Ask how fast

Bolt did the training and ran 19.75, then said, "You've got to keep your promise."

He ran 10.03.

Nearly twenty years on, Kenny Bednarek describes much the same arrangement with Dennis Mitchell. Ask to run the 400 and you get a "hell no", then a list of what you'd have to do first.

I think that's close to the right answer: the coach should have a view and say it out loud, and the athlete should have a real say.

We could just decide for them, but we're not as good at it as we think.

Stephen Francis read sprinters as well as anyone, and in 2015 he called Shericka Jackson "not a 100m type sprinter". Six years later she ran 10.82.

Junior results aren't much help either. Of the sprinters ranked in the world's under-18 top 50, only 17% of the men and 21% of the women made the senior top 50. That's rankings rather than event choice, but if a 16-year-old's ranking says that little about where they'll end up, I wouldn't settle their event on it.

And look at how some of the big switches really happened. Wayde van Niekerk moved up because of his hamstrings. Fred Kerley says money was a big part of it, but he'd never have ended up in the 100 if his ankle hadn't been hurting.

Hamstrings and an ankle. Try planning for that.

So when an athlete wants something you think is wrong, I'd do one of two things.

Say what it would take. A time, a block of training, whatever you'd need to see. They get something to chase, and nobody has to back down.

Let them find out somewhere it won't cost them. The first couple of meets of the season are ideal. If you were right, they've seen it for themselves. If you were wrong, you've learned something.

There isn't much evidence for any of this. I can't find a study comparing athletes who chose their event with athletes whose coach did. Research on giving athletes a say links it more to motivation than performance, and none of it is about events.

So this is mostly judgement, yours and theirs.

Game Speed

Do players lose speed in season?

Do team-sport players lose speed in season? It's less clear than you've probably been told.

In one US college soccer team, the starters got 4.3% slower during an 11-week season. The non-starters didn't. But a 2022 review of professional soccer found mostly trivial changes between in-season tests, with a lot of variation between teams. The samples are small, and a player who's injured can't do the last test.

I wouldn't tell you speed falls off a cliff, but I also wouldn't assume that playing keeps a player fast.

One study tracked 35 elite soccer players over four seasons. Attackers didn't reach 90% of their top speed at all in 35% of their matches. For midfielders it was 65%.

Training doesn't necessarily fill the gap. A later study from the same lead author looked at three-day turnarounds between matches: 97% of the time, a player got no run above 95% of top speed.

If that's the top of the sport, I wonder what a small-sided game on a Tuesday night is giving your players.

So I'd add a small amount of fresh, flat-out sprinting each week, early in the session before they're tired.

In one study of 36 young soccer players, one sprint session a week over 10 weeks brought the same sprint gains as two, on the same total volume. Handy if you only see your team twice a week.

Build it up gradually, though. A review of 22 soccer studies found that sudden jumps in fast running, relative to what the player had been doing, lined up with injury more often than weekly volume did. Most of those studies carry a high risk of bias, but it's good sense anyway.

And this is where it gets oversold.

As far as I can find, nobody has shown how little is enough, or that this protects hamstrings. The three-day-turnaround study covered more than 600 players and 96 match hamstring injuries. Across all turnarounds, rates were no different with or without near-top-speed running in training. They did see no injuries when that running came two days before a match, but on small numbers, and they warn it could be a false positive.

I'd still do it. If a player is going to hit top speed at some point in a game, I'd rather it wasn't the first time in months.

Just don't sell it to your head coach as hamstring insurance.

Reader Question

How much rest do you need between sprints?

Here's where I'd start:

• Acceleration: 1 minute per 10m, so a 30m gets 3 minutes.
• Top speed work: 1 to 2 minutes per second of effort.
• Speed endurance: 10 minutes after an 80m, 12 after a 100m, 15 to 20 after a 150m.

Those are starting points. The real test is whether the next rep is as fast.

If you have a timer, it'll tell you. When a timed fly is about 5% down on the day's best, that part of the session is over. If your best fly is 1.00, stop at about 1.05.

Why not cut the rest and get more done? Balsom and colleagues had seven men run 15 x 40m with 30, 60 or 120 seconds between sprints. Speed over the last 10m dropped every time: after 3 sprints on 30 seconds, after 7 on 60, after 11 on 120. The first 15m only slowed on 30 seconds.

So the start of the rep can look fine after top speed has already gone. If you only time the first bit, you'll miss it.

If time's short, I'd cut a rep rather than the rest.

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

​