This is incorrect in my analysis

Tremblant and LP are nearly identical in terms of times. I have done LP 11x and Tremblant 3x. I have probably coached around 30 LP finishes and 25 Tremblant finishes too in some cases same athletes on both courses.

The only reason why your stats suggest Tremblant is that much faster is your data gets skewed by the average 1:19 swimmer drafting forever on the Tremblant course in massive pack.

In LP, right out of T1, things start breaking up quickly.

If you take the guys and girls who are riding solo at the front of the race, you will see they have similar race times. Look at Amber Ferreira's times from last year. She did both 3 weeks apart and had the identical bike time in Tremblant as LP and ran a touch slower. If anyone had to the both races solo, times would be very similar. I have done both courses at the same wattage in training on the bike in the same year with very similar fitness and equipment and end up with almost the same bike split. They have almost the same amount of vertical. Tremblant actually has more vertical, but better pavement so they tend to cancel each other.

Interesting. I wanted to look at the top finishers in all age groups but was getting too far into the data. Your answer makes sense. And I did see more drafting in IMMT than IMLP.

That said, for a middle-of-the-pack M45-49 age grouper it's tough to ignore such a huge margin. No matter how I cut the data, IMMT for an average guy in this age group is faster.

So my answer only applies to M45-49.

i haven't done IMMT but i don't see how the courses can be the same when IMMT has pro women going sub 5 and many more AG men going sub 5 on the bike. IMLP also is easier now with the new out-and-back on the bike.

IMMT bike is 110 miles so the bike times are inflated

that explains a lot. the bike and run both are long at IMLP. why is dev arguing with a straight face that they are equal when it doesn't appear that they are?

Ive done IMMT x3 and LP this year.
1:06,1:05,1:08 swims in MT
5:22,5:19,5:25 bike at MT
3:40,3:33,3:29 runs at MT.

1:04,5:30,3:49 at LP this year.

I would say that the LP bike is harder, I read a lot and consulted Dev before i raced and I really tried to keep both bike loops even and I was within 2 min of each loop. A lot of the hills at MT are rollers you can get to the top of if you carry your speed. There are not a lot of those at Placid

I thought the run course at MT was a bit easier also. That 1.5 mile hill after each loop at Placid is no joke. I was fried this year in the heat at LP on the run. We had great weather all 3 years I did Tremblant.

I loved both races, they are both in beautiful spots, you really cant go wrong racing either one, but if I had to wager on where i would put down the fastest time I would say IMMT.

Well, after 18 years of racing, the course record at IMLP is Lessing's 8:23 in 2004. Not sure if anyone else has been within 15 minutes of that. And, after only 4 years of racing, the course record at IMMT is 8:16 by TJ Tollackson. So, at the pointy end at least, IMMT is certainly a faster race course.

I agree, especially as a 222-lb athlete. The rollers at IMMT favour everybody but particularly the heavier riders. The 3 bears at IMLP were a tough haul. No such thing at IMMT.

Would love to get athlete weight stats to prove that theory.

(For the record, my bike times were 6:26:19 at IMLP 2012 and 6:28:04 at IMMT 2013. So I was slower in Tremblant but I was less trained, felt like crap and couldn't stomach my nutrition. Should have been more than a 2-min difference.)

Sorry, I left out a few things that even the two races out or make them close

• Tremblant swim is generally slower than LP. LP is like a pool. I have always been 3-6 min slower in Tremblant
• LP T1 is much faster than Tremblant. Around 1-2 min quicker depending on when you leave he water
• LP pavement now is good and the new out and back is faster than the old

If you add these three together the two races are about the same. Some will say LP run is harder run with two big climbs per loop, but others find the Tremblant run to be hard in its own way. Tremblant run has a long 1-2 percent incline from the turnaround at the 10K point. It is hard to really see the "climb" but starting at around 31K and having to run up that for 5K is draining in its own way. On the outbound section the first 5K is generally all mainly uphill. Nothing killer, but they add up.

