Advertisement IC Markets

Slippage tails on GBPJPY: histogram from my deals

Discuss 1-minute to 15-minute price action setups, fading intraday momentum, key support/resistance zones, and proven short-term trading methodologies.
Post Reply
LondonScalper
Posts: 965
Joined: Sat Sep 05, 2026 7:54 am

Slippage tails on GBPJPY: histogram from my deals

Post by LondonScalper »

Slippage tails on GBPJPY: histogram from my deals

GBPJPY can look like free range until you histogram slippage. I pulled a few hundred discretionary scalps and plotted signed slippage in points. The centre of the distribution was boring; the right-hand tail was where quiet weeks died.

Desk response after seeing the shape:
  • Harder spread filter before entry on GBPJPY than on EURUSD.
  • Smaller size when Asia is already wild or when London opens into a messy overnight range.
  • Time-stop enforced — tails love traders who “give it room” without a clock.
Averages lie; tails bill you. If your cross feels “fine on average,” check whether a few disasters are funding the marketing story in your head.

If you have charted slippage on a volatile cross, did the histogram change your size, your session hours, or whether you trade that name at all?

I revisit the histogram quarterly. If the tail thickens, I do not “try harder” on GBPJPY — I shrink size or shorten the hours I allow that cross. Tools change; humility stays.
Recommended broker for automated trading & scalping IC Markets
PropScalpDesk
Posts: 521
Joined: Sat Sep 19, 2026 7:50 pm

Re: Slippage tails on GBPJPY: histogram from my deals

Post by PropScalpDesk »

Worth splitting that histogram by order type if you haven't already. On my own crosses, entries by market order had a fairly tight distribution. Almost all of the ugly tail came from stop-loss exits, which makes sense, since a stop fills exactly when liquidity is disappearing. If the tail lives in the stops, the fix is mostly smaller size or avoiding certain hours. A tighter entry filter won't touch it.

The shape around zero also tells you something about the broker. A fair venue should show some positive slippage as well as negative. If the left side of your distribution is almost empty while the right side has a proper tail, positive slippage probably isn't being passed on, and that's a question for the broker rather than for your strategy.

I'd also plot it in R instead of points. A 12-point slip is nothing on a wide swing stop and a big bite out of a 60-point scalp.
Shadow Trader
Posts: 5594
Joined: Wed Sep 09, 2026 5:19 pm

Re: Slippage tails on GBPJPY: histogram from my deals

Post by Shadow Trader »

PropScalpDesk's split by order type changed things for me too, and the next cut that helped was time of day. Once I looked only at stop exits, most of the GBPJPY tail came from a narrow window around the London open.
img_12556_r1.png
img_12556_r1.png (47.89 KiB) Viewed 121 times
The chart is a made up version of what mine looked like, with stop exits from 07:00 to 07:30 London on the left and 09:00 to 11:00 on the right. In the open window 10 of 60 exits slipped more than 2 pips, about one in six, while later in the morning it was 1 of 140.

On an 8 pip stop, a 2 pip slip is a quarter of R gone before commission. That is why the tail felt so much bigger than the averages suggested.

The fix was not a tighter entry filter. I moved the start of my GBPJPY window to 07:30 and kept EURUSD from 07:00, which removed most of the right side without changing the strategy at all.

LondonScalper, did your tail move when you changed hours, or was it spread across the whole session? I would expect the messy overnight range days you mention to show it most.
Shadow Trader
Posts: 5594
Joined: Wed Sep 09, 2026 5:19 pm

Re: Slippage tails on GBPJPY: histogram from my deals

Post by Shadow Trader »

PropScalpDesk's point about the left side of the distribution was the next thing I checked, and the answer depended on order type again. On market entries my GBPJPY fills looked fair, but on stop exits the positive side was almost empty.
img_12556_r2.png
img_12556_r2.png (70.05 KiB) Viewed 117 times
The chart splits a made up version of my log the same way. Of 320 market entries, 58 filled better than requested and 74 worse, while of 200 stop exits only 3 filled better against 61 worse.

I would not call the stop side proof of an unfair venue, though. A stop becomes a market order at the moment price is running through it, so there is rarely a better price waiting, and almost any honest feed will show that lopsided shape.

The fairer test is entries and limit orders, where price is not moving against you at the trigger. If those also show a bare left side, that is when I would start asking the broker questions.

LondonScalper, did your histogram mix entries and exits together? Pooled, my 61 bad stop fills made the whole account look worse than the entry side deserved.
Shadow Trader
Posts: 5594
Joined: Wed Sep 09, 2026 5:19 pm

Re: Slippage tails on GBPJPY: histogram from my deals

Post by Shadow Trader »

Average R per trade is a single number, and a single number hides shape. Plotting every closed trade as a multiple of its initial risk, in bins of half an R, shows where results actually cluster and how long the tails run.

A healthy scalping profile often shows a tall spike near minus one R, a scatter of small partial results around zero, and a thinner right side. The questions worth asking are specific. Does the spike sit exactly at minus one, or does a second bump appear near minus one and a half or beyond, which points to slippage or stops that were widened mid-trade? Is the right tail made of planned exits or of a handful of lucky runs?

Skew matters as much as the mean. Two histories can share the same average while one depends on rare large winners and the other on frequent small ones. The first tolerates long dry spells poorly, and the second is more sensitive to costs.

Medians and the share of total R contributed by the best five trades add context. If removing those five trades turns the average negative, the edge is narrower than the headline suggests. Histograms also need a reasonable count of trades before the outline means anything.
Post Reply