Module 4 · Lesson 12 of 12 · Volume and indicators · About 7 minutes
Putting Volume Tools Together: One Session, Three Tools, One Plan
On this page
Before you read: Overfitting and Repainting: When Indicator Results Look Too Good
What you will learn
- How to give each tool one job in a plan: context, condition or distance.
- Why tools built from the same data can repeat each other instead of adding information.
- How each extra condition shrinks the number of trades a test can count.
- How one fictional plan using VWAP, relative volume and ATR played out on three sessions.
What job does each tool do?
If three indicators all point the same way, have you learned three things or one?
The earlier lessons in this module explain what each tool measures and what it can’t tell you. A plan has a different question: what is each tool for? One useful way to sort them is by job:
- Context: tells you which kind of session it is, and whether the plan is active at all.
- Condition: a specific event that has to happen before entry.
- Distance: sets how far away the exit and the reference points are, and so the position size.
A tool with no job in the plan is decoration. Two tools doing the same job are often one tool counted twice.
When do tools repeat each other?
What does a second volume-based line add if the first one already uses volume?
Many indicators are built from the same few inputs: open, high, low, close and volume. Session VWAP weights price by volume. On-balance volume and the accumulation/distribution line add volume up by different rules (OBV, A/D and relative volume). Moving averages, RSI and MACD are all built from closing prices. When two tools share their input, agreement between them is partly arithmetic. It isn’t two independent confirmations.
Giving tools different jobs keeps this visible. In the plan below, VWAP gives context, relative volume is the condition, and ATR sets the distance. The plan names each tool’s settings, so someone else could rebuild every number.
John's plan: VWAP, relative volume and ATR
Fictional example. John, Harbor Index futures, the prices, times, costs and results are all invented for teaching. This shows how a written plan can be built and tested. It is not a recommendation, and nothing here says how often this idea works.
Try it on a real chart. Test this idea yourself on past sessions with bar replay. Keep your own count, including the days with no trade.
John trades a fictional market: tick 0.05 = $5, so 1 point = $100 per contract. His costs are $5 per contract round trip, plus 1 tick of slippage on market and stop orders. Limit orders fill at their price.
The idea. On a session that has stayed above VWAP, John waits for a quiet pullback to VWAP, followed by a busier bar that closes up. ATR sets how far away his exit and references are.
John's plan
- Setup
- every 5-minute close from 08:30 to 09:30 is above session VWAP
- Enter if
- price pulls back to within 0.10 of VWAP on quiet bars (relative volume under 1.0). Then a bar closes higher than it opened, with relative volume of 1.2 or more. Buy.
- Wrong if
- price trades half an ATR below that bar's low (stop order)
- Manage
- take half at entry + 1 ATR, the rest at entry + 2 ATR or by 14:30
- Risk
- $150 maximum, costs included. Size depends on the distance.