One place where Tremblant wins in terms of speed is the minimal amount of braking. All the potential energy you gain going up, returns as speed coming down. In LP there are multiple places that you have to waste speed in the brakes coming down Keene or riding through town behind the oval and then the T1 exit and the hard left turn at the bottom of that hill. That's just a ton of energy you don't get back. You don't have any of that in Tremblant, so you guys are likely correct that outside of the drafting LP has a bunch of things that makes the bike slower.

One interesting way of looking at this is whether Rapp's 4:25 (or Tollakson's) compares with Steve Larsen's LP record. I don't think these guys quite in the same category as Larsen. Larsen clocked a 4:33 on the old LP course and was probably as aero as Rapp if not more on 2001 equipment. That 4:33 probably translates to a 4:30 on the new LP out and back. I wonder if you put the same Larsen on the Tremblant course on the same equipment as Rapp how fast he goes? 4:20?

dev, you normally make great points but you are talking pure gibberish here.

(1) the swim times are not 3-6 minutes slower in IMMT. that's completely ridiculous and i challenge you to provide any empirical data to support that other than your anecdotal evidence. they both are easy swims.
(2) t1 is NOT faster in IMLP. here are the t1s of the first 5 AGers in both this year (IMLP didn't have pros this year so i don't think it's fair to compare pro t1s to non pro t1s):

5:35
4:37
4:13
3:52
4:17

5:43
3:47
4:29
4:43
3:57

could these t1s be any more similar? 1-2 minutes quicker in LP is preposterous (btw, IMMT is the first group and IMLP is the second group).

(3) IMLP's pavement is not good. yes, the new out and back is faster but that just means that the current IMLP course is a little faster than the old IMLP course but both are A LOT slower than IMMT's bike course. if IMMT really is 2 miles short, that's 5-7 minutes right there plus IMLP's bike is long (the entire part past the top of the oval and behind the school on the second loop is after 112 miles) by probably a minute. that gets you to 6-8 minutes faster even before accounting for the fact that IMMT bike is easier. based on the people who have raced both, it sounds like IMMT's bike rides at least 5 minutes faster so you are looking at 11-13 minutes faster for just the bike.

(4) IMLP's run also is long and the hill in town x2 makes it harder than IMMT's run.

I think to measure transition times you have to take the same person's time on both courses, not 5 guys randomly (even if they are the first 5 people). The physical distance in transition at Tremblant is substantially longer. Easily 100m+ longer. LP transition is downhill run too. I have done a 3:42 T1 transition in LP just jogging downhill and doing things quickly in transition. Same strategy in Tremblant with a ton of congestion in the tent, barely break 5 minutes. It's just physically longer


I think if you are racing pro, the LP and Tremblant swim will be the same. If you are doing it in age group certainly with the wave starts and speed differentials and having to swim through or be swim over multiple people you'll understand why the Tremblant age group swim ends up being somewhat slower than LP. Come and do Tremblant and assess for yourself. I don't think you have done it. Pretty well everyone who raced both will tell you that Tremblant ends up being slower. Something as simple as congestion at the two turn buoys for some of the faster swimming women who start in the back results in them having to take big wide turns that adds 30 m to their distance. Then you have less people to draft off out of your wave in Tremblant vs at LP where after the initial combat (even in the rolling start) you end up swimming with a ton of swimmers of the same ability. All of these add up. I am yet to see anyone who swam faster at Tremblant than LP, but maybe I have not spoken with enough people.

Also on LP pavement, have you ridden there this year? It is pretty darn good now! In any case, on similar fitness (when I did not flat as I did in 2013), my bike times in LP have been a bit sub 5:30 on the old course and Tremblant have been around 5:25....so that's why I was mainly saying between the faster new course, LP's slightly faster swim and transitions, the two races end up being a wash and the BIGGEST difference you see in race times are pack/slingshot/drafting related at Tremblant. I even did a test on the Tremblant course and rode it totally solo one day in training at IM goal watts of 182 and did a 2:47 for the loop, with race gear on. On race day, starting in an older wave and slingshotting past a ton of riders, the same wattage gives me 2:38 on loop 1 and the same 2:47 on loop 2 when I have ridden past everyone and now riding solo. At LP, I am riding solo pretty well after climbing Jackrabbit the first time...and for that entire climb, pretty well going at my pace with no benefit from other riders. I purposely went and acquired all this data on the Tremblant course just to understand how much loop1 slingshotting helps me.....and then there are people sitting in massive draft packs of 30-50 people free wheeling to 5:30 splits after swimming 1:2x. Those same athletes would barely break 5:55 in LP. I see power files from athletes in Tremblant at the 70.3 and there is no way you can ride that fast off that low wattage without a big effect of people around you (legal pacing, slingshot, drafting).

