Training Specificity Part 1: crits and sprint metabolism

The capacity of sprinters?  It takes more than just a killer sprint to win a race (like positioning, but my power meter doesn't record that), we need to consider other factors.  I will do so by considering some interesting power metrics, but they'll tie back into well known exercise physiology, if not by the end of this post then in the next one.  Today we look at neuromuscular power.

As always, charts are from WKO4, build 260.  Unlike usual, I will mostly focus on my own data.  Not for ego, but because I coach no track racers with power data, I end up as the purest sprinter for whom I have data.  That is to say, I'm the only athlete whose WKO4 phenotype didn't change from "sprinter" to "all-rounder" during base training.  To be fair, I'm not even a great sprinter.  This is my power curve from 2015, a year in which I sucked at criteriums.

In crits, repeated accelerations are arguably the most important factor.  This year I started training that ability and it's paying off.  The following is the power readout from two criteriums last sunday (I doubled up).

The course was 0.6 miles with two gentle corners and two right angles.  In each race there was no more than 6 minutes of coasting.  I did over 90 laps of the course this day, and the routine went: turn, accelerate, gain position, repeat.  You can tell that to haul my 163lb bulk back up to speed I had to hit about 600-1000w repeatedly.  The 1200w peak in the first race was an attempt to establish a break, and in the second race it was snagging a prime.

The highest average power for 5 minutes was less than my FTP.  Since FTP can be held for an hour, 5 minutes isn't really a big deal.  I selected a random 20 minutes from the first race and looked at the normalized power: 285w, right near FTP.  Random 20 minute chunk from the second race: 275w NP.  This is where the training specificity principle comes into play.  In order to meet the demands of an event, you have to prepare for them in ways that replicate those demands.  Hence this workout:

Since races are about 45 minutes long at the non-elite level, that workout is 40 minutes of 10 second sprints, one every minute.  Besides matching the neuromuscular demands of a criterium, the normalized power of this workout was 297w, over FTP by a few watts.  Excellent criterium specificity.

I've been learning that one mark of a sprinter or pursuiter (besides the obvious) is the ability to create normalized powers well over their FTP.  If you're well steeped in traditional power analysis, you may be thinking that NP represents the physiological demand if you had been working at a steady state, and you're right.  This doesn't mean that if you can normalize a power that you can recreate it steady state.  Believe me, I've tried.  So let's get into why.

The metric I've been working with is 1 hour NP way over FTP, which really only applies to sprinters and pursuiters, and modestly to all-rounders.  How does this happen?  It comes down to stored energy.  As you can see from the sprints above, I'm pretty good (but not amazing) at recharging PCr stores in my muscles, hitting between 950-1050w consistently.  Were I any better at this I would have normalized a higher power, or perhaps there's an optimal rest-sprint ratio to maximize NP for the workout.  Doing 12 sprints in an hour also normalizes 300w for me while hitting about 1250w each time.

What's this have to do with stored energy?  Phosphocreatine is recharged from two molecules that have higher stored energy, phosphoenolpyruvate and 1,3-bisphosphoglycerate (PEP and 1,3BPG).  PEP is shunted off from the TCA cycle and 1,3BPG is a glycolysis intermediate.  As the reaction ADP+PCr->Cr+ATP proceeds for each sprint, we can safely assume that the PCr levels in a muscle cell are well below normal levels.  This increases the demand for PEP and 1,3BPG and thus the glycolytic and aerobic systems.  See here (page 5) if you're rusty on your standard free energies and why this works.

Glycolysis is probably more taxed in sprinters with good anaerobic power since glycolysis can create energy more rapidly than aerobic metabolism.  Looking at it another way, because the normalized power a sprinter can create is much higher than FTP (here representing the aerobic ability of the athlete), the difference must be made up by stored energy in the form of PCr and glycogen.  This ability to rapidly mobilize non-aerobic energy (during and while recovering from high intensity efforts) is why sprinters and pursuiters can't simply put out their normalized powers at a steady state.  It should be obvious that FTP is also important for criterium racing.  Greater demands on stored energy during the race means there will be nothing left for the final sprint.

Next post will look more closely at the ability of some sprinters and all pursuiters to chew through glycogen at rapid rates.  I'll also look more closely at a couple cases of 1 hour NP > FTP.

