After riding 70.3 World Champs at StG, I realized most of my training was on terrain where I could maintain a power number almost all of the time (lots of flat parts, short ascents, short descents.) Training with 80/20 Endurance training plans, I had a small power range in Zone X that I targeted for all of my race sim intervals as I closed in on the race. But how should I race a hilly course with a avg power target for the whole course, considering the descents with zero or near zero power? Is there a rule of thumb for this, such as target 15% higher than your full course target for the ascents to balance out the zeros on the descents? Overall, I rode at an IF of about .83 which is on the high end for me I think, but I felt like I had more left in the tank when I finished the bike and wondered how to better manage and apply power in a hilly course. And does some app like Best Bike Split address this issue too?
but I felt like I had more left in the tank when I finished the bike
How was your run?
Logella asks the right question.
But yeah, BBS does profile your power plan with elevation taken into account. My plan for a very hilly full distance race below as an example if it helps. Mostly it was asking for some power on nearly all of the descents, but with a couple of exceptions where it suggested freewheeling due to, I assume, diminishing returns on exceptionally fast downhill parts.
Cheers, Rich.
https://live.staticflickr.com/65535/48576386992_44277ea642_c.jpg
It’s unclear to me from your post whether you’re just concerned about whether you can use the power saved on decents elsewhere, or whether you’re considering the fastest way to ride a hilly course as a whole. The fastest strategy on uneven terrain or when you have a headwind, will always be to raise your power when climbing, or when riding into a headwind. (They’re not the same situations but there are overlaps - I’ll put headwinds aside from here on.) The question is how much you can realistically raise your power on climbs without hurting your endurance to a point where it’s not worth it. I have a tendency to overdo it on the climbs and have to be pretty disciplined to avoid suffering later. However, I’ve often seen riders trying to hold a constant effort throughout and that’s just a bad plan unless you’re riding on a flat course in calm conditions.
I don’t know the course at StG but of the hilly courses I’ve done, the fastest way for me was to roll uphill near threshold and continue to apply pressure on the downhills within reason.
That strategy had people passing me slowly uphill and me passing them back going downhill and never seeing most of them again. You’ll have to be confident in your descending skills to do that.
You’ll also have to accept that your run will be compromised more with more time spent at threshold.
On hillier courses, as has been said, you need to target a higher output on the uphill sections, then try and power over the top of a hill into the descent. I tend to draw the line somewhere in the 37-40mph range for coasting versus continuing to spin. I also raced St. George, and looking back, I had 19 minutes at 0 watts. That’s quite a good chunk of time (14% of my total bike time). This was compounded a bit when the storm hit. I was just getting to the top of Snow Canyon. I ended up coasting a lot of the time down the long stretch into St. George - in better weather I would have pedaled up to around 40 mph, but it was sketchy as hell and I only averaged 32mph on the final descent. Looking at files of others, I lost at least two minutes in that 8-9 mile stretch.
It’s also a good idea to go more off of Normalized Power on a hilly course then AP. My VI at St. George ended up being a 1.10 (around 25 watts between AP and NP), which wouldn’t be great on flatter courses. However, my NP was pretty close to what I would aim for in a normal 70.3 that would not have as much coasting.
I will say that it’s a hell of a lot easier to pee during the bike at St. George then a course like Eagleman! Those downhills make it very easy. 
I have struggled with this too on hilly courses which is pretty much all I have ever done. I had a coach telling me to keep the VI low like under 1.05, but that is nearly impossible on hilly and punchy courses. I rode a race last week at a higher VI (1.14), lower AP, higher NP, and had a 5.5 minute PR vs prior attempts on the same course. Ran better off but that was probably more overall fitness than bike effect.
I think the trick is to climb hard, ride hard up and over, but get some recovery on the downs where you can only soft pedal or not pedal at all.
I would like to know if there is any real research on this topic. I know coaches are preaching keeping NP close to AP and VI low but at some point that has to go out the window on courses where there is more that maybe 60-70 feet of climbing per mile, and where the grades are greater than 5%, like punchy climbs of 8-12%. The pacing strategy cannot be the same. There has to be some recovery benefit from the low power or no power descents, more like an interval workout.
BestBikeSplit as mentioned above is a pretty phenomenal tool. You can look at the splits they’ll give for different sections, match them up with the type of terrain, and then use that as a guide. They also have available a cheat sheet with power targets based on your FTP, the course, and a couple other parameters. The cheat sheet has targets for something like: flat, steep uphill/stiff headwind, steep downhill/stiff tailwind, moderate uphill/headwind, and moderate downhill/tailwind. It’s really useful as a guide, and I think you’ll find that it’s the fastest strategy. As speed goes up, your power creates less speed. Because of that, you want the highest power when your speed is lowest, to a point, of course. Aside from corners and other such technical aspects (which are somewhat accounted for in BBB), I can’t find any fault with BBS as a tool to help guide your plan. It’s accurately predicted most of my races within a couple minutes when I’ve set up all the details (weight, FTP, IF, weather, course conditions, etc.) with careful consideration. I recommend first using a flat race (or one you feel you executed as well as you can) to verify/estimate the CdA, Crr and other numbers by lining up the power, weight, and other details, and then take those numbers to a hilly goal race plan to see what it might look like. I think you’ll find it extremely useful. VI in my opinion is often a poor measurement. The case of an uphill time trial with a technical descent stands out: you’d likely have a huge VI by necessity. Obviously, you can’t ride a normal rolling course at 1.5 VI and still succeed, but I think anything below 1.2 on some more extreme courses isn’t necessarily the worst plan. You can set a minimum VI on BBS, and my guess is that the highest they’d take you is something like 1.15-1.2 on normal settings, but I’m not sure. For any course I’ve done, that would be pretty wild.
