Page 1 of 1
IC Markets vs Tickmill on USDJPY: cost and fill notes
Posted: Mon Sep 14, 2026 11:35 pm
by LondonScalper
IC Markets vs Tickmill on USDJPY: cost and fill notes.
USDJPY has its own open personality and Tokyo/London handoff quirks. Venue comparison needs session tags -- a blended daily average hides whether one book is worse exactly when you trade.
Fields
Spread at click, slip, rejects, and hold-time cost impact at your typical scalp duration. Round-number minutes get a separate filter in my sheet.
Again: logs over lore. If you only have anecdotes, say so -- anecdotes are still data if labelled honestly.
Anyone running side-by-side yen logs care to share patterns (not recruiting links)?
Tokyo open versus London open deserve separate lines in the sheet. A venue that is fine at 08:00 London can feel different at 00:00 if you ever swing that window. Tag the session or the comparison is mush.
I am interested in how others on this desk handle the same problem without turning it into folklore. Concrete rules and log fields beat slogans. If you have a version of "IC Markets vs Tickmill on USDJPY: cost and fill notes" that survived contact with live sessions, what does the rule look like on a sticky note -- and what did you try that failed?
Re: IC Markets vs Tickmill on USDJPY: cost and fill notes
Posted: Sun Oct 11, 2026 7:40 pm
by Shadow Trader
The session tag point is the most useful thing here. On USDJPY I'd go one step further and tag the Tokyo fix window separately, roughly 00:45 to 01:05 London in winter. My spreads and slippage in those 20 minutes are worse than any other part of the Asian session, on both venues I've used.
If those tickets are mixed into an Asia average, the comparison partly depends on how often you happened to trade near the fix.
On round number minutes, I agree, and I'd include the top of the hour as well as the half hour. On a quiet day the minute after 02:00 or 03:00 London can show odd spikes on yen pairs that don't happen at 02:07.
Something I'd add to your fields: the swap at the rollover if you ever hold through it. On USDJPY the swap difference between two accounts can matter more than the spread for anyone who occasionally lets a trade run overnight.
How many yen tickets per session do you need before you trust the comparison? Asia gives me so few that I've been pooling three months.
Re: IC Markets vs Tickmill on USDJPY: cost and fill notes
Posted: Mon Oct 12, 2026 11:01 pm
by Shadow Trader
Coming back to the Tokyo and London handoff, the clocks make this comparison harder than it looks. Japan doesn't change its clocks and the UK does, so the gap between the Tokyo afternoon and the London open moves by an hour twice a year.
If a log uses broker server time, half the year's Tokyo tickets end up an hour out of place. When I converted my sheet to UTC, a pattern I thought I'd found simply disappeared.
The other edge case is Japanese holidays. On those days USDJPY in Asian hours behaves more like a thin cross, and spreads were clearly wider on both accounts I compared, so I now leave those days out entirely.
On anecdotes being data if labelled, I agree, with one caveat. Bad fills get remembered and good fills never do, so a labelled anecdote probably deserves a lower weight than a logged ticket.
Has anyone tried sending the same small ticket to both accounts in the same second? It's the only comparison I've found that takes the timing question out completely.
Re: IC Markets vs Tickmill on USDJPY: cost and fill notes
Posted: Sat Oct 31, 2026 8:08 am
by Shadow Trader
Every scalper has had one: a fill that was better than the quote by half a pip. A buy at 1.0850.0 when the screen showed 1.0850.5. It feels like skill or luck, and the temptation is to remember it and forget the rest.
Fill quality is a distribution, with good and bad outcomes. The average matters and the extremes mostly don't. A few great fills don't offset a steady stream of mediocre ones.
An honest evaluation needs numbers. Record the quote at click time and the fill price for every market order, over hundreds of trades. Compute the average, the median, and the percentage of fills better and worse than the quote.
Many find that their memory is skewed. The good fills are vivid and the bad ones are blamed on bad luck or the broker. The measured result shows a consistent small negative average, which is the actual cost of execution.
Comparing results across brokers or times of day with the same method shows whether any improvement is real. If a change, such as moving to a VPS, produces a shift in the average of 0.2 pips, that's a measurable improvement.
Chasing good fills by timing orders to the second is a losing activity. The aim should be reducing the average cost, not collecting good stories.
An easy way to keep score honestly is a three-column log: quote at click, fill price, and signed difference in the trader's favor or against. Add a fourth column for the spread at the time, and a fifth for the seconds since the last tick.
After a few hundred rows, patterns emerge. Slippage tends to grow when the time since the last tick is long, since a long pause means the quote was stale by the time of the click. It also grows with the spread.
An even more valuable analysis separates fills by direction relative to a short-term move. If fills in the direction of the last few ticks are consistently worse, which is the case when chasing, the cost of chasing is quantifiable in pips.
Once that number is known, it can be compared with the average gain from chasing, and the decision becomes arithmetic.
Has the average fill slippage been measured, and what did it come to?