Usain Bolt's 9.58: The Physics of the Fastest 100 Meters Ever Run
Inside Usain Bolt's 9.58 world record in Berlin in 2009: 20-meter splits, a laser-measured top speed of about 27.5 mph (44 km/h), 41 steps, a slow reaction time, and how he finished.
In this story8 sections
Usain Bolt's 9.58 has stood since August 2009. German biomechanists filmed and laser-tracked every meter of it, so we know exactly where he won it. Hint: not at the start.
On August 16, 2009, in Berlin's Olympiastadion, Usain Bolt ran 100 meters in 9.58 seconds with a legal tailwind of +0.9 m/s (about 2 mph). He took 0.11 seconds off his own 9.69 from the Beijing Olympics exactly a year earlier. World Athletics called it "the largest chunk ever – by far – to be sliced from the 100m World record." More than 17 years later, it's still the world record.
It's also one of the best-documented races ever run. A team of 18 researchers from six German institutions, working with the German athletics federation (DLV) and the IAAF (now World Athletics), filmed the final from the stands and aimed laser speed guns at the sprinters' backs. Here's what the data show.
The race, 20 meters at a time
The official biomechanics report broke the final into 20-meter splits:
| Distance | Bolt's time | Time for that 20 m |
|---|---|---|
| 20 m | 2.88 s | 2.88 s (includes reaction) |
| 40 m | 4.64 s | 1.76 s |
| 60 m | 6.31 s | 1.67 s |
| 80 m | 7.92 s | 1.61 s |
| 100 m | 9.58 s | 1.66 s |
Bolt led from 20 m on, but only narrowly. At 20 m, Tyson Gay was just 0.04 s behind. The real gap opened in the middle of the race. By the finish, Gay ran 9.71, then an American record, and still finished well back.
Where the 27.8 mph figure comes from
You'll often see Bolt's "top speed" given as 27.8 mph (44.72 km/h). That number is arithmetic, not a direct reading. Bolt covered the 60–80 m segment in 1.61 seconds. Twenty meters divided by 1.61 seconds is 12.42 m/s, or 27.8 mph (44.72 km/h). It's an average over that stretch, based on video split timing to two decimal places.
Lasers tell a slightly different story:
- Graubner and Nixdorf (2011), writing up the Berlin laser data for New Studies in Athletics, put Bolt's measured maximum at 27.6 mph (12.34 m/s, 44.4 km/h), reached at about 67.9 m.
- A 2023 reanalysis of the same laser data in Bioengineering (Štuhec et al.) found a peak of 27.6 mph (12.32 m/s), first reached at about 52 m, with Bolt averaging above 26.8 mph (12 m/s) in five 10-meter sections of the race.
So the honest version: Bolt's peak was about 27.5 mph (12.3 m/s, 44 km/h) by laser, and the popular 27.8 mph (44.72 km/h) is the 20-meter average implied by the splits. Both are real; they just measure different things. That's the kind of distinction we spell out on our How we verify page.
He wasn't fast out of the blocks
Bolt's reaction time in Berlin was 0.146 seconds, sixth-quickest of the eight finalists. Richard Thompson reacted fastest at 0.119 s, so Bolt was already 0.027 s behind him before he'd moved.
Bolt, 6 ft 5 in (1.96 m) tall according to the 2023 study, still reached 20 m first. Everything after that was about the speed he could build and hold.
Fewer steps, longer steps
This is the headline from the biomechanics report. Bolt needed 40.92 steps to cover 100 m. Second-placed Gay needed 45.94. Asafa Powell, third in 9.84, needed 44.45.
Bolt's average step length grew as the race went on:
- 0–20 m: 5.8 ft (1.78 m) per step (including the crouched start)
- 20–40 m: 8.3 ft (2.52 m)
- 40–60 m: 8.8 ft (2.67 m)
- 60–80 m: 9.1 ft (2.77 m)
- 80–100 m: 9.4 ft (2.85 m)
His step frequency, about 4.5 steps per second through the middle of the race, wasn't higher than his rivals'. Gay, for example, hit about 4.9 to 5.0. Bolt simply covered far more ground with each contact. The 2023 study put his average step length at 8.0 ft (2.45 m) against 7.3 ft (2.23 m) for the rest of the finalists, with almost identical step frequencies.
Speed is step length times step frequency. Most tall athletes gain the first and lose the second. Bolt kept his frequency close to that of much shorter men, and that's the whole trick.
Did he really slow down less?
Every sprinter slows in the last 20 meters. Muscles can't keep producing peak power for 10 seconds. Bolt did too: his 80–100 m split was 1.66 s, up from 1.61 s.
It's often said Bolt "decelerated less" than his rivals. The Berlin splits don't really show that:
- Bolt: 1.61 s → 1.66 s (about 3% slower)
- Gay: 1.63 s → 1.69 s (about 3.7% slower)
- Powell: 1.68 s → 1.74 s (about 3.6% slower)
- Daniel Bailey: 1.70 s → 1.75 s (about 2.9% slower)
- Richard Thompson and Marc Burns: 1.72 s → 1.76 s (about 2.3% slower)
Most of the field slowed by 0.04–0.06 s over the last 20 m. Bolt's slowdown (0.05 s) differs from Gay's and Powell's (0.06 s) by just 0.01 s, and Bailey, Thompson and Burns slowed slightly less than he did. Graubner and Nixdorf caution that differences this small are within measurement error: in their words, "a difference of 0.02 sec in total for any result should not be worth discussing." So there's no clear evidence Bolt held his speed better than the others. What really separated him was that his peak was so much higher that even his slowing finish, 1.66 s for the last 20 m, was the fastest closing 20 m in the race. Interestingly, his step frequency dropped over the last 20 m (from 4.49 to 4.23 steps per second) while his steps got even longer. He wasn't spinning his legs faster to hold on. He was striding out.
Physics: where does all that energy go?
A 2013 paper in the European Journal of Physics (Hernández Gómez, Marquina and Gómez) modeled the Berlin run as a constant driving force fighting air drag. It's a simplified model, so treat these as estimates, not measurements:
- Bolt's peak power came just 0.89 seconds into the race, at about 3.5 hp (2,620 W), while he was moving at only 14 mph (6.24 m/s), about half his top speed.
- Of roughly 81.6 kJ of mechanical work, the model says only about 7.8% went into moving him forward. The rest, about 92%, was "absorbed by the drag."
For contrast, the 2013 Nature study of wild cheetahs calculated a stride-averaged power of about 25 W per kg of body mass (about 11 W per pound) for Bolt's 9.58. Hunting cheetahs often exceeded 100 W/kg (45 W per pound). More in our piece on why cheetahs can't keep their top speed.
Claimed, calculated, measured
- Official result (World Athletics): 9.58 s, wind +0.9 m/s, Berlin, August 16, 2009.
- Average speed over the race: 23.4 mph (100 m ÷ 9.58 s = 10.44 m/s, or 37.6 km/h).
- Fastest 20 m average (from splits): 27.8 mph (12.42 m/s, 44.72 km/h), 60–80 m.
- Laser-measured peak: about 27.5 mph (12.3 m/s, 44 km/h), reported as 12.34 m/s (2011) and 12.32 m/s (2023).
Compare Bolt with the rest of the fastest 100 m sprinters, or with his 200-meter world record.
Sources
- World Athletics: Bolt again! 9.58 World record in Berlin!worldathletics.org
