A slippage log does not need a database. A spreadsheet with a few columns is enough, as long as the same fields get filled in every time:
- Requested price at the moment of the click, taken from the chart or the order ticket, not from memory.
- Filled price from the deal history.
- Direction, so slippage can be signed. Negative is against the trade, positive is in its favor.
- Order type: market, limit or stop. They behave differently and mixing them hides patterns.
- Spread at entry, and the time to the second.
- Context tag: news, session open, quiet range, fast candle.
Convert everything to pips or points in the sheet so pairs are comparable. After thirty or forty trades, a couple of things usually show up: average slippage by order type, and average slippage by context. The tag column is where the interesting part lives, because slippage tends to cluster around a handful of moments instead of spreading evenly.
Positive slippage matters as well. Some brokers pass it back and some do not, and a log that records only negative events gives a skewed picture.
The log should also record stop fills separately from entries. A stop that fills three pips beyond its level behaves like a bigger loss than planned, and that never shows up in a plain win-loss count.
Which tags or columns has anyone added to a slippage log that turned out to be more revealing than expected?
A slippage log that takes thirty seconds per trade
-
Shadow Trader
- Posts: 5594
- Joined: Wed Sep 09, 2026 5:19 pm
-
FXS London Scalper
- Posts: 978
- Joined: Sun Oct 18, 2026 12:25 pm
Re: A slippage log that takes thirty seconds per trade
Taking the requested price from the chart, not memory, is the key point. Adding the spread at the moment of the click as a separate column makes it easy to see if slippage is worse at certain spreads.
FXS Official Team | London Session Desk
-
Amanhasnoname
- Posts: 2060
- Joined: Fri Oct 30, 2026 9:45 am
Re: A slippage log that takes thirty seconds per trade
The column that surprised me most was seconds-to-fill on market orders. I added it almost as an afterthought, measuring from click to deal ticket. On GBPUSD during the first hour of London, fills under 180 milliseconds averaged 0.2 pip adverse slip, while anything over half a second averaged 0.9. Same broker, same size, different patience of the book.
I also split stop fills into a separate sheet with a column for how far past the level the deal printed. Stops that slipped more than 1.5 pips clustered on Mondays after thin weekends and on days when I sized above my usual 0.4 lot. Mixing those with entry slip had hidden the pattern for months.
Positive slip showed up too, about one trade in six on limit entries during quiet mid-morning. Recording only the painful fills made the broker look worse than the net picture. The tag that earns its keep for me is still latency plus size, not the vague news-or-quiet label I started with.
I also split stop fills into a separate sheet with a column for how far past the level the deal printed. Stops that slipped more than 1.5 pips clustered on Mondays after thin weekends and on days when I sized above my usual 0.4 lot. Mixing those with entry slip had hidden the pattern for months.
Positive slip showed up too, about one trade in six on limit entries during quiet mid-morning. Recording only the painful fills made the broker look worse than the net picture. The tag that earns its keep for me is still latency plus size, not the vague news-or-quiet label I started with.