Scalping vs Mean Reversion
Scalping is a holding period of seconds to minutes and supplies no reason to enter. Mean reversion is a reason: the claim that a move away from a typical value tends to come back, which is most reliably true over exactly the short horizons a scalper trades.
One of these is a holding period and the other is an edge, so they are not rivals. They are worth writing about together because unlike most duration-and-edge pairings, the logic of the two suggests this particular combination is well matched.
What each one is
Scalping is a holding period of seconds to minutes, taken repeatedly, with a small target. Scalping covers it.
Mean reversion is an edge: price stretched away from a typical value tends to return. Mean reversion covers it, and momentum trading covers the opposing claim.
One says when to leave and the other says why to arrive. Whereas the holding period is a constraint you impose, the edge is a claim about the market that either holds at that horizon or does not.
Where they differ
Whether it selects anything. A holding period does not. Mean reversion produces a list — whatever has stretched furthest — which is a complete instruction and the part a scalper actually needs.
Which timeframe each favors. The holding period is a personal constraint. Mean reversion is a claim whose truth varies by horizon, and it is most defensible at short ones, where price genuinely does change direction quickly and often. Even the S&P 500, which has risen over decades, closed higher on only 53.1% of days since 1950 — from one day to the next, direction is close to a coin toss.
What the tail looks like. Most runs last a bar or two, but every so often price moves the same way for many bars in a row. That gap is the whole risk — a strategy that works on the average is one that has to survive the longest run, and a long streak of consecutive bars against a scalper is a very long time.
Where the losses come from. A scalp loses to costs and to the occasional long run. Mean reversion loses to the position getting more attractive as it goes wrong, which is a psychological failure the holding period does nothing to prevent.
Where they agree
Both want a range. The edge needs a mean to revert to and the holding period needs frequent opportunities, and a ranging market supplies both.
Both are ruined by a sustained trend, which removes the mean and turns every entry into a fight with the direction.
Both pay a round trip per trade — and because the target is small, the cost is large relative to it. Say a scalp aims for 0.25 and the round trip costs 0.01: that is 4% of the target, paid whether the reversion arrives or not.
And both need external position sizing. Neither the schedule nor the edge says how much to risk.
Which one to use: Scalping or Mean Reversion?
Use them together when the market is ranging and your costs are low. This is the configuration the reasoning actually supports, and it requires both conditions rather than either.
Use a hard stop that the edge did not choose. Mean reversion’s own logic argues against every exit, so the stop has to be imposed from outside it — and at a scalping frequency, one unstopped trade can undo a week of correct ones.
Use the scalping holding period only when you can be present continuously. That constraint is about your life rather than the edge, and it does not become less true because the edge is sound.
And when a trend establishes itself, stop trading the pairing rather than adjusting it. Widening the stop to survive a trend converts a scalp into a position you did not size for.
Why the longest run matters more than the average
Because you experience the tail, not the mean. A short average run makes the strategy look comfortable, and the rare long run is what actually arrives — and at a scalper’s size and frequency, that run is where the account damage happens rather than in the ordinary trades.
And because costs bind before the edge does. Roughly 4% of a half-bar target goes to friction on every round trip, so a genuine edge can be entirely consumed by execution without ever being wrong.
The original data
Scalping appears in 709 titles at a median of 23,380 views across 353 channels. Mean reversion appears in 157, at a median of 3,857 across 106.
Five times the videos and six times the audience on the holding period. The duration draws the attention and the edge — which is the half that determines whether any of it works — draws almost none, which is the consistent shape of this corpus.
On the chart above the edge says add and the run length says wait. Four bars in one direction is already a long run for a short-horizon market, and nothing says it has to stop there, which is the exact interval in which this pairing destroys accounts.
When it fails
The characteristic failure is averaging down inside a long run. The edge tells you the trade improved as price moved against you, the short holding period makes each addition feel low-risk because it will be over soon, and the two together produce a growing position in the middle of the longest run the market is going to hand you — a streak many times longer than the typical one. Every individual decision is consistent with the method, the position size is the only thing that was not, and by the time the reversion arrives the account may not be there to benefit from it.
A second failure is running the pairing in a trend. The mean stops existing and every entry fights the direction.
A third is scalping at retail costs, where roughly 4% of each target is consumed by friction regardless of edge.
A fourth is defining the mean after entering, which makes the reference value a function of your position.
And a fifth is scalping part-time, which selects your trades by when you happened to be available rather than by the setup.
Related
Scalping covers the short holding period. Mean reversion covers the contrarian edge. And momentum trading covers the opposite claim about the same moves.
This is the one place where the standard advice and the evidence line up. Short-horizon price genuinely does oscillate, changing direction far more often than it trends, so the edge is plausible. What kills people is not the edge failing, it is paying a round trip on every one of a very large number of trades.
— Michael Whitman
This page is educational, not financial advice. Test every idea on your own charts before risking money.