Are there other good apps/resources to plan power on a race course besides BBS?
Is there any simple math to establish a power ceiling for inclines? Going into a 70.3, I know my FTP, I have guidance from 80/20 Endurance (my coach/plan) that I should aim for about 78-81% of FTP for my Normalized Power for the race overall. So if that were a flat course with fixed headwind the whole way, I’d aim for that NP as my number for every segment, every mile (you all with me so far?) But then when I introduce hills with steep enough descents that I can’t produce any power (zero power), that time at 0 watts will reduce my avg power and NP if I kept my flat and incline power the same. Is there a maximum % above FTP to “punch it” on the uphills? Is there any math to predict when I would overdo it?
There’s some math you COULD use. It would be some sort of FRC analysis, probably. As far as I’ve seen, this is available from TrainingPeaks’ WKO, and perhaps a couple more places. Ultimately, it’s just so in depth, and more focused on preservation of glycogen as it relates more specifically to the dynamic efforts in bike racing, as to be unhelpful in most cases for a triathlon bike split. The ceiling I have heard of and been happy with is something like 110% FTP for 5 minutes or so. Obviously you don’t want to do that 8-10 times or your NP will be through the roof. The super basic gist of the best pacing I know of: ride 10-15% above target on hills, somerhinf like 5-10% low on flats and 10-15% low on downhills. I think getting TOO involved is not useful. The difference in 2-3% on race day MIGHT be important, but at that point I would mostly argue you should practice riding enough to go on feel for that.
I don’t totally get the hesitancy for BBS, but if it’s about paying for it (it’s admittedly expensive if you’re not on the absolute cusp of a KQ or something like that) you can use it for free. You only get one “race†and there are a few things that get into the weeds pretty good that you can’t use, but for the basics, it’s all there. It takes a little more work to do all the analysis on different courses, but the free version has more than enough to get it done.
This paper looks at the problem of headwinds/tailwinds and analyzes the problem in terms of optimizing average speed for a given energy expenditure: https://arxiv.org/ftp/arxiv/papers/1309/1309.1741.pdf - but fails to account for limitations imposed by human physiology (e.g., inability to ride 150% FTP for a long stretch).
Using a critical power - W’ model of endurance physiology (critical power is the power you can sustain for an indefinite amount of time; W’ is your store of matches to burn in shorter efforts) you could figure out how long the climbs were and then set a limit for what fraction of W’ you were willing to expend on a single climb. If the descents were long enough you might be able to assume that your W’ store then refilled in between each climb, but if the descents were not sufficiently long, you might need to account for incomplete refilling of W’. Also you’d need to figure out by testing what your own W’ reservoir was.
Based on this model, you would certainly ‘overdo’ it if you used 100% of W’ on any single climb (that would mean riding an effort comparable to the maximum power you’ve done for the duration of the climb).
I hear where you are coming from, and I’ve been there, racing on hilly courses and unsure how to handle the hills.
The key for me is to shift into really easy gears going uphill, thereby not burning any matches. On the downhill, pedal, and keep your power up (you go fast, right, that’s good!).
The guidelines are to keep your VI below 1.05 so as to not toast yourself for the run.
Joe Friel explains it here: #17: Get IF and Vi right, by Joe Friel | IronmanHacks
In his book “The Power Meter Handbook” he also has guidelines about what speeds and power to follow when going downhill.
Are there other good apps/resources to plan power on a race course besides BBS?
Is there any simple math to establish a power ceiling for inclines? Going into a 70.3, I know my FTP, I have guidance from 80/20 Endurance (my coach/plan) that I should aim for about 78-81% of FTP for my Normalized Power for the race overall. So if that were a flat course with fixed headwind the whole way, I’d aim for that NP as my number for every segment, every mile (you all with me so far?) But then when I introduce hills with steep enough descents that I can’t produce any power (zero power), that time at 0 watts will reduce my avg power and NP if I kept my flat and incline power the same. Is there a maximum % above FTP to “punch it” on the uphills? Is there any math to predict when I would overdo it?
Bear in mind that going over average on the uphills and under average on the downhills might get you a net zero change in your AP, but the NP will be higher. That’s kind of the whole point - to model the physiological cost vs simply calculating your energy output.
I’m not claiming that the BBS model is perfect, but if you look at my plan above this was based on FTP of 265 watts. This was a very hilly course (Norseman) and nowhere in the plan does it ask me to hit or even approach FTP. You can see the maximum power on that plan is <240 watts. I’m not saying that this is right, but that is how BBS models the optimum profile for putting out my target TSS on that course.
At the end of the day you are trying to moderate NP to manage your TSS for the bike leg, and protect your W’ on the climbs.
Cheers, Rich.
I made this video in 2020 about climbing and made some examples using Zwift including time analysis using ave. power and NP. Most of my athletes and friends live in the Houston area where it is flat. It’s about 25 minutes. Hope it helps.