The Paradox of Skill

Here is a puzzle that sounds backwards but is true. When everybody becomes more skilful, luck becomes more important, not less.

Think about a school running race. In the first year, some children have trained hard and some have never run at all. The trained runners win easily, every time - a slow child has no chance. Here skill decides everything, and luck barely matters.

Now imagine years pass and every child trains hard with the same excellent coach. Now they are all fast, and all nearly equal. When they line up, the times are almost the same. So what decides who wins? A tiny gust of wind, a slightly better lane, a perfect start by a hair - in other words, luck. When everyone is brilliant, brilliance no longer separates them, so the small random things decide.

Michael Mauboussin gave this a name: the paradox of skill. A "paradox" is something that seems to contradict itself but is actually true. Here it is in one line: as skill rises and spreads to everyone, the gap in skill shrinks, so luck decides more of who wins. This study is about seeing why more skill can mean more luck - and why beating a crowd of experts is so much harder than beating a crowd of beginners.

Picture all the players in a contest as a spread - a little hill showing how good everyone is. The middle of the hill is the average player; the far right edge is the very best. What matters is not how high the average is, but how wide the hill is - how far the best sit above everyone else.

On the left, the hill is wide. The best player stands far to the right of the average. That big gap means skill easily decides the winner - a small nudge of luck cannot close such a distance. On the right, everyone has improved and bunched together, so the hill is tall and narrow. Now the best player is only a whisker ahead of the average. That tiny gap can be closed by the smallest gust of luck - so luck decides.

Notice the strange thing: the players on the right are better than the players on the left. Everyone trained, everyone improved. And yet, because they all improved, skill stopped separating them, and luck took the wheel. Getting better did not make winning more about skill. It made winning more about luck.

The read, then, is to look not at how good people are, but at how bunched together they are. When a field is full of beginners, back the skilful - skill will out. When a field is full of experts who are all nearly equal, expect luck to scramble the results, and do not read too much into who happens to win any single time. The tighter the pack, the louder luck speaks.

illustrative Aayra's spelling club holds a yearly contest. Let us give each child a "skill score" out of 100 and add a small random wobble of luck - say up to 5 points either way on the day, for a lucky or unlucky word.

Year one. The children are all different. Haridya scores 90, the next best 80, then 70, then 60. On the day, luck adds or subtracts up to 5. Haridya might drop to 86 or rise to 95; the second child might climb to 85. But even with the best luck, the second child at 85 cannot pass Haridya at her worst of 85-and-up - the gap of 10 is bigger than luck's reach of 5. So Haridya wins almost every time. Skill decides.

Year two. Everyone has practised hard with a great coach. Now the top four all score around 90 - 91, 90, 90, 89. The skill gap between them has shrunk to almost nothing. But luck still wobbles by up to 5 points. Now the child who truly has 89 skill can jump to 94 on a lucky day and beat the 91-skill child who slips to 88. Suddenly the winner changes from contest to contest, and it is no longer the most skilful child who wins - it is whoever catches the luck that day.

The number lesson: nothing about luck changed - it was a 5-point wobble both years. What changed was the gap in skill. In year one the gap (10) was bigger than luck (5), so skill won. In year two the gap (1 or 2) was smaller than luck (5), so luck won. When skill gaps shrink below the size of luck's wobble, luck takes charge - even though every single child is now more skilful than before.

More luck at the top does not mean skill is gone. This is the easiest mistake. Just because luck decides who wins among a bunched-up crowd of experts does not mean skill stopped mattering. To even reach that crowd of experts, you needed enormous skill. Skill got you to the start line; luck decides the final placings among those who are already there. If you conclude "skill is useless, it's all luck," you will stop trying to get good - and then you will never even reach the pack where the prizes are.

A single winner is not proof of the most skill. In a tight field, the one who wins today may just be the one who caught the luck today. So do not crown the winner of one expert contest as "the best." Over many contests, the truly most skilful will win a little more often than the rest - but only a little, and only over a long run. Reading one result as proof of top skill is exactly the trap this idea warns against.

Not every field is bunched. The paradox bites hardest where nearly everyone is highly trained and equal. In fields where people are still very different in skill, the old rule holds and skill decides. So before you say "it's all luck now," check whether the field really is a tight pack of near-equals, or whether real skill gaps still remain. Assuming everything is bunched when it isn't is its own error.

This idea needs no special study - you can watch it in any activity that many people take seriously. Cricket is a good example: as coaching, fitness, and practice spread, the top players bunch closer together, matches get tighter, and small moments of luck - a dropped catch, a dodgy bounce - decide more big games than they used to. The players are better than ever, and yet luck seems to matter more, exactly as the paradox predicts.

In our markets, this reading is quietly powerful. A very large number of clever, well-trained, hard-working people study the same shares with the same tools and the same information. That is a bunched-up field of near-equals. In such a field, the paradox warns that beating the crowd by skill alone is extremely hard, and that whoever tops the list in any single year may be riding luck more than genius. This is a reason to be humble, not to chase last year's winner as if their top spot proves unbeatable skill. It also explains a puzzle: as everyone gets smarter, standing out gets harder, not easier, because you are no longer racing beginners - you are racing a whole pack that has trained just as hard as you.

• Look at the spread, not the average. Ask how bunched together the players are, because that gap - not how good they are - decides how much luck matters.
• Expect luck to rule tight fields. When nearly everyone is highly skilled and near-equal, small chances decide the winner.
• Don't over-read a single winner in an expert pack. In a bunched field, today's winner may just have caught the luck.
• Remember skill still got them there. Luck decides among experts, but only skill gets you into the expert pack in the first place.
• Check the field is really bunched. Where real skill gaps remain, skill still decides - don't cry "all luck" too soon.
• Stay humble as everyone improves. The better the whole crowd gets, the harder it is to stand out, and the more luck scrambles the results.