Footstrike type and zero drop shoes


My usual footstrike metric is a mid-foot strike. This is both what the RunScribes show and what video shows. However when I run in a pair of Altra’s, which are zero drop shoes, my measured footstrike metric is heal strike. Video shows, as you would expect, that running in the Altra’s is closer to a forefoot strike than a mid-foot strike. I see this with three different models of Altra’s.

Just curious if anyone else running in Altra’s and what do you see for a footstrike.

Maybe the RunScribe is interpreting my forefoot slap as a heal strike - the slap is more pronounced in the Altra’s than in other shoes.


@runnerdave I run primarily in Altra’s and can attest that my metrics are quite different in Altra’s than in other shoes, namely Saucony’s, ASICs or Hoka’s. There is apparently something with the construction of those shoes that channel energy very differently from other shoes. I primarily run in Paradigms, but the results in One 2.5, Torin and even the Olympus are similarly different from other shoes.

The foot strike number with the pods heel mounted appears to be very accurate in my case, and correlates with my shoe wear pattern. When the pods are lace mounted, I get a more forefoot number. In other shoes my foot strike number is nearly always more midfoot when heel mounted.


Thanks Alang -

I’m a mid foot striker with all of my Hoka’s and my Sketcher Ultras but RunScribe records me as a heal striker with the Olympus, the One 2.5 and the Provisions even though video shows a forefoot strike in those shoes. I suspect too, that the shoe construction might have some effect on the data capture - I have mine on the heals.


@runnerdave, @Alang The shoe thing is interesting. I will run a double check between Altra zero-drops (purchased last December) and an old zero-drop NB Minimus Trail Zero’s. Relative to the NB’s, The Altra’s have a much higher stack height and soft soles.


  1. If either of you experiment with your endurance pace gait to shift from slightly more rear foot to slightly more forefoot, how exactly do you do that? What cues do you use? I would like to be able to go out an run 1/2 mile forefoot, and 1/2 mile rearfoot and see that show up on the rS graph.

  2. I’m still searching for a zero drop or near zero drop set of running shoes with a low stack height and slightly more cushion than available in the NB Minimus series. Recommendations for trials?


Zero drop shoes are supposed to more naturally move your footstrike forward. That is why I was questioning whether or not the RunScribe was accurately picking up the footstrike in the Altras - as I am normally a mid-foot striker and the Altras all register as heal striking. I don’t do anything different, or try to do anything different.

If you want to try and change your footstrike for the sake of measurement, I guess I’d try to increase my cadence as much as possible while decreasing my stride length to try and force more of a forefoot strike, and conversely try to decrease cadence and take as long a stride as possible, swinging at the knee, to try and go more towards a heal strike.

As far as the shoes - I’ve never run in the NB Nimimus so don’t know what they are like. In the Altra - I think the One 2.5 is the lightest, or near lightest and smallest stack height. They have very minimal support.


@runnerdave out of curiosity what does your shoe wear pattern show? My wear patterns have lined up quite well with the rS indications over the past 4-5 months.

This is exactly what I do too. For me, it requires a lot of focus to keep it up and consciously try to land on my forefoot. Possibly another characteristic of Altra’s is that going by feel is pretty deceiving for me. A Heel strike in the 3-4 range as measured by rS doesn’t feel much different to me than an actual Mid foot strike.

Tangentially, in my case maintaining a higher step rate and shorter stride length requires less power to maintain a pace than does my more natural stride, as measured by a power meter.


@Alang, OK, just out of curiosity I will do the following experiment (which will have to be on a track I guess just to mnimize confounding variables):

warm up, which for me is a slow mile or two.

equip with rS on each foot and power meter.

start stopwatch
run 1/2 mile @ maybe 9:30 pace with great concentration on landing with tibia vertical, midfoot strike. attention span has been a challenge for me on this kind of thing.
click lap button
run next 1/2 mile at same pace, but higher cadence, concentrating on more forefooty strike (this way is more natural to me)

should be able to see the shift on footstrike in rS data and see whether there is a measureable difference in power output or surrogate VO2 (heartrate surrogate).

will post graphs


results of rehearsal run on .38 mi pavement loop, gentle grade(s) as opposed to track.

things don’t always go to plan. stopwatch malfunctioned, only got time for one of the loops, rS reports wicked fast pace (7:20’s) vs actual (8:57) . Graphs showed two running segments of approximately equal pace separated by a pause as planned, but it was hard to distiguish first segment (aimed to shift toward midfoot) from second segment (aimed at running normally).

The thing that did show up was that BOTH segments showed both feet squarely in the midfoot strike numbers, which has never happened in the entire time I’ve been using rS (since Jun 2015).

