There's a well-known piece of research that gets wheeled out every few years, usually during a tournament. Somebody looked at several hundred penalties, worked out where they went, and found that roughly a third of them go down the middle — but goalkeepers stay in the middle less than 6% of the time.
The obvious conclusion is that keepers are leaving saves on the table by diving. And the obvious follow-up question is: if everyone knows this, why hasn't anything changed?
The maths, first
The numbers vary by study but the shape is consistent. Penalties are distributed roughly evenly between left, right and centre, with a slight bias towards the taker's natural side. Goalkeepers commit early — they have to, because a well-struck penalty takes about 0.4 seconds to reach the goal and human reaction time plus movement takes longer than that.
So the keeper is essentially guessing, and if he guesses left or right he covers maybe a third of the possible outcomes. If he stays central he covers a smaller area but a disproportionately common one, and he can still stretch a leg or an arm towards the corners.
On paper, staying still looks like the better bet, or at least a bet that should be made a lot more than 6% of the time.
Why it doesn't happen
The explanation that's usually offered is action bias, and I think it's basically right. A keeper who dives the wrong way has done something. He tried. Nobody blames him much — the penalty was going in anyway, most people think.
A keeper who stands still and watches the ball go into the top corner looks, to everybody in the stadium, like he didn't try. Even though the outcome is identical, the perception is completely different, and keepers are extremely aware of this. The regret from a passive failure is much worse than from an active one.
I'd add a second reason that gets less attention: the takers would adapt. The reason a third of penalties go down the middle is precisely because keepers almost never stay there. If keepers started standing still at a meaningful rate, takers would stop going down the middle within about a season, and the advantage would evaporate.
This is a game-theory equilibrium, not a mistake. The current distribution of keeper behaviour and taker behaviour is roughly stable, and the "keepers are wrong" analysis treats one side as fixed when it isn't.
What's actually changed in the last decade
The real shift hasn't been in whether keepers dive. It's in when.
Older penalty-saving technique involved committing early and hoping — a full dive initiated before the ball was struck. What good keepers do now is delay the commitment as long as possible, using a small split-step and a wide, low stance that lets them push off in either direction from a balanced position.
This is why you see more saves where the keeper looks like he's fallen over sideways rather than making a graceful dive. The graceful dive is a guess. The awkward collapse is a reaction, and reactions save more penalties.
The other change is information. Every professional keeper now has a folder on every likely taker — preferred side, whether they change it under pressure, what their run-up looks like when they're going one way versus the other. Some of it is noise. Some of it, particularly the run-up tells, is genuinely predictive.
The run-up tells
This is the part I find most interesting, because it's where the keeper's job stops being a coin flip.
The common ones: the angle of approach tends to correlate with direction, since it's mechanically easier to hit across your body than to open up. The position of the non-kicking foot at plant is a strong signal — it points roughly where the ball is going more often than takers realise. And the last glance before the run-up, though the better takers have learned to fake this or simply not look at all.
A keeper who reads one of these correctly isn't guessing anymore. He's reacting to information a fraction of a second before the strike, which is exactly enough time to matter.
The counter is the stuttered or paused run-up, which exists specifically to make the keeper commit before the plant foot lands. It's an arms race and it's been going on for decades.
The mental side, which is most of it
Something that gets underplayed: the keeper has almost nothing to lose and the taker has everything. That asymmetry is the keeper's main weapon and the good ones know it.
Everything a keeper does before a penalty — the delay, the standing tall, the pointing to a corner, the conversation with a defender — is aimed at converting a technical task into an emotional one. A taker who is thinking about the keeper is not thinking about the strike, and penalties are missed by bad strikes far more often than they're saved by good goalkeeping.
Roughly a quarter of penalties that don't result in goals are misses rather than saves. That's an enormous number and it's the taker's error, not the keeper's skill. Which means the highest-value thing a keeper can do is make the taker uncomfortable, not guess better.
What I'd tell a young keeper
Don't worry about the statistics. Worry about the split-step timing and the plant foot.
Stay up as long as you physically can, get your weight balanced and low, and watch the standing foot rather than the ball or the eyes. If you can delay your commitment by even a tenth of a second you'll save more penalties than any amount of clever pre-match analysis will give you.
And occasionally — not often, maybe one in ten — just stand there. Not because the numbers say so, but because word gets around, and a taker who has seen you do it once will hesitate about going down the middle for the rest of his career. That's worth more than the save itself.