Advertisement IC Markets

Asymmetric Slippage: How to Detect It

Compare ECN/Raw spread brokers, analyze execution speeds, report slippage, and evaluate commission structures for high-frequency traders.
Post Reply
FXS Prop Desk
Posts: 1690
Joined: Wed Oct 07, 2026 9:52 am

Asymmetric Slippage: How to Detect It

Post by FXS Prop Desk »

Fair execution produces both positive slippage (better price) and negative slippage (worse price). If positive slippage almost never happens, the execution may be asymmetric.

How to check:
  1. Collect at least 100–200 market orders and stop fills.
  2. For each, calculate slippage: fill price minus requested price, adjusted for direction.
  3. Count positive, zero and negative outcomes.
  4. Compare the average size of positive vs negative slippage.
What to look for:
  • Some imbalance is normal in fast markets, especially on stops.
  • A complete absence of positive slippage on market orders is a red flag.
  • Slippage that only appears when it hurts the trader deserves a direct question to the broker.
If you find a clear asymmetry, contacting the broker with your data is the first step, and comparing with a second broker the next.

Have you analysed your slippage distribution, and was it balanced?
FXS Official Team | The Idea Lab & Strategy
Recommended broker for automated trading & scalping IC Markets
Fairman
Posts: 6955
Joined: Tue Jul 21, 2026 7:11 am
Location: Abuja

Re: Asymmetric Slippage: How to Detect It

Post by Fairman »

FXS Prop Desk wrote: Some imbalance is normal in fast markets, especially on stops.
That's the bit that makes it important to split the sample before counting. Stop orders trigger in the direction of a move, so by design they tend to fill worse than the trigger, and mixing them in can make perfectly fair execution look asymmetric. Limit orders go the other way, they can't fill worse than the limit, so they inflate the positive side. The clean test is market orders only, ideally separated by session and by whether price was moving toward or away from you at the click. If market orders in quiet hours still show almost no positive slippage across 150 or so fills, that's when the question to the broker has real weight behind it.
It’s Fairman :geek:
Shadow Trader
Posts: 5594
Joined: Wed Sep 09, 2026 5:19 pm

Re: Asymmetric Slippage: How to Detect It

Post by Shadow Trader »

FXS Prop Desk, splitting the sample the way Fairman describes is the only clean test I trust. Mixing stops with market orders made my first pass look terrible, because stops are designed to fill worse than the trigger in a moving market.

Once I kept market orders only and separated quiet hours from news windows, the picture changed. In quiet London my positive and negative slip counts were nearly even across about 160 fills, with average sizes close enough that I stopped worrying.
img_13617_r1.png
img_13617_r1.png (63.23 KiB) Viewed 107 times
The illustration compares a balanced quiet sample against a mixed sample that folds stops in, so the negative bar dominates for the wrong reason. That is why I keep the two order types in separate columns now.

If quiet market orders still show almost no positive slip across 150 fills, that is when I would raise it with the broker. Until then I treat news window asymmetry as normal cost.

My own split removed most of the apparent imbalance. Did yours change the reading as well, or did the skew survive on quiet market orders alone?
Shadow Trader
Posts: 5594
Joined: Wed Sep 09, 2026 5:19 pm

Re: Asymmetric Slippage: How to Detect It

Post by Shadow Trader »

Fairman's split into market orders only fixed the counting for me, but it leaves out one form of asymmetry that never shows in the slippage column. A fill can only be measured if it happens, and rejected or requoted orders do not appear in a fills export at all.

So I now log every order the platform refused, along with where price went during the delay. Over my last 200 market orders there were 7 rejections, and on 6 of the 7 price had moved in my favour between the click and the refusal.

That pattern is what last look tends to produce. Slippage on the fills can look perfectly balanced while the trades that would have gone my way quietly never happen.

Latency is the other thing worth separating from the broker. If both sides are symmetric but large, I check the ping from my VPS to the trade server first, because 80 milliseconds in a fast London minute produces slippage that is honest but expensive.

Seven is a small number, so I am still collecting before I raise it. FXS Prop Desk, do you count rejections and requotes in the same sheet, or treat them as a separate test?
Shadow Trader
Posts: 5594
Joined: Wed Sep 09, 2026 5:19 pm

Re: Asymmetric Slippage: How to Detect It

Post by Shadow Trader »

Slippage gets a bad name, but it's supposed to go both directions. Price moves in your favour between click and fill, and symmetric execution hands you the better price.

Some brokers do exactly that. Others run asymmetric policies: negative slippage is passed on, while positive slippage is either withheld or the order is re-quoted at the requested price. The execution policy document usually spells it out. Hardly anyone reads it.

Contrarian bit. A trader with 30 fills who sees only negative slippage might blame the broker, when 30 trades is a tiny sample and streaks happen. Meanwhile a trader with several hundred fills and zero positive slippage has found a real pattern, and that deserves a question to support.

A worked example, with invented numbers. Over 200 market orders, a trader logs 80 fills worse than requested, averaging 0.4 pips, and 6 fills better than requested, averaging 0.1 pips. Natural noise would normally put those two counts much closer together, so the lopsidedness is the interesting part and not the size of any single fill.

Absence of evidence isn't evidence of a rigged setup. Absence across hundreds of fills is at least worth a polite email.

Execution mode matters as well. Market execution usually fills at whatever the liquidity provider gives, so slippage can arrive in both directions. Instant execution tends to quote a price and then fill it or re-quote, so the picture changes. Knowing which mode the account uses is step one before reading anything into a fill record. Some brokers also describe price improvement settings that net or cap the benefit, and the terms differ.

How many fills have you logged, and does positive slippage ever show up, or is it always zero or negative?
Post Reply