Two corrections: the hill, then the stroller

SAP starts from the pace your watch measures and corrects it twice, once for the slope and once for the stroller. Both corrections come from published lab studies, and the stroller one can be tuned as we learn more.

The hill

Running uphill costs more energy per mile and running gently downhill costs less. In 2002 Alberto Minetti and colleagues measured how much, with runners on a treadmill tilted anywhere from 45 % down to 45 % up, and published a curve for it. On a 5 % climb a mile costs about 1.3 times what it costs on the flat. On a 5 % descent it costs about 0.76 times.

Parent Pace reads the slope from your watch's barometric altimeter, over the last 10 seconds or 25 metres, whichever is longer. It looks up the cost on Minetti's curve and divides your pace by it. The result is grade-adjusted pace, shown on the watch as PP GAP.

The stroller

A stroller costs you in two ways. On the flat, pushing locks up your arm swing and changes your stride, which adds about 7 %. Going uphill you're also lifting the stroller and your child up the slope, so the extra grows as the hill gets steeper and as the load gets heavier compared with you.

• F is the flat penalty, 0.07 by default.
• H is the hill coefficient, 3.5 by default.
• Load ratio is everything you push divided by your own weight. A 32.3 lb stroller and a 33 lb child pushed by a 153 lb runner gives 0.43. Your weight comes from your Garmin profile; the stroller and child weights are settings.
• Grade only counts going up. On a descent the hill part is zero and only the flat 7 % applies.

Four examples

A 153 lb runner pushing a 32.3 lb BOB Alterrain Pro with a 33 lb child, using the default F and H.

On the 5 % climb, GAP and SAP are 70 seconds per mile apart. That gap is the stroller. On the 10 % climb it's 95 seconds.

Where F and H come from

Only one study has measured stroller running uphill. Sandbakk and colleagues (2020) had 17 runners push a 24.3 kg (54 lb) loaded stroller on a treadmill at 1 % and 10 %. Pushing raised their oxygen cost per kilometre by about 8 % at 1 % and by 19 to 20 % at 10 %. Fitting the formula to those two points gives F = 0.07 and H = 3.5.

The flat figure agrees with an earlier study. Alcantara and Wall-Scheffler (2017) estimated that a stroller adds 5 to 8 % at a steady speed on a track, depending on how you push. Gregory and colleagues (2012) found smaller and mixed changes in oxygen use at 10 minutes per mile.

That's not much data: one uphill study, 17 people, one stroller and two slopes. Nobody has measured the 3 to 6 % grades most of us actually run. So F and H are settings, not fixed numbers, and the anonymous run summaries people choose to share are how we'll check them.

What it doesn't do yet

A stroller gives some energy back on a descent, but you brake it, and nobody has measured the net effect. Until someone does, only the flat 7 % applies downhill.

How you push matters a little: one hand or two changes the cost by about a percentage point in the 2017 model. Most people switch during a run, so the field ignores it.

The field works at walking speed, but Minetti's curve is for running. A walking mode is planned.

Sources

1. Minetti A.E., Moia C., Roi G.S., Susta D., Ferretti G. (2002). Energy cost of walking and running at extreme uphill and downhill slopes. Journal of Applied Physiology 93:1039–1046. doi:10.1152/japplphysiol.01177.2001
2. Sandbakk Ø., Perl R., Holmberg H.-C., Steiner T. (2020). Energetic cost and kinematics of pushing a stroller on flat and uphill terrain. Frontiers in Physiology 11:574. doi:10.3389/fphys.2020.00574
3. Alcantara R.S., Wall-Scheffler C.M. (2017). Stroller running: energetic and kinematic changes across pushing methods. PLOS ONE 12(7): e0180575. doi:10.1371/journal.pone.0180575
4. Gregory D.A. and others (2012). Physiologic responses to running with a jogging stroller. International Journal of Sports Medicine 33(9):711–715. doi:10.1055/s-0031-1299748