Seriously, go ride both courses, totally solo with a powermeter and the same gear and you will be pretty darn close. Tremblant has more elevation and is a touch shorter, LP has less elevation and a touch longer. What makes one course faster than the other is mainly the dynamics of how other riders on the course affect the splits at Tremblant....they affect the splits "less" at LP.

Don't let actual empirical data sway your opinion, Dev.

Is there anyway to get full Ironman results as a single page or txt/html file anymore instead of 60 or so pages? I'm happy to import some results data into a database and do some real analysis, but need full results file and ironman.com no longer provides that from what I can see.

I was hoping http://www.trirating.com/course-ratings/ would still have LP, but it looks like it was removed (due to no pro race?). Either way, for professionals, IMMT has a course rating of 7:08 while IMLP did have something right around -4:00 if memory serves me correctly, meaning LP is about 11 minutes slower according to the data Thorsten gathered.

how is looking at the first 5 AGers random? it's not. the times of each 5 are almost identical and you are embarrassing yourself continuing to argue that LP is 1 to 2 minutes faster.

you are right, i have not done IMMT but the bike times are MUCH faster there than at IMLP. has a woman ever split a sub 5 on the bike in LP? 2 miles short and 1/2 long is a big difference even if the courses are identical (they aren't).

rather than using anecdotal evidence, why don't you try and use some facts to support your outlandish claims?

it's not like i have a horse in this race and have a faster PR at IMLP and want people to think that it's a lot slower. i just had to respond to your completely preposterous claims that IMLP is FASTER than IMMT where every empirical fact says otherwise.

I couldn't find one file to download either. Ironman.com lists 20/page and Sportstats lists 40/page (https://www.sportstats.ca/...s.xhtml?raceid=26181).

The debate can be pretty easily solved with about an hour of cutting/pasting. My conclusions on M45-49 aren't opinion. They're fact, based on 1,191 IMMT finishers (2012-2015) and 1,555 IMLP finishers (2011-2015) in that age bracket.

Here are the average M45-49 finisher times again:

Swim was identical to within a fraction of a second at 1:19:06. (Strange that it's that close despite the chaos of a mass start and the run over the timing belt at IMLP, isn't it?)
Bike was 6:17:08 for IMMT vs 6:38:11 for IMLP = 0:21:03 faster
Run was 4:53:28 for IMMT vs 5:04:48 for IMLP = 0:11:20 faster
Transition was 0:15:48 for IMMT vs 0:13:58 for IMLP = 0:01:50 slower
Finish was 12:45:31 for IMMT vs 13:16:02 for IMLP = 0:30:31 faster

Here's the data: https://docs.google.com/...gGY/edit?usp=sharing

Personally, I don't care which course is faster for the AG winners or pros. I wanted to see the results of my competitive set and I'm targeting top half. Overall averages are misleading for everybody.

My main point is that Tremblant appears to be faster than it should be due to all the drafting and major slingshotting going on, so pulling down data from results files will skew the reality. Probably better to take performances from the same pro athlete on each course and you will be able to compare better assuming said pro had similar performance days.

I really don't get your animosity and and aggression when I post though. It's been like this for years and you seems to be carrying it over to this thread. You have not even been on the Tremblant course once and gathered first hand data. I am just saying results files will paint a largely skewed view because of draft packs. If all these guys at Tremblant raced close to solo, the times would be a lot slower.

