Route: Presidential Traverse (NH)

New Hampshire, US
18 mi
Vertical Gain
8500 ft

Wikipedia describes the Presidential Traverse as follows: "A Presidential Traverse, as it is known to hikers in the Northeast, is a strenuous and sometimes dangerous trek over the Presidential Range of New Hampshire's White Mountains. Contained almost entirely in the 750,000-acre White Mountain National Forest, the Presidential Range is a string of summits in excess of 4,000 feet. To complete the traverse, one must begin at either the northern or southern terminus of the Presidential Range and finish at the opposing end. As there are multiple definitions for the Presidential Traverse, what happens in between is a question of debate..." While many hikers have claimed that doing the traverse without tagging the summits is valid, this seems counter to the definition of a range traverse. The minimum traverse, from Madison to Pierce, is 18+ miles, according to recent GPS tracks from some of the FKT runs. For most hikers, the northern trailhead is near the town of Appalachia, and the southern trailhead is Crawford Notch. A good topo map is here. This page gives distances, elevation gains and "typical" hiking times.

The Picnic:  Run the Presi, close the loop by bike.

Note about Mt Clay:  As far as we can tell, most recent FKT aspirants have gone over the summit of Mt Clay via the Mt Clay Loop Trail.  Malcolm Groves did not summit Clay, but rather just contoured around the summit via the Gulfside Trail. Groves seemed to indicate that since Henry Clay was not a U.S. President summitting his namesake peak would not be required for a "true" Presidential Traverse. Groves beat Ben Thompson's earlier time by 2m43s.  However, Thompson did summit Clay, so it is worthwhile to try to estimate how much time Groves might have saved by skipping that peak.  Since both runners submitted GPS tracks via Strava, this is relatively easy.  It took Groves around 14m13s to do the Clay section via the Gulfside Trail, and it took Thompson 17m34s via the Mt Clay Loop Trail, a difference of 3m11s.  This comparison suggests that Thompson should still be credited with the FKT, but only by about 30 seconds!  We would suggest that future FKT aspirants should summit Clay, as seems to be more standard.

Further route note (IMPORTANT):  As best we can tell, the Presi has never been defined, except that you have to do all the peaks, and it's TH-TH.  Therefore, you can start and end at any TH.  So you should obviously do whichever route (and direction) you think with be the fastest for you.  Most of the earlier FKT runs used the Daniel Webster Trail.  More recently, Valley Way has been used by Ryan Atkins (August 2018), Lee Berube (September 2018), and Jordan Fields (September 2019). Comparing Daniel Webster with Valley Way, the distances and vert aren't that different, but the Valley Way trail seems to be faster.  Here's a comparison of distance / total elevation gain for some recent FKTs from Strava:

-Jordan F 30.11km/2536m

-Ryan A 28.96km/2662m

-Ben T 29.44km/2702m


Judith Maddock Hudson wrote of the early history of the Presidential Traverse for the Randolph Mtn Club Newsletter:  

The first "marathon" in Randolph's records was a relatively relaxed walk, undertaken in September 1882; its participants were Eugene B. Cook and George Sargent. 3 The pair left the Ravine House at 5:03 a.m., traversing Madison, Adams, Jefferson, Clay and Washington, where they spent an hour and a quarter, "during which time we luxuriated in refreshing rest, feasted with the greatest relish and desire, and ended by enjoying the fascination of the inexhaustible view." They continued over the Southern Peaks, descending to the Crawford House, and then returned to the Ravine House via the Cherry Mountain Road, arriving at 1:24 a.m., where "our worthy host appeared and welcomed us in, [and] we found an inviting table awaiting the prodigal walkers." Total distance, 42 miles, 1,212 yards; total actual walking-time, 17 hours, 33 minutes, with an additional 2 hours, 45 minutes devoted to rests as well as meals at the summit of Washington and at the White Mountain House (at the junction of the Cherry Mt. Road, northwest of Fabyans). At the time, Cook was 52 years old, and Sargent was still a young medical student (who had first come to Randolph in order to improve his frail health). Their goal had been to accomplish the trek in a single day, and they were not walking for speed (they averaged 2.1 miles per hour for the total time elapsed).

Cook's and Sargent's "audacious tour de force" (so termed by Appalachia's editor) was not replicated for many years. The next documented range run occurred on July 12, 1904, when Appies Herschel Parker and Warren Symonds began the trek from the Crawford House at 6 a.m., first ascending Webster by trail and then having to bushwhack to Jackson and on over to Clinton. Parker wrote: "The exertion required to reach the summit of Mt. Clinton via Mts. Webster and Jackson appears to the writer to be fully equal to that required to traverse the remainder of the Crawford Path and all the Northern Peaks to the Ravine House." 4 The two reached the Ravine House at 8:15 p.m., "where supper was taken and a lantern kindly furnished us by Mr. Watson...we set out for our long walk back to Crawford's in inky blackness with the rain descending." After sheltering from a thunderstorm in an empty freight car, they made their way over the Jefferson Notch Road 5, down the other side, and reached the Crawford House at 4:15 a.m.. Total mileage, about 43 miles; total time, 22 1/4 hours, an average of about 1.93 miles per hour total elapsed time, though with less than 18 hours actual walking time. For both parties, the range walk was a gentlemanly pursuit, more a test of endurance than a quest for speed.

