HFT. Sort of Review: The Quietest Kind of Bybit Copy Trading

Most cards on Bybit's copy-trading board are shaped like weather. This one is shaped like a ruler.

HFT. Sort of is the strategy people mean when they ask for low-drawdown copy trading: across every reading we have taken of it, the maximum drawdown field has shown the same small number, and the 90-day curve has never sat more than a fraction of a point below its own peak. The account's page on the exchange looks almost featureless next to the board's usual fireworks - a line that mostly refuses to dip.

We keep a rating of Bybit's leaders built on our own accumulated readings, and this account holds a place in its upper rows. This review is what those readings say up close - including the one piece of arithmetic that the calm surface is made of, which anyone copying it should understand before sizing anything.

What the card showed

On our reading of 5 September 2026: 325 days old, about $90,000 of the leader's own assets under management, 138 followers against a cap of 1,000, and a total follower profit of $17,922. The published 90-day return was 26.08% - modest by the board's standards - and the max drawdown 2.81%, a figure that has not moved by a hundredth of a point across our readings of 28 August, 2 September and 5 September. The Sharpe printed on the card, 2.76, is among the better ones we see; the profit share is 10%, the cheap end of what the board's upper rows charge.

That is the quiet part. Here is the machinery.

Where a 95% win rate comes from

The card reports a 94.57% win rate: 179 closed positions, 173 of them winners, 6 losers. It also reports a profit-to-loss ratio of 0.4 - meaning the average losing trade is roughly two and a half times the size of the average win.

Read those two numbers together, because separately they mislead. The strategy harvests many small wins and accepts rare, larger losses - the classic quiet shape we walked through when we took apart the leader card itself. On accounts like that, the win rate is not a skill metric; it is a description of where the risk is hidden. What makes this one better than most of its kind is the evidence that the hidden part has stayed genuinely small: six closed losses in eleven months, a drawdown that our own record - not just the card's rolling window - has never seen worse than 2.81%, and a curve that has not spent meaningful time below its own peak.

The honest limit of that evidence: a record with six losses has had six stress tests. It shows discipline; it cannot show what happens when the market takes the other side for weeks rather than trades. Quiet strategies fail rarely and abruptly, and no card announces the transition in advance.

Eight days of our own record

Between 28 August and 5 September we read this card three times, and the three readings tell a small, instructive story.

The audience only grew: 117 followers, then 128, then 138. The ledger breathed: $17,359, up to $19,072, back to $17,922 - the twenty-one people who arrived across those days joined an account whose collective profit ended the stretch slightly lower than it began. And the published return wobbled from 31.5% to 33.1% to 26.1% for no reason a follower would feel, because the trailing window was rolling. Of everything on the card, the number that did not move - the drawdown - was the one telling the truth about the week.

The twin

The board also lists TopROI. Sort of, and the platform's own data says the two are siblings rather than strangers: both accounts are exactly 325 days old, both charge the same 10% share, and their user ids sit a hair apart in the platform's sequence - registered together, run as a pair. The twin trades the louder dial: an 89.33% win rate with 17 closed losses, a 5.72% drawdown, 65 followers.

One operator offering two risk settings is not a red flag - if anything, the twin's existence makes the pair look like a deliberate product line rather than a lucky streak. The caveat is arithmetic, not suspicion: copying both is not diversification. It is the same decision-maker twice.

The accounts wearing the name

Search "hft sort of bybit" - or the name on the board itself - and more than one result comes back. There is an "HFT. Sort of2", under an unrelated user id, with five followers and a drawdown that swung from 38.6% to 13.8% across the same readings in which the original's never moved at all. A "TopROI. Sort of2" rides the twin's name with one follower. We found the same pattern beside bluntz, and the advice does not change: follow the exact account, not the name. The one this review covers is the one our rating row links to.

What copying it actually risks

Size against the shape, not against the history. The 2.81% drawdown is the record of eleven good months; the 0.4 profit-to-loss ratio is what a bad stretch is built from. A copier who allocates as if 3% is the worst case is trusting that the six-losses-in-179 pattern holds forever, and no quiet strategy has ever promised that. We worked through how much of your capital belongs in any one strategy separately; for this shape, that number matters more than for the loud ones, precisely because the card gives you so little warning to react to.

For the outcome-shopper, the summary is short. This is the board's honest version of calm: cheap, disciplined, transparent about its losses, watched by our own record for weeks without a wobble - and quiet in the specific way that must never be mistaken for safe.

Readings quoted are ours, taken 28 August, 2 September and 5 September 2026. The mechanisms - the win-rate shape, the rolling window - are how the board works on any date. Copy trading carries real risk, and nothing above predicts future performance.