I used to write this exact sentence myself: retail options traders are gamblers, meme-stock chasers, dumb money that dealers happily take the other side of. I wrote it because it was true in 2021. I kept writing it for two years after it stopped being true, mostly because nobody in market structure likes admitting the joke’s on them.
Here’s the thing that finally changed my mind. Citadel Securities’ most recent trading report shows retail options volume running at a record clip this year — north of $6 billion a day in premium, up more than 60% from where it sat just a year ago. That number alone isn’t the story. Big numbers get thrown around in this business constantly and most of them mean nothing. What matters is what retail is actually doing with the money, and that part doesn’t look anything like 2021.
Speed is the first tell, and it cuts against retail at first glance. Something close to half of all retail options volume now trades same-day — zero days to expiry, the contracts that expire before dinner. Back in 2021 that figure was barely above one in ten. On paper that’s the reckless story writing itself: shorter holding periods, faster decay, more room to blow up an account. I thought that too, for about as long as it took to look at what’s actually inside those short-dated trades.
It’s not meme stocks anymore. Go back to peak mania — late 2021 — and retail’s ten most-traded names were dominated by a single stock everyone remembers, with that one name outtrading the other nine combined. The concentration in the top ten hasn’t really changed. What’s inside it has. Index products and broad ETFs now make up the majority of that top-ten premium, and index-linked options overall have climbed to somewhere around 60% of all retail volume — up from a rough historical baseline near half. That’s not a crowd betting on the next squeeze. That’s a crowd hedging a portfolio.
You saw it clearest during the growth-stock selloff earlier this summer. Demand for broad market puts from retail accounts spiked to something like three times the normal run rate, and — this is the part that actually surprised me — retail didn’t just buy the protection and sit on a loss when the bounce came. They sold into it well. By some measures it was one of the more profitable stretches for retail put-selling since the tariff scare last year. I’ll admit I expected the opposite. I expected retail to buy the top and dump the puts at the bottom, the way the story always goes. It didn’t go that way this time.
None of this means the speculative appetite disappeared, because it hasn’t, and I don’t want to oversell the “retail grew up” narrative either — plenty of that money is still chasing momentum, just in a different sector. Semiconductors are where the aggression lives now: well over a billion dollars a day in premium through the first half of the year, and closer to two billion in June, mostly on the put side rather than calls, which tells you retail isn’t just chasing chip stocks up anymore — they’re pricing the downside too. Given how much implied volatility in that group has climbed over the past decade, that’s arguably the correct read, whether or not the people making the trade could explain why in a research note.
What comes after chips is where it gets speculative in the old-fashioned sense. Space names, mining and metals tied to the physical side of the AI buildout, and — new this year — a real jump in drone and robotics options, a corner of the market almost nobody was pricing back in January. Some of that is just rotation chasing whatever’s hot. Some of it looks like retail trying to get ahead of the next leg of a trade rather than late to the last one. I’m not fully sold on which it is yet, and I’d be lying if I said the drone number specifically didn’t make me raise an eyebrow.
So here’s my actual complaint, and it’s aimed at people in my own business more than at retail: we kept the “dumb money” label because it was convenient, not because it was accurate. It let dealers explain away flow they didn’t fully understand, and it let commentators write the same lazy paragraph every earnings season. The data doesn’t support it anymore. Whether that’s because retail got smarter or because the dumb money simply got outcompeted and left, leaving behind a more sophisticated core, I genuinely don’t know.
What I do know is this: if you’re on a trading desk still modeling retail flow as noise, you’re mispricing your own risk. The question worth asking isn’t whether retail knows what it’s doing. It’s whether the people still calling them dumb money do.