Couldn’t really distinguish segments by comparison of footstrike type or flight ratio.
Power output was likewise very nearly identical.

Shoes: the original NB Minimus Trail Zero’s with zero drop (not to be confused with later Minimi that have 4 mm drop)


results of second rehearsal run on .38 mi pavement loop.
Another incomplete - this time the stopwatch worked fine, power meter app did not…

…and the runScribe…

despite carrying the rS to the pavement loop test track in a little slung camera pouch, inserting before test, and removal right after post test standstill, it apparently kept taking data in the pouch on the walk back and got so confused it is still ‘processing recently uploaded run’ 24 hrs later.

So the rS does not stop taking data after running activity even in a pocket or slung pouch.

In the test of the post immediately prior to this, I carefully carried the two rS’s in my hand trying not to swing it or let it vertically oscillate while walking back. That worked better, - it kept taking data, but at least it uploaded and digested the non-running part, which could be trimmed off.

Are there any tricks to stopping the data taking after activity?

Next I’m going to drive the quarter mile, do the test, maintain vibration-free transport back for BT connect & upload.


My recent routines ending a run are to either walk about 400ft up my driveway or about 50 ft to either my car or hotel.

The rS usually do a pretty good job determining when I stopped running. For the level of precision you’re looking at for this test I would likely have to trim a little off the end.

A couple of thoughts are to stand still for 60-90 sec at the end of your test, or to grab your phone, connect to the pods with the app, and hit the sync button. You don’t have to actually sync the runs at that point, I believe simply starting the sync will tell them you’re no longer running.


Thanks Alan. Thanks a lot!
I run with my iPhone for these tests ( side-strapped to the rigid portion of my hips), and that would be quite convenient.

Reiterating the goal here: capture metric showing distinct difference in intentional midfoot vs forefoot landing on short runs at 8-9 minute pace. And hopefully identifying the difference in how it feels.


Alan is correct, starting a sync is something I regularly use during my testing to end the run.


@rcarter1000 I think the bouncing around in your pocket or pouch may be causing some of the issues but I can’t say for sure. In our experience, the transition from running to walking is enough for the post processing routine to clip out the walking at the end of a run as Alang suggests.

Not sure if you are carrying your phone but Alangs technique is a good way to end a session quickly. So if you want to end a session but not upload at that time you can open the App, make sure both scribes are connected, hit the synch button and as soon as you see the progress bar appear, cancel the synch. This will stop the collection of data and leave the scribes ready to start another trial if you want. Then later, when you are ready to upload the files, perform a synch as normal and let it complete. All the files will be uploaded to the server. Hope that helps.


OK, thanks to all the above advice, I finally got all devices to work for two consecutive short runs
AND uploaded (i.e. Stopwatch, Runscribe, Stryd). The ‘stop rS’ using canceled sync works GREAT!

As planned one run was attempted midfoot landing, and the other was
forefoot (up on the balls of my feet, no heel touch).

See summary of results below and screenshots for addtional data.
Link to run if you are interested is:

My comments:
Stopwatch was the primary timing device & was clicked on cue. Others were picked off graph with endpoints
as close as possible to estimted stopwatch start/stop points.

Shoes:early model NB Minimus Trail Zero’s (zero drop)

On the midfoot stretch, I was trying hard to focus on landing with shin upright and feeling a square foot landing.
On the forefoot stretch, I was trying to land squarely on the ball of the foot every step and did. I do not
have calves of steel, and could feel the heels dip down everytime after impact however they (almost) never touched
Since most of my other rS run records show footstrikes of 12-13 for BOTH feet, it’s clear that I affected the
gait by intention, and that rS picked it up. But I am not sure what it’s telling me about my footstrike.
Does not appear that forefoot strike on right foot was detected. Why might that be?

runScribe reported pace was a little fast, could be reduced by calibration. Normal running gait on this
same road results in accurate pace reporting. Agrees fairly well with Stryd GPS based pace report.

cadence report also agrees between the two devices.

Flight ratio was higher for forefoot strike, which seems reasonable - there was more ‘launch’.
Stryd wattage indicates that possibly (I take all this with a grain of salt) it cost me
more energy to make that launch than I saved by being in the air instead of on the ground.

That’s the report. Have to think about what to try next. Other experimental results and/or comments are

    midfoot        forefoot

distance 0.39 0.39 mi
ave pace 8:46 8:43

ave pace 8:16 8:10
ave cadence 192 187
flight ratio L=7%|R=7% L=10%|R=9%
footstrike L=13.8|R=9.9 L=14.0|R=8.4

ave pace 8:59 8:40
ave cadence 190.8 185.2
ave power 186.7W 202.3
HR N/A N/A run not long enough to stabilize