Race prep: off season training targets

There are a few different sides of a coin when it comes to early season training, which maybe makes it more like a die.  I like to pick two things for an athlete to work on at the same time, and not just to shore up the coin analogy.  Over the winter with lots of trainer time, the obvious priority is FTP (because I don't like to assign 4 hour trainer rides).  The other thing is dependent on the athlete and their goals.  This will go through the process for a few athletes as well as the results.  Any charts are from WKO4, build 255.

Athlete 1, CX focus.  Goals: FTP and force production

The first person we'll look at has been doing road training for about two months now, and after finally having gotten a power meter in late December, his needs are exactly what I expected they would be but there's not deep historical data to compare with so I can't show you any pretty graphs illustrating progression, but here's the graph of his weekly peak values.  The first set of points in the 12/28 week is the testing protocol to establish baseline maximal values.  He has not done a maximal 1 minute effort since, but I doubt the anaerobic power has much increased.

This athlete's main focus is cyclocross, so with half a year before that racing starts, there's a lot of time to work on many things.  His slim 70kg build and tendency to ride at high rpm indicate a lot of slow twitch fibers, so the goal is to increase his threshold and muscular power, even if it means putting on weight.  These improvements will be the foundations on which we'll build his high end power for CX, but right now he needs to improve his musculature and maximal force production.  For the last two months he's been doing almost entirely endurance and tempo riding with strength training on and off the bike.  His training weeks are about 13 hours, plus one 20 hour week.

This training alone has been enough to bring up his FTP from 240w to 270w and raise his Pmax by 150w.  In just two months the two training targets are improving dramatically, and he's on track for a great season.

Athlete 2, road/stage race focus.  Goals: FTP and muscular endurance

The next athlete is a 61kg road racer who excels at climbing.  She's been training since October and, like the previous athlete, has a naturally high cadence but felt a lot of muscular strain and fatigue while climbing steep grades.

The last couple months have seen her intervals build from 3x10' to 3x20' at threshold.  In addition there was a lot of strength work in the gym and low cadence climbing on the bike, but it was different from Athlete 1's program.  We focused on muscular endurance for climbing, but it had the benefit of increasing her anaerobic power as well since the lifting was similar in demands to a short anaerobic effort.  Here's the graph of her build since October.

We can easily see that she's natural athlete, able to maintain high workloads and recover well.  Whether doing endurance, sub-threshold, or threshold efforts, her FTP has built from 210w to 250w.  The purple dots are 20' maximal power values, and they show steady progression upwards.  Also worthy of remark is that she did no anaerobic work until December with her first 1' power test (450w) around 20 December, which was already a full 100w better than her 2015 best.  After some time off in January for exams, she spent most of February maintaining the weight lifting and reestablishing her endurance, and her 1' power is still the same.  In terms of meeting the goal of increased muscular endurance and power, she is able to handle VO2max hill repeats at low cadence (up to 40 minutes of climbing) and her Pmax increased from 680w to a high of 820w in December.

Athlete 3, road race focus.  Goals: w/kg and endurance

This last athlete went through a real transformation in the last couple months.  His goal was to excel in road racing, so his goals were increasing FTP and losing weight.  In October, he was 10kg heavier, with a 1400w Pmax, 700w 1 minute power, and 280w FTP (not shown in graphic).  His Pmax dropped to 1200w, but his w/kg at FTP went from 3.8w/kg to 4.7w/kg.  This is the power curve not normalized to weight (the w/kg graph is a bit muddy with several lines on it, but this is clearer).

For the long road races he'll be doing, this is what I wanted to see.  Would he have 10-20w greater FTP if he hadn't lost the weight?  Probably, but a 345w FTP at 80kg that would only leave him at 4.3w/kg.  Fortunately his Pmax hasn't suffered as terribly as I expected, and will serve him well in reduced bunch sprints.  We've done some 3' VO2max work, but no anaerobic or sprint work yet.  It's just been burning calories and increasing threshold, and this is reflected well in this graph.  In fact, despite regular rest weeks, he spent from October through Christmas with a negative TSB, having ridden 6000 miles.  There's been reduced workload since, but it's the foundation of a successful season.

Two types of CX riders: training analysis

Not many of the athletes I coach have power meters on their cyclocross bikes, but two who do have them are quite different riders.  What follows is an analysis of the metrics I used when determining their workouts within the scope of their goals for the season.  All graphs come from WKO4, build 204.

