Advertisement IC Markets

Who publishes real average execution reports you can trust?

Compare ECN/Raw spread brokers, analyze execution speeds, report slippage, and evaluate commission structures for high-frequency traders.
Post Reply
LondonScalper
Posts: 965
Joined: Sat Sep 05, 2026 7:54 am

Who publishes real average execution reports you can trust?

Post by LondonScalper »

Some brokers publish execution statistics: average slippage, percentage of orders filled at the requested price, average execution speed. The question is which of these you can trust.

What I look for in a report:
- A defined time period and sample size. "Average execution 30 ms" with no dates or order count tells you nothing.
- Breakdown by order type. Market orders, limit orders and stop orders behave very differently. A report that only mentions limit orders is hiding the interesting part.
- Breakdown by time or market conditions. Execution during news is where problems happen.
- Positive and negative slippage reported separately, not just a net figure.
- Some form of external audit, or at least a clear methodology.

In practice, very few retail broker reports meet all of these. Some regulators require execution quality reports, and those are more structured, but they can be hard to read and are not always specific to retail CFD accounts.

My approach: I treat published reports as a starting point and compare them with my own data. My deals tab over a few hundred trades is the only report I fully trust, because I know exactly how it was collected.

Does anyone know of a broker whose published stats matched their own experience closely? I would be curious which ones.
Recommended broker for automated trading & scalping IC Markets
Fairman
Posts: 6955
Joined: Tue Jul 21, 2026 7:11 am
Location: Abuja

Re: Who publishes real average execution reports you can trust?

Post by Fairman »

LondonScalper wrote: Some regulators require execution quality reports, and those are more structured
On the regulatory reports, it's worth knowing that most of them have gone. In Europe, MiFID II required firms to publish quarterly execution quality reports, known as RTS 27, and annual top-five venue reports, RTS 28. RTS 27 was suspended in 2021 because the reports were barely used, the UK FCA removed both requirements for UK firms around the same time, and the EU has since dropped them as well. So a broker regulated in the UK or EU today usually has no obligation to publish the kind of fill statistics you describe, and the ones that do are choosing to, which makes the methodology questions on your list even more important. In the US, the Rule 605 execution reports cover equities, not retail forex. That leaves your own logs as the main source.
It’s Fairman :geek:
Shadow Trader
Posts: 5594
Joined: Wed Sep 09, 2026 5:19 pm

Re: Who publishes real average execution reports you can trust?

Post by Shadow Trader »

Fairman's point about RTS 27 going away matches what I've seen. The few brokers still publishing stats now do it as marketing, and the methodology section is usually a single sentence.

The bigger issue for me is order mix. A headline of 92% filled at the requested price can be completely honest and still say nothing about my account, because it blends every order type the broker handled.

I checked one published monthly figure against my own deals tab, 410 trades over about three months. Limits filled at price 99% of the time, market orders 81%, and stop entries only 58%. Since most of my breakout entries are stops, my real number sat well below the headline.

The attached illustration is that comparison, with the news window split out as its own bar.
img_13330_r1.png
img_13330_r1.png (57.32 KiB) Viewed 149 times
Inside two minutes of a red folder release, my stop entries dropped to around 30% at price. None of that shows in a monthly average.

So I only trust a report if I can rebuild its number from my own log using the same buckets. If the buckets aren't published, it's a sales page.

Has anyone seen a report that gives positive slippage per order type, not just one net figure?
Shadow Trader
Posts: 5594
Joined: Wed Sep 09, 2026 5:19 pm

Re: Who publishes real average execution reports you can trust?

Post by Shadow Trader »

LondonScalper, since the published reports are mostly gone, the practical question is how to make your own deals tab answer your checklist. The fiddly part is the reference price, because the platform only shows the fill, not the price you saw when you clicked.

I log it with a small script that writes the bid and ask at the moment the order is sent, then compares that with the fill. Without that column, slippage is a guess based on whatever the chart showed a second later.

Sample size is the other trap. My slippage per trade has a spread of roughly 0.4 pips, so to pin an average down to within 0.1 pip I need about 60 trades in each bucket, and the news bucket took me close to five months to fill.

Your point on positive and negative slippage is where the log earns its keep. I count how often a fill improves on the requested price when the market moved my way, compared with how often it worsens when the market moved against me.

If those two rates are close, I am fairly relaxed. If improvements almost never happen while the bad fills arrive every time, that asymmetry tells me more than any average, and it is the first number I would ask a broker for.
Shadow Trader
Posts: 5594
Joined: Wed Sep 09, 2026 5:19 pm

Re: Who publishes real average execution reports you can trust?

Post by Shadow Trader »

Two monthly returns: +10% and -10%. The arithmetic average is 0%. The account, starting at 100, goes to 110, then 99. It's down 1%. The geometric average, which reflects actual compounding, is slightly negative.

Reporting arithmetic averages overstates what an account actually earns. The more volatile the returns, the bigger the gap.

For scalpers reviewing monthly or yearly performance, the geometric return, the compound growth rate, is the honest measure of what happened to the capital. It's calculated by multiplying the growth factors, taking the nth root and subtracting one.

When comparing strategies or projecting the future, the geometric view prevents flattering illusions. A strategy with a 3% average monthly return and wild swings may grow more slowly than one with 2% and calm.

The distinction also matters when reading other people's claims. "Averaged 8% a month" could hide a sequence like +40%, -20%, +10%, which compounds to far less.

Spreadsheet functions like GEOMEAN make the calculation easy once the returns are converted to growth factors.

A related trap is quoting "average return per trade" without compounding. A series of +1R and -1R trades that nets out to zero in R terms can still shrink an account when position sizes scale with equity, since the same percentage loss after a gain is larger in currency terms than the gain was.

The practical rule: report growth as a compound rate whenever the claim is about capital, and as expectancy in R whenever the claim is about the strategy. Mixing them is a common source of confusion in forum threads.

Which average is used to report results, and how much lower is the compound rate than the arithmetic one?
Post Reply