I was browsing through athlinks race results as I usually do on a Sunday morning (or any other morning for that matter) and I noticed something that I had a hard time explaining. I was wondering if our collective wisdom here can give me some insight into this.
Why is there so little difference in top bike times relative to other disciplines? And how should this affect our training?
For example, let’s take Top 30 (overall) results for Accenture 2007: Swim:
Fastest of the Top 30: 21:54
Slowest of the Top 30: 31:13
Difference: 9min 19sec % Difference: ~42%
Bike:
Fastest of the Top 30: 58:17
Slowest of the Top 30: 1:05:03
Difference: 6min 46sec % Difference: ~11%
Run:
Fastest of the Top 30: 35:08
Slowest of the Top 30: 45:25
Difference: 10min 17 sec % Difference: ~29%
Why are we seeing these numbers and what do they mean?
A) Once you get close to a FOP cyclist, should you focus your time on running? or better yet, swimming?
B) Is it easier to get within throwing distance of top cyclists? (relative to top runners and top swimmers)
C) Cycling times are more influenced by external factors (weather conditions, hills, road conditions) than other disciplines, therefore their times are somewhat normalized?
Of course, it’s possible that I’m missing something really obvious here or the way I’ve logically interpreted the data is somehow flawed?
I think the %difference you are seeing is an artifact of the relative duration of the legs. For example, if you were 42% off the pace for the bike, all of a sudden you are over 25 minutes behind - that is a difficult gap to make up and still be in the top 30.
Also, due to the exponential factor by which power must increase for a unit increase in bike speed, it is going to be easier to get “close” to the top cyclists times (if not their power output). Looking at analytic cycling, it is pretty easy to make a model where a 7 minute difference (between 58 minutes per 40K and 65 minutes per 40K) is a difference of 95 watts of power output for the same size rider and same aerodynamics. So, even though there was only an 11% difference in speed, there was a huge difference in power output.
Thats what makes non-drafting triathlon interesting. The bike is the longest leg, but to go faster on the bike takes a lot more effort. If you save it on the bike, you might run well, but you also might be out of contention by T2.
I think the %difference you are seeing is an artifact of the relative duration of the legs. For example, if you were 42% off the pace for the bike, all of a sudden you are over 25 minutes behind - that is a difficult gap to make up and still be in the top 30.
Also, due to the exponential factor by which power must increase for a unit increase in bike speed, it is going to be easier to get “close” to the top cyclists times (if not their power output). Looking at analytic cycling, it is pretty easy to make a model where a 7 minute difference (between 58 minutes per 40K and 65 minutes per 40K) is a difference of 95 watts of power output for the same size rider and same aerodynamics. So, even though there was only an 11% difference in speed, there was a huge difference in power output.
Thats what makes non-drafting triathlon interesting. The bike is the longest leg, but to go faster on the bike takes a lot more effort. If you save it on the bike, you might run well, but you also might be out of contention by T2.
Post disappeared for some reason. Re-writing…
I think the %difference you are seeing is an artifact of the relative duration of the legs. For example, if you were 42% off the pace for the bike, all of a sudden you are over 25 minutes behind - that is a difficult gap to make up and still be in the top 30.
Also, due to the exponential factor by which power must increase for a unit increase in bike speed, it is going to be easier to get “close” to the top cyclists times (if not their power output). Looking at analytic cycling, it is pretty easy to make a model where a 7 minute difference (between 58 minutes per 40K and 65 minutes per 40K) is a difference of 95 watts of power output for the same size rider and same aerodynamics. So, even though there was only an 11% difference in speed, there was a huge difference in power output.
Thats what makes non-drafting triathlon interesting. The bike is the longest leg, but to go faster on the bike takes a lot more effort. If you save it on the bike, you might run well, but you also might be out of contention by T2.
If what you said in the first paragraph is true, that means that you cannot be far behind the leaders in the bike due to the duration of its course. Which also means that you have to be a top cyclist to do well.
I calculated the overall positions of the top 30 finishers in each discipline:
Average swim position of the top 30 finishers: ~138
Average cycling position of the top 30 finishers: ~36
Average run position of the top top 30 finishers: ~54
What means that you are right. And this seems to reflect conventional wisdom that the bike leg is the most important leg. Followed closely by the run.
As a swimmer, I think this sorta sucks. But the numbers indicate that to be in the top 1% (top 30 of 3000 participants at accenture) you essentially have to be top 1% of all cyclists (36 out of 3000). It doesn’t get less ambiguous than that…
it was trying to rank the times/differences by importance.
