no choice but be forced to ride under threshold in order to recover the O2 deficit. This is why pacing your effort is so crucial in timed events, and in mass start racing why dosing out the hard efforts at the right time is so important. The cost of "blowing up" is considerable in performance terms.

It's also why improving threshold power is so crucial. When you do go into the red zone in a race, you don't incur as much O2 deficit, or can last for longer at that level. And when the pace eases up again and you dip below threshold more quickly, you recover faster meaning you are ready for the next attack before someone else is. Counterattack anyone?

How do we determine this aerobic/anaerobic contribution with a power meter? Well as per the book, it's matter of looking at O2 kinetics of a well paced pursuit:


Andy showed that we can plot, along with the actual power output from a pursuit, a line representing a rider's theoretical maximal aerobic power output based on lab tests of a rider's VO2max** and efficiency***.

Except that in my case, I don't have the latter. Never mind, since the steady state part of a well paced pursuit represents power output at VO2max, we can simply adjust those VO2max and efficiency values so that they match the steady state portion of the pursuit power file. I assumed an efficiency of 22.5% and adjusted my VO2max value until it fairly represented my steady state power output in the pursuit. It came out at 58 mL/min/kg. If you change the VO2max (or efficiency) value, it moves that maximal aerobic (red) line up and down accordingly.

OK, so that's pretty funky, I can estimate my VO2max (or at least a range given that we assume efficiency is in a range typical for trained cyclists).

But by then directly comparing the difference between the maximal aerobic power, and what power I actually produced, we can then attain an estimate of the proportion of energy output from anaerobic contribution.

In my case, it estimates about 17% of my energy was from anaerobic supply. That's a little lower than typical for a pursuit, but my race time was 5:08, which is longer than the 3.5-min to 4.5-min efforts for elite riders in 3km and 4km pursuits and so it's not entirely surprising.

It also means that my MAOD was estimated at 4.16 L. We'll tag that number for now.

OK, so how about those cumulative O2 deficit lines?

In the WABAC machine we saw the way the O2 deficit would climb at different rates when riding a team pursuit as a rider alternately takes a pull on the front (O2 deficit line increases at a faster rate) and then gets back in line and recovers (where the line either rises more slowly or can even fall if the rider is quite powerful and not overly challenged by the team's pace).

If a rider exceeds their MAOD, then there is a pretty fair chance they will crack, which in a team pursuit means they are unable to continue and pull out after their turn on the front, or as sometimes can happen they cannot even maintain the pace of the rider(s) in front and they end up creating a gap in the line, which is bad news.

So I plotted the cumulative O2 deficit line from my recent individual and team pursuits and they shows the same pattern as in 2007. The team pursuit line is much shorter of course since the event is half the distance of the individual pursuit, and in a team, so it is considerably faster.

I also plotted the same line from a section of my points race on the weekend. I chose a starting point very early in the race, it was about lap 6, with 4 to go to the first sprint. My team mate was on the front at the bottom of the track, he slowed the speed down a little in the preceding half lap and then launched an attack, I was on his wheel and went with it.

I had to go pretty hard, with peak power reaching 1184W in order to cover it (he's a world class masters sprinter but not on form right now). The idea was to see what we could get from it - either get a break happening or at least pick up some early points for later strategic benefit.

Problem was, he cracked pretty quickly and I was left with about 3 laps to the sprint line. I was committed, had a gap, so went for it. The cumulative O2 deficit line shows just how deep I went. Very deep.

Once the sprint line was passed I then had to do everything I could to ensure I stayed in the race. You can see how the cumulative O2 deficit line drops away as I reduced my power output and went on the hunt for good wheels to follow. Not long later you can see the line begin to rise again as the next sprint was approaching. I sat that one out just making sure I got through unscathed and could cover any counterattacks.

When you look at the blue line tracing my cumulative O2 deficit from the team pursuit, it reaches a maximum of 4.26 L (about 2.5% higher than from my individual pursuit) and in the points race I reached 4.46L (7% higher than in the IP). What's going on there?

Well, a few things:

- firstly, there is normal day to day variability in performance.

Given that in this analysis we are keeping VO2max and VO2 kinetics**** constant, then the performance is wholly expressed as a difference in MAOD. And since anaerobic contribution to power output is still only 25% or less of total over several minutes, then it still only means a difference in performance of ~ 20-25% of 7% or less than 2% of the total power/energy.

- the next obvious difference is group versus solo efforts, and the influence of motivation/psychology

I would never discount the role that motivation can have on performance and perhaps I am capable of pulling just that little bit more out of myself in a team or a mass start event than I can in an individual pursuit. I can't imagine how I could go any harder in the IP, but it is interesting nonetheless to see if there's any more blood to get from this stone.

- thirdly, as Andy mentioned to me, lab studies indicate that MAOD is independent of the duration of the effort, although he doesn't recall any studies looking at efforts quite as long as this (~5-min). Perhaps that is a factor as well.

