Help me model "Least Watts" vs "Always on the gas" riding
I like my coasting. In my racing brain, I think back to some early guidance on “least watts” race execution, and remember – or selectively recall – the reasoning that above 32mph (or whatever), it just doesn’t make any sense to push watts. I remember being persuaded by the reasoning that the power required to increase or maintain speed when already descending at that speed is just not worth it, and on balance, it makes more sense to bank the heartbeats / ts / watts for a later point. Again, I like coasting, so I’ve decided to keep remembering it that way.
As I start to put my mind to an upcoming race and look for time savings here and there, I’m challenging that recollection. To get to the bottom of it – and maybe to test my assumptions - are there any suggestions from the team about how to model this? Someone will probably suggest good ol’ math, but there are so many variables that I can’t imagine a model. Would Best Bike Split or Craig Willett’s speed given watts calculators let me get at this? If my premise is “least watts” riding, it pushes the question of the cost* of those downwhill watts … I guess I could just identify the difference between coast vs non-coast power from the ^above^, and estimate the tss difference between the two styles of riding on a course, right? So I get something like “coasting Dave’s bike cost is 265tss; pedaling Dave’s bike cost is 277.” Certainly, putting it in those terms would be the preference, if only to peg it to an impact on the run.
Thoughts?
*one more thing to complicate it: I’ve posted in the past that I don’t think a watt is a watt. My example: if I were to go from riding on the flat, no wind, at just riding along effort, and then went a minute or 2 at 100% ftp, I would probably see my HR creep up by ~20bpm, and my Rate of Perceived Exertion increase to maybe 8. However, if I were similarly just riding along in the same conditions, came to a descent, and then rode the descent at exactly the same 100% ftp, my HR would maybe creep up by maybe 5bpm tops, and RPE might be a 6. There’s just no way that a downhill watt is the same as a flat/climbing watt (or a tailwind watt is a headwind watt, for that matter). I know that physics doesn’t agree, but tens of thousands of hills tell me otherwise.
As I start to put my mind to an upcoming race and look for time savings here and there, I’m challenging that recollection. To get to the bottom of it – and maybe to test my assumptions - are there any suggestions from the team about how to model this? Someone will probably suggest good ol’ math, but there are so many variables that I can’t imagine a model. Would Best Bike Split or Craig Willett’s speed given watts calculators let me get at this? If my premise is “least watts” riding, it pushes the question of the cost* of those downwhill watts … I guess I could just identify the difference between coast vs non-coast power from the ^above^, and estimate the tss difference between the two styles of riding on a course, right? So I get something like “coasting Dave’s bike cost is 265tss; pedaling Dave’s bike cost is 277.” Certainly, putting it in those terms would be the preference, if only to peg it to an impact on the run.
Thoughts?
*one more thing to complicate it: I’ve posted in the past that I don’t think a watt is a watt. My example: if I were to go from riding on the flat, no wind, at just riding along effort, and then went a minute or 2 at 100% ftp, I would probably see my HR creep up by ~20bpm, and my Rate of Perceived Exertion increase to maybe 8. However, if I were similarly just riding along in the same conditions, came to a descent, and then rode the descent at exactly the same 100% ftp, my HR would maybe creep up by maybe 5bpm tops, and RPE might be a 6. There’s just no way that a downhill watt is the same as a flat/climbing watt (or a tailwind watt is a headwind watt, for that matter). I know that physics doesn’t agree, but tens of thousands of hills tell me otherwise.
0
Comments
Will be interesting to see what the smart people say here!
Agree with your 'watt is not a watt' statement (i think we talked about this before). I wonder if the increased speed of downhill and tailwind watts for a given output results in increased cooling due to the higher relative wind and therefore lower HR and perceived exertion?
I think if you can pedal down a hill and get some more speed but keep your HR low, why not do it- yes a few more TSS points but worth it , no? Plus, up to a point, my bike feels more stable and maybe less prone to wind/speed wobbles when I am pedalling. But then again I am a shrimp and seem to be getting shrimpier.
I tend to think a watt is a watt on a given bike, but not on different bikes. So 200 watts on my road bike seems easier than 200 on my tri bikes. Different muscles maybe?
If you are thinking about Kona, there must be a ton of files out there to look at.
At the end of the day, you want the fastest overall time possible. Implicitly, you recognize the value of coasting downhill - it provides an opportunity for rest and recovery mid-race. The question is, does that come at a cost of a slower time than you might otherwise get overall - bike + run.
The factors are way too complex to allow for a reasonable "model" which is usable in real-world conditions. Just for starters, there're wind, gradient, length of hill, aero coefficient, soft pedaling vs pure coasting. Then there's the issue of VI, which is buggered by coasting, but may not fully reflect the recovery value of doing so. And on and on.
A couple of experiments you might try to help you think through this morass. First, go back to your own races which included hills, both on HI, and in other races. Pore over the files, to see if you can find examples of both coasting and not coasting down the same grade. See what the difference in speed was, and calculate how much time you "lost" as a result, and correlate that to the overall result.
