Scarb Glossary
live
← All sections
Tested — does not work

Exchange news and token unlocks as a reason to close a short early

The idea: hold the short, but get out early if the coin has exchange news coming or tokens coming out of vesting. It did not hold up: those events do not explain the violent moves against a short, and in one case the news is a reason to hold, not to leave.

Why it sounded right

A long-held short has one real problem, and it is not the average result — it is the rare but huge move against the position. A move like that can eat the profit from dozens of normal trades, and a protective order is poor defence against it: put it close and it gets knocked out almost every time, put it far and it is no longer protection. So it is natural to look not for the size of the protection but for its cause: if the violent moves up have an external trigger, then they can be seen coming and you can leave before rather than after.

The triggers are right there and known in advance. Exchange news — a listing on another venue, a perpetual contract being added, inclusion in some programme — is almost always announced publicly, and some of it comes with a delay before trading starts. Tokens coming out of vesting are known weeks ahead: the schedule is published and the date is hard. So you get a tidy scheme: hold the short by the usual rules, but know in advance which days are best not to be in, and leave.

The scheme is also backed by what it feels like from the inside. When a short really does get blown out, you can almost always find a piece of news that explains it. In hindsight the link looks obvious, and it is very hard to believe it is an illusion: we go looking for news exactly where we already know there was a move, and we find it — because an active coin almost always has news.

How we tested it

The test is simple in principle and not obvious in practice. You have to compare not "was there news before the move" but the frequency: how often news shows up in windows where there really was a violent move against the short, versus how often it shows up in random windows of the same size on the same coins. If news explains the moves, the first frequency should be clearly higher. If the frequencies match, the news is just background — it is always there, and the sense of a connection comes from looking for it knowing the answer.

We did exactly the same with tokens coming out of vesting: compared how often an unlock date falls into a window with a violent move against the short with how often it falls into a random window. Separately we tested the move itself — "exit on the event": for each type of news we computed what exiting on the news day would have given versus simply holding the position by the usual rule to the end of its term.

How it ended

The frequencies matched. News shows up next to violent moves against a short about as often as it does in random windows on the same coins. Tokens coming out of vesting: the same, and with a slight lean toward the random windows. So there is no basis for the scheme: the events we planned to step around do not distinguish dangerous days from ordinary ones. The feeling of "there's a reason every time" turned out to be exactly what it looked like — finding a cause after the result.

One exception did show up, and it flips the move on its head. A listing on the Korean venue Upbit really does produce a violent pump on the day it is announced — but it happens in literally minutes, so closing "on the news" is something you cannot do in time anyway, you are already inside the move. And then, over the following week, the coin gives that pump back, and in almost every case the short would have been in profit if it had simply been held. So this piece of news is not a reason to get out, it is a reason not to flinch — and sometimes to add. Exiting on a timer after such a pump lost to holding in the overwhelming majority of cases.

Where this could be wrong. Some news is not published instantly, and not in one place, so our list is incomplete — rare triggers could have been missed. And we counted an event as "falling into the window" by date; there is no fine matching to the hour here.

What this teaches

More in this section

A long shelf, then a break upward
The idea: price sits in a well-worn range for months, and leaving it means a run to the next such range. When we tested it, breaking out of the shelf paid exactly as much as a random day on the same coin.
Big orders in the book as a reference point
The idea: a big order in the book is a level price will bounce off, or a level it will aim for. When we tested it, price reached such a wall no more often than it reached a randomly picked level at the same distance.
Buying into a vertical spike
The idea: the coin has gone vertical, so a move has started and you should get on board while it runs. When we tested it, buying the spike was not a zero but a loss: far fewer trades finished up than with a random entry.