forex-scalping.com

I've been experimenting with a tighter stop on some EURUSD entries. Roughly a quarter of my usual stop distance, entered at a better price on a limit order. The tricky part hasn't been the trade, it's been the size.

The maths is simple. If my normal stop is 8 pips and I risk 0.5% of the account, a 2-pip stop at the same 0.5% risk means four times the lot size. On paper the risk is identical.

In practice, it isn't, for two reasons.

Slippage and spread are a much larger share of a 2-pip stop. Losing 0.3 pips of fill against a stop eight pips wide costs me less than 4% of what I put at risk. On a 2-pip stop it's 15%. So my real risk per trade is higher than the 0.5% I think it is.

And the psychological problem. Seeing four times my normal size on the screen does something to my head. After a tight-stop loser, my instinct is to re-enter at the same large size, immediately. It feels like the same 0.5% risk, but it's revenge trading in disguise.

What I've settled on. Tight-stop trades risk 0.35% instead of 0.5%, to allow for the extra slippage. The lot size is pre-calculated and written on a card for each stop distance I use, so I'm not calculating in the moment. And after a tight-stop loss, the next trade must use my normal stop and normal size.

So far, 40 trades in, the tight stops are slightly profitable.

LondonScalper wrote: If my normal stop is 8 pips and I risk 0.5% of the account, a 2-pip stop at the same 0.5% risk means four times the lot size.

There's a third cost in that maths worth adding: commission scales with lots too. Four times the size means four times the commission, so it grows exactly like the slippage does as a share of risk. Say the raw account charges $7 per lot round trip, which is 0.7 pip on EURUSD. Against an 8-pip stop that's under 9% of the risk; against a 2-pip stop it's 35%. Add the 0.3 pip of slip and a stopped 2-pip trade really loses about 3 pips, so the actual risk is around 0.75% rather than 0.5%. One fix is to size off the stop plus expected costs, 3 pips here, which brings the size down to about 2.7 times normal instead of 4. That also takes some of the shock out of the big number on screen.

The commission point in the second post is the one that finally made me change size. I was treating slip as the only extra cost and ignoring the fact that commission scales with lots the same way.

On my raw account the round trip is about $6 per lot. Against an 8-pip stop that is under 8% of the risk. Against a 2-pip stop it is over 30%. Once I sized off stop plus expected costs of 3 pips, the on-screen lot fell from 4x to about 2.7x, which also took some of the revenge urge out of the next click.

The attached chart is an illustration of the same EURUSD M5 scalp with an 8-pip stop at 1.0842 versus a 2-pip stop at 1.0848, both from an entry at 1.0850, so the size difference is obvious on one screen. I also keep a hard rule that after any tight-stop loss, the next trade is normal stop and normal size, even if the tape still looks perfect. That matches what you settled on, and it is the bit that actually stopped the revenge upsizing.

Have you tracked how often the tight-stop winners still need a wider stop because the first fill was already half a pip off?

LondonScalper, the fill quality question on tight stops is what I tracked next. On my sheet, about one in four of the 2-pip winners still needed the stop widened by half a pip because the entry fill was already off, which quietly turns a planned 0.35 percent risk back toward 0.5 percent.

I now write the planned stop and the allowed widen on the same card. If the fill is worse than half a pip from the limit, the trade is scratched at market rather than nursed with a bigger stop at large size.

That rule hurt win rate a little and cut the revenge urge a lot. The big lot on screen is less tempting when the first action after a bad fill is flatten, not adjust.

Fairman's cost-inclusive sizing still sits underneath. The card just adds a fill gate so the maths does not get renegotiated after the click.

I keep the tight-stop book on its own journal tab so the stats do not blur into normal stops. Curious whether you have separated them or still mix both books.