I decided to spend a little time breaking down the performance of American Pharoah in the Travers Stakes. The common theory is that he “bounced” and didn’t perform at his best in the race. The data doesn’t seem to support this theory though as his Beyer Speed figure and his Timeform Figures were in keeping with his other performances at races beyond 9 furlongs.
Being subject to an inefficient race pace is what did him in.
As I have previously stated, American Pharoah has a longer than normal stride. His average stride length (not his peak stride) in his races at a mile and beyond is 7.69m/25.23ft. What makes American Pharoah really hard to beat is that he uses this long stride efficiently and does so by carving out quarters consistently with about 52 strides in the quarter, which equates to times around 24 seconds and speeds of around 38mph. What makes him special is that if he is not asked to do something extraordinary in a race, he can do four of these together which only the really good dirt horses can do resulting in halves done in 104 strides at 48 seconds and a mile completed in 1:36. Given that the stride length of a horse is linked to inspiration/expiration, that is, there is generally a 1:1 coupling of stride to breath, the more strides that a horse takes the more breathing and work that it has to do to get itself from point A to point B so having a very long fluid stride which American Pharoah has separates him out from the rest who on average will take between 58-61 strides over the same quarter mile distance.
According to the Trakus Data from the Travers Stakes, which you can access here, American Pharoah ran the first quarter mile in 24.29 and the second quarter mile in 24.36.
I ripped the video of the race from YouTube and used the freeware program Kinovea to roll the video to the half mile pole and then subtract the time off the video (Kinovea has a very useful stopwatch feature to time the video on) and counted his strides from the start (actually it was the pole just after the start) to the first quarter and then from the first quarter pole to the half mile pole. As I have said earlier, he is amazingly consistent with 52 strides taken for both segments. He does that first half of the race as efficiently as he likes.
Once they reach the half mile pole is where it gets interesting.
American Pharoah completed the quarter from the half mile to the ¾ pole in 23 seconds flat at 39.3mph. He did this by taking 51 strides. To go faster, American Pharoah lengthens his average stride by half a foot from 25.23ft to 25.73ft. The next quarter, from the ¾ pole to the mile pole, is completed in 23.66 at 38.4mph. He does this section in 50.5 strides which works out at an average stride length of 25.98ft. It shouldn't be lost on you that he extends his average stride to 25.73ft at 39.3mph but then extends it further to 25.98ft but his speed drops 38.4mph...a longer stride but slower time.
The first half was done in 48.61 seconds, 104 strides for 2.13 strides per second. The second half was done in 46.66 seconds, 101.5 strides for 2.17 strides per second. American Pharoah ran his final 2f split in 26.38, three seconds slower than his previous two splits, which highlights the extent of his fatigue. Compare this to his Belmont Stakes win over 1m4f, where he might have been expected to tire but didn't at all, reeling off closing splits of: 24.58, 24.34 and 24.32.
In studies on the thoroughbred and the relationship between stride length and stride frequency, both stride length and stride frequency almost linearly increased with increasing velocity. However, at higher velocities, stride length tended to show a decreased rate of increase, whereas stride frequency tended to indicate an increased rate of increase.
To gain greater velocity, American Pharoah doesn’t increase his stride frequency as much as he needs to. He has a long stride (and no doubt an amazing cardiovascular system) but most likely through a mechanical limitation, he doesn’t lay down strides more frequently to increase his velocity, rather he grinds out at a similar rate resulting in between 2.1 to 2.2 strides per second.
This makes him a particularly pace sensitive type of horse as he never reaches 40mph in a race, so he can’t break 23 seconds for a quarter and a horse can only ‘lengthen its stride’ as American Pharoah does for so long in a race. As a comparison, Secretariat on the other hand was able to break 23 seconds regularly, and did so with an internal fraction in the Belmont itself. Secretariat had a shorter average stride than American Pharoah does, but could turn them over quickly and keep doing it resulting in him taking closer to 2.4 strides a second in races.
Had Frosted not taken American Pharoah out of his comfort zone like he did, resulting in him running 1.4 seconds faster than he normally does in a quarter, I am sure we would have seen him win easily. Had he been allowed to put an efficient second half of the race in 48 seconds/104 strides he’s have gone through the mile at the same pace as the winner Keen Ice (1:36.00/38mph), he’d have closed out the race like he did in the Belmont in 24 to 25 seconds and won the race in roughly the same time as Keen Ice did. Because he can’t truly sprint in races beyond a mile and run a 40mph sectional, the pace of the race can be his undoing. If he is allowed to settle however, and carve out the comfortable fractions that he enjoys rather than being asked to run at velocities that he isn’t comfortable with, he is a very hard horse to beat.