So there you have it.

As for the points race, well that attack was a very big risk and a large match to burn so early on. I really needed it to either form a successful break or net 5 points in the opening sprint (3 at least given the race favourite was always going to be very hard to beat). I was overhauled on the line and ended up with 1 or 2 points (I forget exactly) and so it meant I had not gained the desired return on investment. It sure wasn't through lack of trying.

It also meant that since I had gone so deep into O2 debt, I would need every ounce of craftiness to stay in the race. Perhaps in going so hard, the chase was not so easy either and everyone else had to recover too and that was just enough to keep me alive. Thereafter I just took my opportunities to collect points as I felt able. I had to gain 3 points in the final sprint to have a chance but didn't have the legs for that last lap to contest. It was enough for a 4th place finish. Had my initial salvo netted 5 points, perhaps the result may have been different and I made the podium instead.

That's bike racin'.

Edit: Since posting this the other day I also added to the chart the data from the 1-kilometre time trial. As we can see, I reached a MAOD of 4.30L, which is consistent with giving it all and with the MAOD values attained from the other efforts. Not sure if it affected the value attained in the pursuit on the next day, but does highlight the day to day variances.

Thanks again to Andy Coggan for his inspiration, Ric Stern for getting my form to such a good stage and all those team mates and competitors and supporters who help bring the best out of me.


* We call them endurance events, since even though they are about as hard as hard can be and relatively short in duration as far as cycling events go, they are still fundamentally aerobically (with oxygen) "fueled" efforts, albeit with some sizeable contribution from our anaerobic (without oxygen) energy systems.
** VO2max is the maximal rate of oxygen uptake by the body, typically occurs when exercising very hard for several minutes, although it can be induced with efforts lasting over longer periods (VO2 slow component). Expressed as litres of O2 per minute, or in relative terms as litres of O2 per minute per kilogram of body mass.

*** the proportion of mechanical energy output delivered to the bicycle crank as a ratio of total energy metabolised by the body - trained cyclists are typically around 19-24% efficient. The balance is almost all given off as heat (which is why we get so darn hot when going hard).

**** Initial VO2 assumed at 0.5 L/min and half life for VO2 assumed to be 25-seconds.

Read More......

Darth Vader Rides the Teams Pursuit

MAOD – Maximal Accumulated Oxygen Deficit is the theme of today’s chat.

This is something I first learned about via The Book (Training and Racing with a Power Meter) and also in subsequent analysis of power meter data kindly undertaken for me by Dr Andy Coggan last year. It’s pretty funky stuff, so hang on for the ride if you can.

Original Wattage list reference here and the Excel file used to generate such analysis is here.

For those interested in delving further, Andy has also prepared a Powerpoint presentation on the topic of the demands and preparation for individual pursuiting which is available for download at the Fixed Gear Fever download page. It's worth a look.

First, let me go back a step or two…

Technique plays a big part in Team Pursuiting

As some would know by now, I’m targeting two predominantly aerobic events, which have an anaerobic twist – the individual pursuit and points racing. Along the way, I get the chance to ride in one of my favourite events, the Teams Pursuit. A description of all these events can be found here. A quick glance at my recent posts and you’ll see that my team had success this past weekend, winning the NSW State Master’s Championship.

Two members of the team (Phil & myself) used power meters during the qualifying and final rides. We also both have power meter data from previous individual pursuit efforts. So, what can we learn from such data, in particular what can it reveal that may assist us?

As is already explained in a discussion about the Individual Pursuit in the book (pp 189-192), the performance of a rider in an Individual Pursuit is primarily determined by the combination of their aerobic and anaerobic work capacities. The discourse demonstrates that power meter data from an individual pursuit can be used to estimate the proportion of a rider’s power that is being generated from each of their aerobic and their anaerobic energy systems.

In particular, it is possible to use this data to estimate a rider’s Maximal Accumulated Oxygen Deficit (MAOD) – the best measure of a rider’s anaerobic capacity.

Based on this information, conclusions can be drawn about a rider’s individual capacities and it can help decide the type of training specific to that individual which is most likely to optimise performance (i.e. what specific training leading into the event will make me go the fastest I can go?).

Of course, in an individual pursuit, a rider typically accelerates up to speed and then settles into a quasi-steady state power output, typically at a level equivalent to their power at VO2 Max. See example here. The time taken to reach that VO2 Max power level does vary by rider and is proportionally longer for athletes with higher anaerobic work capacities.

In a Team Pursuit however, the demands are subtly different. While the overall aerobic and anaerobic demands are similar to an individual pursuit, the Team Pursuit also requires a greater degree of technical skill (for riding at 55+kph in an aero pursuit position just inches from the wheel in front, riding a good line in the bends and to effect smooth change overs of the lead rider).

It also places a greater emphasis on neuromuscular power (as the power demands are significantly variable compared to an individual pursuit – e.g. going from following a wheel to being on the front without changing pace demands a significant & rapid change in power).