Then, experiment in real life. Take several RPP interval bike days ("Saturday Ride") during training, and do them on the same hilly course. Play with all the variables - coast, don't coast, differential power going uphill vs down, same power up as down, whatever you can think of that you might want to do on race day. Do a brick after each one. Than ponder the resulting data and your own sense of how you ran and felt.
Science aside, my own perception, thinking about hillier IMs I have done multiple times (Coeur d'Alene and Kona specifically, but even AZ, with that one 4 mile "hill" done three times) is that some judicious coasting allowed me to periodically recover, then both bike and run *faster* than trying to get the lowest possible VI by pedaling hard downhill. But it required some discipline to cap my uphill efforts.
Dave,
You have several Kona ride files. What's your amount of coasting time on those rides 10', 20' more? If you cut that time in say half then make an assumption on how much extra speed you would average miles per hour then you could calculate the reduction in time this would save.
Of course this is a blanket assumption of your increase in speed do you push your max speed up to 33 and then soft pedal or is it 34. I think part of the thought process here is your discussion of switching to a bigger gear and how much that will get you on the top end with out spinning out. As you have mentioned this will have an effect on you and your available heart beats. Also will this have an effect on your run given you are now pushing without those micro breaks. Hard to say but I think your 5 hour power and your big bike weeks will put you in a good position to not slow down but hard to say. As Al mentioned you are giving up some recovery.
If my quick thinking is right you would travel 5.334 miles in 10' at 32MPH/hour. If you upped your speed say coasting down Hawi to 33MPH you would cover 5.5miles travelled in 10'. To cover 5.5 miles at 32 MPH is takes 10' 18" so this effort for 10' saves you 18". Did I get that right???
I used this calculator http://www.csgnetwork.com/csgtsd.html
Alternatively if you think of your overall average 112miles @21MPH is 5H 19' 59", if you decreasing coasting and increase your overall speed for the race by 0.1MPH that saves you 1'30". Now what's the opportunity cost of that 0.1MPH on your run. I don't recall you IM bike times off hand and I need to get back to work if you complete the 112 @ 20.5 that 5H 27' 48" ride and increasing to 20.6 given 5h 26' 12" or a savings of 1' 36".
These are not huge chunks of time but you are looking for the margins to make 10 hours you might want to look at some sections of the course to cut down soft pedaling to get 30" to 1'. Your combination of more gearing to not spin out and less soft pedaling could also help, perhaps getting you 1'-2'.
Hi Dave - Here is a idea about how to model this. BestBikeSplit will give you a very detailed power plan for the course. Here is one that I have started modeling for my upcoming IMWI race (Rich's IMWI BestBikeSplit Link) and the RideWithGPS model for the course (showing ascents and descents).
I've effectively set descents > 4% to 25% IF. The TSS numbers do not add up for to the exact numbers that BBS shows due to rounding - but maybe this gives you a starting point for thinking through the problem. Non pedaling should just result in reduced TSS... The link to a google sheet is below.
TSS Model for Non Pedaling Descents
Not alot of experience on my side, but would like to add my tests done so far since I own a PM.
I've seen when I ride over 70%, my runs are alot stronger than when I am riding at 70%. Today, I did my camp day 1, IF was 66% (per TP) but was able to run at 7:40/m, run was ok but not regular pace
at 70.3 MT, I rode at 80% and was able to run at 7:29/m (I know you cant corelate to an IM).
my question is : is it better to have a VI of 0.94 (which was today) or be a little over 1 (ie: 1,03-1,05) at 70.3MT VI was 1.17 !! and I guess its coming from the fact im coasting downhills cuz I pee in my pants going fast on downhills since my crash.
Sorry to jump in with questions !!
ok. this puts me ahead in my thinking. I have about 15 hours on airplanes in the next few days, so I'll bring the data with me and get my math on at 40,000 feet (or wherever I am). This plane will not touch down until I can answer "is it better to coast or pedal."
I'm especially interested in you thoughts on @Joe's thoughts of not including zero's. As an aside, I was looking at a friend's bike file from Placid. He had over-ridden his bike and ruined his day. Interestingly, his watts were only marginally above his target... but when you looked at the histogram of the power, he had lots of minutes at 0w and way too many watts above his target. The Pavg and Pnorm were only slightly high, but the matches were still burned.
I think that this lends an interesting twist to this discussion - we know that watts above are target cost us more and that TSS doesn't scale linearly - have to be careful how we apply the TSS saved by not pedaling.
I know this is not a very geeky way of determining what works best, and it may be different for individuals, however it has changed how I ride, and I have set pr's on the run for two 70.3's in a row this year.
As you may know, IMLP'16 last week a COULD vs SHOULD exercise for me. I basically applied my power to the course in a manner that would get me the fastest bike split vs set up the run. Specifically,
When I find some time I'd like to plug my numbers into BBS then do the same modeling what my SHOULD bike split would have been to determine that delta between COULD and SHOULD that we talk about in the 4k talk. My gut feeling is that the 10-15' difference between COULD and SHOULD that we describe is probably accurate but the experience off the bike is night and day.
Not really sure what I'm saying here
other than you know that the EN Steady way of riding isn't the fastest way around a bike course but is an excellent method for setting up the run. You'll just need to do some smart guy stuff to figure out how far towards the COULD side of the scale you can ride and still have a good run....I think 