Ryan Welts seems to have ignited recent FKT activity on the Presi when he ran the Traverse (including Pierce) in 5h7m44s on Sept. 5, 2009, about 4 weeks after setting the FKT on the Pemigewasset Loop.  Prior to Welts, the fastest time we are aware of was from an unofficial "race" of one version of the route (not really the "classic" route) on Sept. 13, 1988. Ian Torrence and Steve Smith ran the race from Dolly Copp campground to Crawford Notch in 5h25m on September 13, 1998, summitting Madison (via the Howker Ridge Trail), Adams, Jefferson, Clay, Washington, Monroe and Eisenhower (but not Webster or Pierce) along the way. The race provided no aid stations, but the runners refueled on top of Mt. Washington. Torrence felt it likely that others had done the traverse faster, but knew of no faster time.

HOWEVER: It has been reported by an anonymous poster that Mike Gallagher and Ned Gillette ran the Traverse in 4h46m way back in 1968, while training for the US Olympic Nordic Ski Team. This is being reported based on Gallagher's training log from the time, though we have not see it. It is not clear if the route they used is the same as what is being done presently.

GPS Track


There's been a defined route for the Presidential Traverse for decades and it involves summiting Clay. The point is not to summit every President's peak, but to follow the Presidential Ridge. And the ridge definitely includes the summit of Clay.

That said, Malcolm's time is impressive and fast! I'd be excited to see what he could do if he gave it another shot on the correct route. Take it under 4:20!

Aside: My impression is that the only choice in the FKT route is which way you go up Madison. Lots of people take Valley Way and Daniel Webster. Ben Nephew tried Watson's path once. I'd bet all three are within a minute of each other and play to slightly different strengths/plans.

I agree with Ben. It's on the runner to stay on route or make any variations at least as difficult. Skipping Clay is a disqualifier in my opinion and something that could have been prevented with a little research and planning. There's clearly some impressive ability on Malcom's part here. Show us what you can do on the correct route.

Also agree with Ben T here. I’d be fine with seeing Malcom’s time stay up since it’s clearly marked and explained, but that’s not my call. Regarding route up Madison, Ben N and I once agreed that we’d be fine with any ascent route (Daniel Webster is tough to get to, and Ryan only chose it initially “because it’s hard”). Glad to see Ben T agrees.

Edit to say Malcolm’s time should be moved down below Ben’s on the list if possible, to make it clear that it’s not the time to beat. 

On Madison ascent routes: I think Pine Link should be off limits simply because it starts several hundred vertical feet higher than the other trailheads (~1700ft vs ~1300ft) and isn't any longer. That'd probably cut several minutes off the ascent.

Agreed. Clay must be included. I was under the impression the power to amend the route only belongs to the current holder. When I resurrected this dormant FKT in 2009, Ian Torrence has ascended via Howker Ridge but had not included Clay. I chose to ascend via DWS because I though it was more aesthetically pleasing. I also chose to add distance by adding Pierce and still bested their time. Malcolm, your time is very impressive, but let’s follow the “rules” to keep the competition and spirit alive! Looking forward to your next attempt!

Sorry for the drawn out comment ?

Ben Thompson's time is the FKT, He ran the full route. If someone runs an ultra course that is short, it is not counted as a record. There is no extrapolation. I also don't agree with the choice of picking a route up Madison. I did always DW to set the FKT. I only tried Watson's Path because I had the FKT at that time. Valley Way is a quicker way up, and it would be too bad to have Ben Thompson lose the FKT to someone doing a different route. Changing almost 25% of the route is quite significant. While Ben is OK with this, if everyone starts doing this, it defeats the purpose of having an FKT where you can actually compare times. Why not change descent routes, or 4 miles in the middle of routes? The logic that allows an optional start would certainly justify something like skipping Clay if there was little difference in elevation gain.

My bad, I guess I misinterpreted our conversation when you planned to try Valley/Watson instead.

But since Ben T is the record holder it’s his call now, right?

I am planning to make an attempt for the Presi FKT in the fall. Due to the logistics of the one way route and shuttle system, I plan on starting the run from the Valley Way trail unless a legitimate reason if given for why this should not be considered an FKT. Aesthetically, I like the Daniel Webster trail for this traverse; however, I believe the Valley Way trail improves the logistics of this FKT for those who do not live in the area. 

Lee, we're having trouble viewing your Strava activity.  I think it must be a privacy thing.  If you log out of Strava and try to view it you'll probably see what I mean.  Please check the privacy settings, and make the activity public.  Thanks!

Hello, I am planning to attempt a prezi FKT this fall, should weather/timing permit. However, I want to clarify a few things about this route here before attempting:

1. Is the Valley Way ascent with the out and back to Madison now the accepted route?

2. Neither Lee nor Ryan seem to have actually summited Jefferson -- they simply ran past the summit cone but did not ascend the rock pile to the summit. Is this acceptable? (See their strava tracks to confirm this). It is not a huge time addition but at least a minute, given the rough scramble to the summit marker. 