The pros race solo, and despite there being only 4 years of IMMT to 18 years of IMLP, the course records for both men and women are well faster at IMMT.

there's no animosity and your posts usually are well thought out and very informative. however, you are dead wrong here and with each additional post of empirical data showing IMLP to be considerably slower you are looking more and more dogmatic on this subject. btw, there's no drafting in t1 to skew the data. it doesn't really matter to me which course is faster but i did feel the need to step in after reading some of your patently wrong claims about the two courses.

LP bike is definitely harder and slower. By how much depends on the individual, and where you are relative to the swim packs.

Just last week I rode the LP full course 1 loop, using the exact same NP target watts I raced MT in last year, and the loop time was nearly identical, and only a couple of minutes slower.

HOWEVER: Caveat Emptor!!!! For those who use power, you know that you can get measured NP up fairly easily with spikes that in my opinion are not the "same quality of watts" that you get when you ride continuously moderate-to-high wattages. In other words, I was absolutely more tired after 1 loop of LP at the same NP watts I raced MT in. So, when I do LP next year for my first time there, I will be conservative in my watts target, and "swing low" just a bit to save for the harder run.

More detailed explanation of wattage rationale: I do a regular 1 hour ride that I call my FTP hammer ride. It's my proxy for the proxy (FTP itself being a proxy), when I don't feel like the 2x20 protocol, and I don't have mountains near by to climb on days I want to test. On this FTP hammer ride, some days I do it by trying to hold the highest *continuous* wattage that I can hold for the 1 hour, and other days I allow myself some respite from time to time, and hammer up the small hills out of the saddle, spiking the watts. For those familiar with how NP calculation works, your high spikes get a "to the 4th power" component to them that quickly jumps the FTP/NP number up. In my humble opinion, for these 2 different protocols and riding styles, watts do not equal watts. Or rather, NP does not equal NP. IMO. I have lots and lots of power files, and many of these FTP hammer rides to call upon, as my meta-data set.

Why this matters to the MT vs LP discussion? => In LP, you don't have any "spikey" climbs anywhere, but you have a LOT of continuous power output. It's hard to hold your target average watts for 5 or more hours. It hurts. By contrast, MT allows coasting pretty regularly, with a lot of rollers in several parts of the course, compared to LP which has one big honkin' descent (into Keene) and then flat-to-up back to downtown Placid. MT lets you get frequent/short periods of resting your legs without a big hit to your NP calculation, and then the NP goes back up when you hit a spikey climb.

My 2c worth. It's not precisely scientific, I'll grant you that. But I do have a science/math/engineering background, so I don't think I'm just making sh*! up. :)

Isnt the swim at IMMT one loop where as Placid is two? I have done LP 3 times and find I loose a lot of time having to swim the shallow water and get out to run across the beach to renter. That would make the LP swim slower I would think.

I did IMMT last year and came up with 112.07 miles. I've done a rollout calibration and generally find my distances match reasonably well with what others are coming up with. Not impossible that I could be a little bit high but highly doubt I'd be 2 miles off.

I've done MT multiple times and come up with pretty close to 56 per loop, too - aligning wheel-revs and GPS. But I didn't want to get off-topic. I don't think distance is a contributor to the discussion.

I had 110 on the number as did every other Strava file I've looked at from the race this year.

I did the race in 2012 and remember a hill in St. Jovite that seems like it's been cut.

That cut hill (from St. Jovite in '12) was replaced by a lengthened stretch on 117. You now go all the way to Labelle before turning around, which was not the case in '12.

One thing that's important to note: depending on what device is used to capture the data loaded onto Strava, it may under-represent the true distance, since GPS only samples periodically. Some more periodically than others. The true distance can only be measure in-line by wheel-rev counting which was been shown to be accurate to GPS over straight-line comparisons. This is obviously more pronounced in MTB riding where the trails are constantly changing, but enough turns left/right, and periodic sampling could under-measure the distance - that turn looks like a straight line when you only sample every 10 seconds or longer for position.

I have no dog in this fight, but I can tell you that the exact same Joule (which uses wheel-revs) and exact same wheel were used to measure both MT and LP, and they were both close enough to 56 per loop to not make it worth discussing. So, arguing in the alternative, if MT is short at 110, my Joule would have also said 110 for LP, negating the one vs the other argument.