Does caffeine make you faster?

Here's what's in this week's edition.

​

A new trial out this week asked exactly that.

Researchers in Beijing gave 21 trained male 100m sprinters caffeine or a placebo an hour before testing, then timed them over 60m. The caffeine came at two doses, 3 or 6 mg per kg of body weight.

On the lower dose, 20 of the 21 ran faster. On the higher dose, all 21 did. The gains were 0.123 s and 0.175 s, about 1.7% and 2.4%. Normal day-to-day variation in 60m times is around 1.4%, so this is more than noise.

What interests me most is where the time came from.

The biggest gain was over the first 10m. That matches the only trial I could find run over a full 100m, in 13 male college sprinters, where caffeine was worth 0.14 s once times were corrected for conditions. Most of it came in the first 60m. Top speed and the last 40m weren't significantly different.

So caffeine looks mainly like a start and acceleration aid. Whether it adds anything at top speed is still open.

More isn't clearly better. The lower dose did most of the work, and the higher one added only a little. In an older dose study, on endurance runners, 9 mg/kg made no significant difference. For a 70kg sprinter, 3 mg/kg is about 210 mg, roughly two small cups of brewed coffee. And it's legal. Caffeine came off the prohibited list in 2004 and is now only monitored, though the NCAA still bans high levels.

I think it's worth trying for races, and it's cheap. I'd start around 3 mg/kg, about an hour before, and try it in training first, never for the first time at a meet that matters.

There are two cautions. The first is evening races. In the general population, caffeine cuts total sleep by about 45 minutes on average. Over a season, sleep matters more than a few hundredths in one race.

The second is under-18s, where I'd follow the youth guidance rather than the adult doses. The Australian Institute of Sport, for example, says under-18s should stay below 2.5 mg/kg across the whole day, which is less than one adult race dose.

And none of this is settled. It all rests on three small trials and about 47 men, none of them elite. I couldn't find a trial on women sprinters, the 200m, or a day of heats and finals. It's promising, but I wouldn't oversell it.

Asafa Powell, now coaching, thinks today's sprinters have a technology problem. He told Athletics Weekly: "They're using too much technology to try and chase fast times."

I agree with half of that.

The chasing part I agree with. Powell's real complaint is consistency: some sprinters run 9.8 one day and 10.0 the next, and he thinks they'd be steadier if they stopped chasing. But he never says which technology he means.

And the old Jamaican system wasn't anti-technology. Glen Mills broke Bolt's sessions down on screen in slow motion to show him exactly what he was doing.

So I don't think the stopwatch is the enemy. It's what we do with it.

Dan Pfaff rarely told his sprinters their times until the session was over. If they knew after every rep, the next one would become a race to beat it, and the mechanics, or whatever the technical focus was that day, would go out the window. Afterwards he linked the times back to the technique.

I liked this so much when I first came across it that I started doing the same. I used to give athletes their time straight after each run, but then the number gets in their head and changes how they run the next one. Now they get their times at the end.

And the research fits with that. Telling rugby league players their 20m time made them faster than giving them technical cues (3.54 vs 3.64 s). But in weight-room studies, take the feedback away and performance drops straight back.

Times drive effort in the moment. That isn't the same as learning to sprint.

I use a timer, but mostly to tell me when to stop. When a timed fly drops about 5% off the day's best, the session's done.

I also set easy running as a percentage of best time, not effort, because effort is hard to judge below flat out. When 18 college sprinters were asked to run 60m at 60% effort, they hit about 90% of their top speed. They were consistent. They just weren't running at the speed they'd been asked for.

Feel can be trained, though. With timing-gate feedback, footballers learned to hit 90% within two sessions. So use the clock to calibrate feel, not to replace it.

One more thing. Powell says that if he could work on Noah Lyles's first 30m, "there would be some madness on the track." Maybe. But Lyles won the Olympic title in Paris from last place at 40m, and Stuart McMillan has argued that spending more energy early would take from his real gift, the stretch from 60 to 90. PJ Vazel, a coach who analyses sprint races, looked at every Olympic and World Championship men's 100m final from 1972 to 2022. In 23 of the 24, the winner was the man who hit the highest top speed.

I'm not convinced the fix is that simple.

Kon Knueppel of the Charlotte Hornets came back from a hamstring strain last week and left the court about six minutes into the first quarter in Brooklyn. It was the same hamstring he hurt over the summer. He'll miss at least four more weeks.

Six minutes is unlucky, but it isn't unusual.

In one group of athletes, the median time from return to re-injury was 24 days. In elite European football, about 18% of hamstring injuries are recurrences, and more than two-thirds of those come within two months of return. And at one AFL club, the risk of another injury was highest in a player's first week back.

So when is a player ready?

I'd bring a player back on landmarks rather than dates. But we should be honest about them: nobody has validated one. In a survey of 131 top football teams, 95% used a criteria-based return, but they gave no consistent numbers. One well-known example, from about 25% of top speed to 50%, then 80%, then about 5% at a time to full sprinting, comes from its authors' clinical experience and earlier rehab programs, not from a trial of the numbers themselves.

Getting the injured leg as strong as the other one isn't a reliable test either. Two-thirds of 52 professional footballers were cleared with a deficit of more than 10% in at least one hamstring strength test, and the ones who got re-injured were no weaker.

What I'd want is speed built back one small step at a time. Before full sprinting, I'd look for a week of running near full speed with no pain or hesitation, and fly times back near the player's recent best.

Fly times tell you top speed is back. They don't tell you everything. One footballer returned with the same top speed as before his injury but about 20% less horizontal force at low speed. The deficit was hiding in the push-off.

Which is why the first weeks back are part of the return, not the end of it.

It's also why I'd keep players sprinting fast all season. Regular sprinting keeps the gap between training speed and match speed small, so a return doesn't have to start from nothing.

A new study tested 182 athletes at the 2025 World Masters Indoor Championships. Reaction time slowed by about 3.3 milliseconds for every year of age. Sprinters were quicker than endurance athletes overall, but they slowed at the same rate as everyone else.

Before that worries anyone, look at the test. Athletes hit a sensor with a hand or foot when a light came on. Average times were 0.75 to 0.95 s. That's nothing like reacting to a gun.

Reaction to a sound slows far less, about half a millisecond a year in a large general-population study. That's roughly 20 ms between 35 and 75.

Now compare that with how much slower the race gets. At the 2022 World Masters Championships, the M35 100m champion ran 10.49 and the M75 champion 13.77, a gap of more than three seconds. Even at the new study's steeper rate, reaction accounts for about 4% of it.

The rest is force. In masters sprinters aged 39 to 96, force, velocity and power over the first 30m fell about 1% a year, and they applied force less effectively, especially early in the sprint. Over ten years, a group of Finnish masters sprinters got 14% slower over 60m and lost 21% of their knee strength, while their muscle fibres held up.

So reaction isn't what masters sprinters should worry about. I'd chase the lost speed with strength and power, not reaction drills. And for a better start, practise the first few steps out of the blocks rather than reacting to a sound.

It's also worth knowing that in masters, your first false start is only a warning. It's your second that disqualifies you.

I haven't coached masters athletes, so this is the research and what I'd do with it.

If there's something you'd like me to cover, just reply to this email.

​