By multiplying the average time of each event by the difference, it creates an importance value
The cycling is longer → more important → less variation
The running is shorter → less important → more variation
so you get
more important → less variation, less important → more variation
when you multiply the average time by the difference you get approximately the same value for the bike and run
indicating the athletes have allocated their time based on the distance of each leg almost perfectly, excluding the swim
I am just making this up so what it means I have no idea
I think the majority of triathletes put in more time on the bike versus the other disciplines. It is pretty obvious just listening to people in this forum that the bike is the “cool” event and matters more than the other. It is also easier to become fast on the bike in a shorter amount of time compared to the other disciplines.
Hard to tell what this means, but if it means what I think it means it backs up what I have been saying for a while - the the run is where you want to really focus on improving all things being equal. Yes the swim could also, be worked on to yeild some significant results improvement as well( redemption for Tom Price - the swim DOES matter!), but due to the placement of the run last, it has far greater importance than these numbers would indicate.
Everyone talks about the bike. Everyone has all the trick bike equipment. Everyone seems to be cycling a lot. Many obsess of bike training . . and so on. However if you are really serious about improving your Triathlon performnace, the swim and particularly the run, are the things to work on.
the difference in time of the #1 finisher and #30 finisher for each discipline
by
the mean time of the #1 finisher and the #30 finisher took for each discipline
I guess the second component might have some correlatation to the distance of the race (although there are probably much better methods to find a more accurate figure), so it’s a multiplier as to how important each variation (first component) is.
And the final figure we come up with seems to be an indication of how significant the difference between the #1 finisher and the #30 finisher is, after trying to account for “distance”.
All these #'s really indicate, to the contrary of the above posters, is that to be in the top 30, one must be within a tight tolerance to the leader, in this case 11%…yes there are diminishing marginal returns but thats not the whole story
.
To give you some more data to analyze, here are some basic stats from the 2007 ITU World Cup Series (average of 15 races). Really all this tells you though is how fast you need to be to compete at this level.
Bike courses are usually technical and are tactical races. More like a road race then a time trial (surges / attacks - never a constant pace).
Run courses are often rumoured to be “approximately” 10km - not necessarily bang-on. That said, it will be close given the stand-alone track or road race times that I’ve seen for the faster triathletes.
Dan, those are great numbers! Would you be able to pull data like that for a USAT event- maybe the Lifetime Fitness series? Reason being that I’d like to see the numbers associated with top overall finishers in open events to help evaluate which disciplines need more work given certain ability levels for each sport. I guess the “Elite” in the title might have been a bit misleading.
My assumption is that being the top x percentage in each category is equally difficult. For example, to have a top 1% time for the swim would be equally difficult as having a top 1% time in the bike. So if my current ability was at the 15th percentile, 30th percentile and 22th percentile for the Swim, Bike and Run respectively, then I know that I need to work on the Bike and to a lesser extent, the run. That much is obvious to most people.
However, I started seeing some flaws in this training approach when I looked at the data I have in my first post. Suppose I was at 25th percentile, 20th percentile and 25th percentile for Swim, Bike and Run, the above logic would dictate that if I should focus more on swim and run for the best returns in overall rankings. However, the data I pulled seems to indicate that to actually do well, you would need to be in the higher percentile of biking and to a lesser degree running, and swimming is relatively insignificant. So according to the data, I probably should spend more time on the bike even though that’s already my strongest discipline by a bit.
I’m hoping that if I explain my thought process it might help explain the kind of data I’m looking for and why.
That’s an interesting point. I tend to agree with Bike times being somewhat more doable relative to the other legs.
Even at the age group level the bike times aren’t too far behind those ITU times. You’d expect that since most people can hit the 60-70 minute mark for the bike leg, the difference of the top finishers would be in the swim and run. However, that’s not the case, since data indicates that being a top overall finisher correlates the strongest with being a top overall cyclist.
Ahhhh, these conclusions are giving me a headache. My data, my logic or my assumptions has to be wrong somewhere!!!
Besides not really knowing what the real distances are, you answered your own question in a general sense. SLow, drafting bike splits= super fast runs. It is often tatical, and the bigger the pack, the slower the ride usually is. You need small 3 to 4 person groups for a really fast ride, and that doesn’t seem to happen much anymore. Usually 8 to 12 from what I’ve seen, and a lot of times two or three of the small groups hook up for one super group. Lots of resting going on in there, so no suprise that some really fast runs follow…Used to be some splits in the swim too, but lately everyone has brought up their game, and it is one long pack, with 1 to 1 1/2 minutes front to back. Very exciting racing in my opinion…