So in a sense, not all aerobic monsters will necessarily make good team pursuiters. Riders like Brad McGee, Stuart O’Grady and Graeme Brown however all possess sensational aerobic engines and have the skill and top end power required for success in such an event.

Meanwhile, back in the Death Star....

So can we apply the MAOD analysis to Team Pursuit power meter data given that you never reach a quasi-steady state in such an event? Well originally I didn’t think it would be valid but as is his way, Dr Coggan showed it was possible (there are a couple of caveats which I won’t go into here) and he came up with some pretty interesting results.

Let’s start with Phil’s data. Rather than rewrite what Andy has already written, let me simply quote him here:

“In a laboratory setting, the gold standard for measuring anaerobic capacity is maximal accumulated O2 deficit (MAOD), i.e., the summed difference between the energy you produce aerobically and the overall energy demand. While we obviously don't know Phil's VO2 during his efforts, his VO2 kinetics, his VO2max, or his efficiency, it is possible to make some reasonable estimates and thus to estimate MAOD, as I did for Phil last year.

Evolution of O2 Deficit
(click/right click on chart to see an enlarged version)

As you can see in the graph titled "evolution of O2 deficit", during the individual pursuit his O2 deficit (the dark blue line) increased progressively for the first ~2 min of the event, after which it apparently became strictly "pay as you go", i.e., all of the power was apparently being generated aerobically.

This is exactly what you expect and what you typically find, with the only real difference between individuals of differing ability being the absolute values and the time point at which all of anaerobic capacity is expended (e.g., for me, it only takes ~1.5 min, whereas for my wife it takes 2.5 - 3 min).

So, what happens when you extend the same logic to analyze the team pursuits? Interesting stuff, that's what! :-)

Specifically, during the qualifier Phil's O2 deficit (the purple line) grew rapidly during the first 40 seconds, then held steady while he was on the wheels, then grew again when he took a pull, recovered a bit, and so on. Notably, however, at no point did it achieve the same value as during his individual pursuit last year. Assuming that he's in roughly the same shape now, this implies that he was never completely at his absolute limit, and thus was able to call upon his anaerobic reserves when he had to elevate his power above his aerobic maximum while taking his turn at the front and then getting back on again.

In contrast, during the final the power requirement was significantly higher from 40 seconds on, such that his cumulative O2 deficit (the yellow line), while flucuating a bit due to being in a paceline, essentially followed the same time course of that seen during the individual pursuit. IOW, in this case he *did* appear to be at or near his absolute limit throughout almost the entire race, so he simply couldn't recover after taking that final pull."

~ Andy Coggan

Now I should add that the final was ridden at a pace ½ second per lap faster than the qualifier and that Phil played the role of lead rider (I knew Phil had the experience to pace the start to schedule). In the final after his third pull, Phil had reached his limit and withdrew from the pace line, leaving the three remaining riders to complete the final three laps (in team pursuits, it is the elapsed time of the third rider across the line that determines the result – assuming you don’t catch the other team).

½ second quicker per lap may not sound like much but as you can see from the chart, it can quickly take someone from being “comfortable” to being right on or over a their limit.

Use the Force, Luke

OK, Andy has shown us something pretty funky with Phil’s data, so what did mine look like? Click/right click on pic to see an enlarged version:

Well at first glance it looks similar to Phil’s chart, however there are some significant differences:

- My cumulative O2 deficit in an individual pursuit (the dark blue line) is of a lower overall magnitude than Phil’s

- In the Team Pursuit qualifier (the purple line), it is apparent that I never fully depleted my anaerobic reserves, whereas Phil did slightly during the initial laps (Phil was the lead rider, so that is not unexpected). Indeed looking at the O2 deficit line, it is apparent that I was recovering quite rapidly when back in the pace line.

- In the Final (the yellow line), once again I did not exceed my anaerobic capacity until it was time to do a pull on the front. But note my recovery when back in the paceline compared to Phil’s. While Phil’s cumulative O2 deficit effectively kept climbing (indicating a depltion of anaerobic reserves), I was recovering sufficiently to enable another two strong pulls on the front, especially the final effort on the last lap and a bit (which took us from behind to in front of the other team).

- So it appears that I too am in at least as good a shape as last year but one should never discount the positive impact that motivation has on one’s ability to find a little more from somewhere within. I have always been a highly motivated rider in a group scenario.

In summary, once again this demonstrates the value of power meter data. Would have I done anything differently armed with such information? Perhaps. With data from all riders I may have decided on a different rider order. Certainly we rode as hard as we could but could have we used our resources more effectively and achieved an even faster time? Next year I expect all squad members will have power meters and perhaps I’ll be able to back up my intuitive assessment with a more objective look at the data.

One thing is for sure, be careful when you ask a sprinter to provide sideline-pacing instructions to a team pursuit squad!

Photo courtesy of Action Snaps photography

Read More......