The first rider is a 6'1 male, ~190lbs, 1500w Pmax, 21kJ FRC.  We'll call him Winston.  It was his first cyclocross season and no big goal races, so we worked on his technical abilities to help with crits in 2016 as well as his steady state power, which would be a limiting factor for him in many races as he moves through the categories.  PD curve:

The other is a 5'11 female, ~130lbs, 730w Pmax, 11kJ FRC.  Let's call her Linda.  Her big goal is winning collegiate nationals, which is tomorrow.  PD curve:

Since these PD curves show the male as a decent all-rounder and the female as having a slightly better steady state power than high end power, I should qualify this and say these are best all time efforts.  Winston is a former lifter and has lot of natural power, so during cross (he only did September and October) we worked a lot on his steady state riding so he wouldn't so heavily rely on his anaerobic reserves during racing.  Linda has very good abilities riding steady state, so we worked on her ability to accelerate and recover as well as her anaerobic power.

It'd take too many graphs to make the next point about cadence, so I'll just tell you.  Winston naturally pedals at a low cadence, and his ability to produce large torque puts him at an advantage in a cyclocross corner where the speed delta into and out of corners is large and even if you shift down twice, you're still accelerating from 75rpm.  This also takes a lot of core strength, which is another thing Winston has thanks to his lifting days.  We didn't do a lot of drills to increase his strength on the bike, so a lot of his targeted workouts were steady state to get him more used to putting out power for 40 minutes as well as getting him used to the body English of riding off road.  A usual interval workout was 45' tempo, or 3x15' sweet spot with accelerations every so often.  A technical workout would be classic 30/30/30/30s (30" sprint, 30" rest, dismount and 30" run, remount and 30" rest), a few hours of riding singletrack, or a recovery day with lots of cornering.

There aren't a lot of graphs that show an improvement in a rider's technical abilities, and since Winston's power numbers stayed steady, what I can show you is an improvement in points: first couple races were low 500s to high 400s, and his last race was low 400s (from crossresults.com).  A good and steady improvement for a first time cross racer, and already with 4 upgrade points.  Regardless, this is the weekly MMP (highest average power for specified time) for his two months of cross.

Linda's much more of a high cadence rider, a practical all rounder in terms of build and abilities.  A lot of her steady state workouts on the road show her pedaling at 95-105rpm, but in cross races the average was about 80, and it took her a few seconds out of corners to get up to speed.  She's quite thin as well, and so doesn't have large muscular power or FRC to dig deep continuously, but her steady state power and technical abilities serve her very well and allow her to ride consistently throughout a race, especially on longer rhythm sections.  Because a lot of cross is coming out of corners at low cadence, she was losing some time.  Most of her drills were a combination of accelerating from low cadence, from steady state, and pure anaerobic power.  An interval workout she did frequently was low-ish cadence threshold intervals with periodic accelerations.  There were many anaerobic power workouts, from 30" intervals to repeated 15"-2' intervals with varying amounts of rest.

To give you an idea of how effective this approach was, here is a quadrant analysis from an early season race from Linda, before I started working with her.  Most of the scatter is relatively low torque and high cadence, but there is still a very large amount of high torque, low cadence.

Next is a late season race.  You can see how the cluster shifts up and to the left.  She's still doing a lot of high cadence, low torque pedaling because it's what she does naturally, but the low cadence, high torque efforts are more repeatable and common.

Another thing Linda's long season allowed us to do was bring her to a nice sharp peak for collegiate nationals tomorrow.  The following is the graph of her build to nationals.

What this graph doesn't show is how hard she worked through late November.  Those power numbers went up, but she was feeling a whole lot of fatigue.  Linda is a great communicator, and I regularly adjusted her workouts to include less volume and intensity as the fatigue went up, so we were able to keep her riding the edge of functional overreaching without burning her out.  Even better was that we knew she would have almost an entire month to rest and maintain fitness between her last regional race and nationals.  During this time she rested, did some low intensity, high volume road riding, and did the absolute minimum amount of high intensity intervals, just to keep the legs sharp.  We can see her hitting her best power numbers of the season the week before the holiday.  When I took this screen cap I didn't have her files after Christmas, but she's been maintaining well and becoming fresher every day.

Something you may notice about the MMP graphs is the 5" power getting progressively lower through both riders' seasons.  This is something I see a lot, but it's not cause for concern.  It reflects the fatigue that cyclocross has on a rider, especially with regular racing and trying to hit those high numbers repeatedly.  Being able to hit that peak at the end of a race matters a lot in road sprinting, where freshness matters.  But like Adam Myerson says, there are no sprinters at the end of a cross race.  It's the combination of technical riding, accelerating repeatedly, and being able to maintain a high pace between those hard efforts that makes a fast cross racer.