There's a number for what your feelings cost you. It's not a metaphor. Dalbar has been measuring it for three decades, and the gap between what the market returns and what the average investor actually keeps sits somewhere between 1.5% and 3% a year, depending on whose study you read and how rough the decade was.

Two percent doesn't sound like much. That's the trap. Nobody blows up their account over 2%. You bleed it, one nervous decision at a time, and you never see the bill.

Here's where the number comes from. The market goes up, but you don't earn what the market earns. You earn what your behavior lets you keep. You buy after a big run because it finally feels safe. You sell in the middle of a drawdown because holding feels insane. Dalbar's 2024 study found the average equity investor trailed the S&P 500 by 848 basis points in a single year — over 8% — mostly from re-entering late and missing the rebound. That's an extreme year, but the pattern repeats in smaller doses every year you trade on feel.

The reason it's invisible is that each individual decision looks reasonable at the time. Cutting risk after a scary week isn't crazy. Waiting for confirmation before you buy back in isn't crazy. But string those "reasonable" moves together across a year and you've quietly handed back a chunk of your return. You didn't make one catastrophic mistake. You made forty small ones that all felt smart.

Now let the number compound, because that's where it actually hurts. Say the market does 10% a year and your emotions cost you 2%, so you earn 8%. Over one year, who cares. Over ten years on $50,000, the 10% path grows to about $129,000. The 8% path gets to about $108,000. That 2% "tax" just cost you $21,000 — more than double the money you started thinking about it. The gap isn't linear. It widens every single year, because the returns you gave up would have been earning returns of their own.

Here's the part most people get wrong. They think the fix is getting better at reading the market — a sharper entry, a better sense of when to step aside. So they add another indicator, another rule of thumb for when it "feels toppy." That doesn't close the gap. It widens it, because now you have more reasons to act, and every action is another place for emotion to sneak in. The behavior gap isn't an analysis problem. It's a behavior problem, which is why it's named what it's named.

Some researchers argue Dalbar overstates it — that the real drag is closer to 1% to 1.5% once you measure it cleanly. Fine. Take the low end. A 1.5% tax on that same $50,000 over ten years still costs you around $16,000. The exact figure is arguable. The direction never is. Emotion is a cost, it's always negative, and it compounds against you.

So what do you actually do about it? You can't feel your way out of a problem caused by feeling. The only thing that reliably closes the gap is removing the moment of decision — deciding once, in advance, when you're calm, and then not touching it. The stop you set on Sunday is smarter than the one you'd move on Wednesday. The entry rule you wrote down beats the "this time is different" voice in your head every time, not because the rule is brilliant, but because it doesn't panic. It just executes.

That's the whole case for trading systematically, stripped of the marketing. It's not that a system predicts better than you. Most of the time it doesn't. It's that a system doesn't sell the bottom, doesn't chase the top, and doesn't renegotiate its own rules at 2 a.m. It collects the return the market offers instead of donating 2% of it back every year out of nerves.

Add up that tax over a trading lifetime and it's not a rounding error. It's the difference between two completely different account balances. If you've never put a number on what your emotions cost you, start there — even a rough one. Once you can see the bill, taking the human out of the loop stops looking like a loss of control and starts looking like the raise it actually is.

We publish all our backtest data — including six years of results on Bybit — so you can see what removing the decision looks like.

See the data →