A working ATAS setup needs four things: a cluster chart on volume based bars, a volume profile showing session and previous session levels, cumulative delta below price, and imbalance highlighting at a sensible ratio. Everything else is preference. This is the layout I use on ES.
ATAS is powerful and its defaults are not what you want. Most people who bounce off it do so in the first week because the out of the box configuration makes footprints look like noise.
Before the platform: the data
Worth stating first because it invalidates everything else if it is wrong.
Connect a feed that delivers unfiltered tick data. Rithmic and CQG both do. If your feed aggregates ticks, your cluster numbers are approximations and the imbalance highlighting is decorating estimates. See market data feeds.
Also decide your session times now. For ES I work on regular US hours for profile levels and view the overnight separately. Changing this later invalidates the levels you have been learning from, so pick one and stay with it.
Step 1: bar type
The single most important setting and the one nobody mentions.
Time based bars are a poor fit for footprint reading. A five minute bar at the open and a five minute bar at lunch contain wildly different amounts of activity, so half your footprints are dense and half are nearly empty.
Use volume bars instead. Each bar closes after a set number of contracts, so every footprint contains a comparable amount of information regardless of when it printed.
Starting point on ES: somewhere in the region of 3,000 to 5,000 contracts per bar for intraday work. Adjust until a bar takes roughly one to three minutes during normal activity. Too small and you drown in bars, too large and the detail disappears.
Range bars are a reasonable alternative. Time bars are the one to avoid while learning.
Step 2: the cluster chart
ATAS calls footprints cluster charts. Add one and configure it.
Mode: Bid x Ask. Both raw numbers side by side. Delta mode is faster to read once you are fluent, but bid and ask is the only mode that teaches you what the others are summarising. Start here even though it is slower.
Cluster size: one tick. Grouping multiple ticks per row smooths away the detail you are looking for. On ES that means 0.25 point rows.
Show cluster delta. The per-price net figure alongside the raw numbers.
Turn off the extras. ATAS will happily display a dozen values per cell. For the first month, bid, ask and delta only. You can add more once those three are automatic.
Step 3: imbalance settings
This is where the defaults cause the most trouble.
Ratio: 3 to 1. A price is flagged when one side is at least three times the diagonal opposite. The default is reasonable. Lowering it floods the chart, raising it hides genuine activity.
Minimum volume filter: on. The setting that matters most. Without it, a cell with 3 contracts against 1 flags as a 3 to 1 imbalance, which is meaningless. Set a floor high enough that only meaningful prints qualify. On ES a floor somewhere in the tens of contracts removes most of the noise, and you should tune it until flagged imbalances look genuinely notable rather than constant.
Stacked imbalance highlighting: on, minimum 3. Consecutive imbalanced prices are what actually matter, as covered in absorption, exhaustion and stacked imbalances. Single imbalances are common and mostly noise.
The ATAS cluster chart settings panel with the configuration described here visible: Bid x Ask mode, one tick cluster size, 3 to 1 imbalance ratio, minimum volume filter set, and stacked imbalance enabled. A straight screenshot of the dialog so readers can match it exactly.
Step 4: volume profile
Add the volume profile indicator and configure three instances.
- Current session, developing. Builds through the day. Shows POC, value area, VAH and VAL.
- Previous session, fixed. Yesterday's completed profile with its levels extended right across today's chart. Early in the session these matter far more than today's, since they are complete and everyone can see them.
- Composite, two to three weeks. The levels that actually hold. Set this to show the POC line only, otherwise the chart becomes unreadable.
Set the value area to 70 percent, which is the standard convention. Extend POC, VAH and VAL as horizontal lines across the chart rather than leaving them attached to the profile, because you need to see where price is relative to them at a glance.
Background on all of this in POC, VAH and VAL.
Step 5: cumulative delta
Add cumulative delta as a separate pane below price, not as an overlay.
Reset: session start. Ties the reading to today's participation. Whatever you choose, keep it consistent, because the same market produces very different looking delta depending on where the count begins.
Display: candles rather than a line. Delta candles show the range of aggression within each bar rather than only the closing value, which makes divergence easier to spot.
Detail in cumulative delta and delta divergence.
Step 6: the layout
What I actually keep on screen:
- Main cluster chart on volume bars, roughly two thirds of the screen.
- Volume profile attached to it, with previous session levels extended.
- Cumulative delta pane beneath, about a fifth of the height.
- A higher timeframe chart to one side for context, no footprint on it.
- A DOM, small, in the corner. Used for execution and for seeing thin areas, not for signals.
That is the whole workspace. No indicators, no oscillators. If you are coming from a heavily loaded chart it will look sparse, and after a fortnight it will look correct.
Step 7: replay
The most valuable feature in the platform and the one most people never configure.
ATAS market replay lets you load a historical session and step through it as if live. That means you can practise reading absorption at value area edges twenty times in an evening rather than waiting twenty days for the setups to appear.
If you do one thing from this article, do this one. Load a session from last week, replay it at moderate speed, and pause every time price reaches a level you marked. Ask what the footprint is showing before you let it run on.
Defaults worth changing immediately
| Setting | Default problem | Change to |
|---|---|---|
| Bar type | Time bars produce uneven footprints | Volume bars |
| Imbalance minimum volume | Tiny prints flag as imbalances | A floor that leaves flags genuinely notable |
| Cluster size | Grouped ticks hide detail | One tick |
| Cluster content | Too many values per cell | Bid, ask and delta only |
| Profile levels | Attached to the profile only | Extended across the chart |
Save the workspace once it is right. Then leave it alone for a month. The temptation to keep adjusting is strong and it prevents you learning what any single configuration actually looks like over time.
Frequently asked questions
What bar type should I use for footprint charts in ATAS?
Volume bars rather than time bars. Each bar then closes after a set number of contracts, so every footprint contains a comparable amount of activity. Time bars produce dense footprints during active periods and nearly empty ones when it is quiet, which makes consistent reading much harder.
What imbalance ratio should I set?
Three to one is a sensible default, but the setting that matters more is the minimum volume filter. Without a volume floor, a cell with three contracts against one registers as an imbalance, which tells you nothing. Raise the floor until flagged imbalances look genuinely notable.
What is a cluster chart in ATAS?
ATAS terminology for a footprint chart. It shows the volume traded at the bid and the ask at every price inside each bar. Cluster chart, footprint chart and order flow chart all refer to the same thing.
How many volume profiles should I display?
Three. The developing session profile, the previous session's completed profile with levels extended across today, and a composite of two to three weeks showing only its point of control. More than that and the chart becomes unreadable.
Should cumulative delta be an overlay or its own pane?
Its own pane below price. Overlaying it on the chart makes both harder to read, and divergence between price and delta is easier to spot when the two are plotted separately on aligned time axes.
How do I use ATAS market replay?
Load a historical session and step through it as if it were live, pausing whenever price reaches a level you had marked. It is the fastest way to build reading speed, because you can practise a setup repeatedly in an evening instead of waiting weeks for it to appear live.