3. Lee used the trail along the cog railway to the summit of Washington, rather than the trail further towards Great Gulf. It certainly saves some time, is this acceptable? 


Your run is super inspiring!! 

Just wanted to chime in and say that:

1) I think Valley Way should be acceptable though not required.  

2) Bummer that folks skipped the Jefferson summit. We shouldn't do that!

3) Yeah, I think people should stay on the trail in the summer!! In the winter, I think going off trail should be okay and only tagging the peaks (Madison, Adams, Jefferson, Clay, Washington, Monroe, Eisenhower, Pierce) should matter since it's fairly standard practice to go off trail in the winter. There's no risk of hurting the alpine vegetation when it's covered in snow... The only reason I bring this up is that I'm fairly convinced that the Presi Traverse would be faster on a perfect winter day with the rocks covered up.

I just set a new FKT on the Presidential Traverse of 3:42:38. Full report soon. I had one snafu and took gulfside by mistake instead of airline up Adams so I had to go back and tag the summit. This added at least 5 min but no other major route issues. GPX file on strava, link:

Since Jordan raised some route questions, I also wanted to ask what people think about a S->N traverse? Would that be acceptable for an FKT? The southern end is over 600 ft elevation higher, so going south to north would involve 1200 ft less elevation gain. I think S->N has not been done because hitting the northern Presis while tired might be challenging and the descent going N->S down the southerns is so gentle and fast. But, I think that for a ridge like this, either direction should acceptable. Similarly, for a loop, clockwise or counterclockwise should be acceptable. 

Yes, you can go either direction.  Part of planning for an FKT is deciding which way is faster for you, and picking the best conditions.

For routes, while there are options for summiting some of the Northern peaks, Valley Way is required at this point since this is the current FKT route.  There has been some choose your own adventure in recent years, with runners taking different trails that make it impossible to directly compare times.  I doubt this is going to be an issue now, as Valley Way is the fastest route, but want to make sure the choose your adventure doesn't spread to other routes.  For staying on the trail, I think this is critical to routes such as this, where going off trail (intentionally....) and posting evidence of this may present some issues with the WMNF.

First, congrats Jordan on the new FKT! That is some seriously fast running on really technical trails, which anyone who has run or hiked the traverse can appreciate. I like that you used Beacon for the effort and think it would be cool for you to run the traverse South to North too to see how your efforts compare. To answer your questions, it was my first time running the traverse, and although I looked over maps, I was figuring out the trail on the fly. I would have a much better understanding of the route if I ran it again. I used the trail along the cog railway because Ben Nephew used that variation in his 2013 effort. I personally think that slight variations and new ways of running a route (like starting from Valley Way or running from South to North) make these experiences unique and meaningful to the individual that runs them. 

Hi all, thanks for weighing in here. I just wanted to agree with Lee that slight variations I brought up are not a huge issue in my eyes. I brought them up before my attempt only clarify the route so as to avoid any potential questions afterwards about whether I was on the correct route, not to question the legitimacy of his FKT, which was impressive!!

However, I agree with Ben Nephew that we should stick to a common, accepted route for the majority of the traverse, i.e. Valley Way, hitting all the summits, and descending Crawford Path. I doubt it will be an issue going forward now that we've clarified it here for future runners.

A common, accepted route is not exactly what I was saying.  To compare times, you need to run the route you are trying to get the FKT on, it doesn't have to be accepted.  This is what is done on the vast majority of the other FKT's, with very few exceptions.  Variations make it difficult or impossible to compare times, which is the main objective here.  It's a race course.  Everyone ran the same route as the FKT holder for many years, this recent issue has only been happening for a couple of years.  Not sure why, I don't see it happening on other FKT's.  Some of the trails to the ridge are vastly different, even if the mileage is similar.  Valley Way is a sidewalk compared to Daniel Webster or Watson's Path.  Very few of us would have chosen Daniel Webster, it was not accepted, but that was the FKT, much to Ryan's amusement.


Not sure if the "Presidential Picnic" is a joke or not, but I did the traverse N-S and bike shuttled a couple days ago solo in 11:22:08 (#covid, no shuttle & hitchhiking being sub-optimal this summer). Hike was unsupported, but I picked my bike up in Crawford, so I suppose that would make this self-supported? If this category is real, unsupported would mean Presi hike-a-bike to Jeff Notch bike, right? Makes me laugh, but I can think of 5 people reading this thinking about strapping a carbon frame to a packboard...

Good hike, fine weather. Left Appalachia at 6:05am, hit Crawford at 3:15pm or so, moving for about 7 hours, elapsed about 9. Didn't really run. Couldn't get hungry all day, which for me means borderline bonk. Cheers to the hut croos / CC that put in the taps at the high huts, despite not opening this summer! Bike over Jefferson Notch took 90 minutes or so. I bet someone fast could do this in 5:00 flat!