Three outcomes
-
Worked: the pullback held → +$140
-
Failed: quiet pullback, busy up bar, then price fell through the stop → −$120, inside the limit
-
No trade: the pullback came on heavy volume, not quiet volume → $0
Show the full replays
| Item | Plan |
|---|---|
| Context (VWAP) | Session VWAP from 08:30 CT. Every 5-minute close from 08:30 to 09:30 is above it |
| Condition (relative volume) | Relative volume = a 5-minute bar's volume ÷ the average volume of the same 5-minute bar over the prior 20 sessions. Pullback bars within 0.10 of VWAP are all under 1.0, then a bar closes above its open at 1.2 or more. Buy 2 at market after that close |
| Distance (ATR) | ATR(14) on 5-minute bars, read at the entry bar's close. Stop order half an ATR below that bar's low, rounded down to a tick |
| Management | Take profit on half at entry + 1 ATR (limit) and move the stop for the rest to the entry price. Rest at entry + 2 ATR (limit). Time limit 14:30 CT |
| Risk cap and size | $150. Cost per contract = (expected entry − stop + 1 tick of stop slippage) × $100 + $5. Contracts = $150 ÷ that, rounded down |
| Not trading if | Scheduled major release within the hour, or no setup by 12:00 CT |
Replay A: worked
| Time (CT) | What happened | John |
|---|---|---|
| 08:30–09:30 | Every close above VWAP | Plan active |
| 09:45–10:00 | Three bars pull back to a low of 100.66. VWAP 100.60. Relative volume 0.7, 0.8, 0.6 | Quiet pullback: watching |
| 10:00–10:05 | Opens 100.68, closes 100.80, low 100.66, relative volume 1.4. ATR 0.60 | Condition met. Stop: 100.66 − 0.30 = 100.36, rounded down to 100.35. Cost about $60 per contract, so 2 contracts. Buys 2 at market, filled 100.85 |
| 10:05–11:05 | Trades 101.45 (entry + 0.60) at 11:05 | Sells 1 at 101.45 (limit). Stop for the other moves to 100.85 |
| 11:05–14:30 | High 101.95, never back to 100.85, never 102.05 | At 14:30 sells the last 1 at market: price 101.80, filled 101.75 |
Result after costs: +0.60 point = +$60, +0.90 point = +$90, commissions −$10. Net +$140. Journal: "Each tool did its one job. The second reference missed by 0.10, and the time limit closed it."
Replay B: failed
| Time (CT) | What happened | John |
|---|---|---|
| 08:30–09:30 | Every close above VWAP | Plan active |
| 10:10–10:20 | Two bars pull back to a low of 100.66. VWAP 100.60. Relative volume 0.8, 0.9 | Watching |
| 10:20–10:25 | Opens 100.70, closes 100.80, low 100.66, relative volume 1.3. ATR 0.60 | Condition met. Stop at 100.35, 2 contracts. Buys 2 at market, filled 100.85 |
| 10:25–10:40 | Turns down, trades 100.35 | Stop triggers, filled 100.30 |
| 10:40–15:00 | Falls to 99.90 | Done for the idea |
Result after costs: 2 × −0.55 point = −$110, commissions −$10. Net −$120. Journal: "Every condition was met, and price still went through VWAP. Three tools agreeing isn't a promise."
Replay C: no trade
| Time (CT) | What happened | John |
|---|---|---|
| 08:30–09:30 | Every close above VWAP | Plan active |
| 09:50–10:00 | Two bars pull back to 100.65. VWAP 100.58. Relative volume 1.6, 1.8 | Pullback on heavy volume: condition not met |
| 10:00–12:00 | Falls through VWAP to 100.10 | No setup. 12:00 cutoff |
Result: $0. Journal: "The pullback was busy, not quiet, so the rule said no. Fine either way, but the rule decided, not me."
Test it yourself
- Write every setting first: the VWAP start time, the relative-volume baseline, the ATR length and bar size.
- Replay every session where the setup was met, including the no-trade days (bar replay).
- Change one setting at a time (a 10-session baseline, ATR(20)). If the results flip, the plan is fragile (overfitting).
Where it breaks
- VWAP and relative volume both use volume. On a thin day, both can be distorted at once.
- ATR changes during the day, so the stop distance and size change with it.
- All three tools can line up, and price can still fail (the failed replay).
Calculators: position sizerisk/rewardexpectancy
What happens when you add more tools?
If one more condition makes a plan look safer, what does it do to your sample?
Each condition you add removes sessions from the test. The count below is invented numbers: this shows how to count, not how the method performs.
| Fictional 60-session test | Sessions that qualify |
|---|---|
| Setup only (every close above VWAP for the first hour) | 31 |
| Setup + John’s full entry condition | 14 |
| …+ “RSI(14) above 50 at entry” | 6 |
| …+ “MACD line above its signal line” | 2 |
With 2 trades, almost any result is luck. Every variation tried also counts as a test, so trying ten filters and keeping the best one makes it likelier the “best” one fits only this sample (multiple testing). A plan with fewer conditions, each with a clear job, is easier to test honestly.
Three-tools diagram
Illustrative diagram constructed for teaching. Not real market data.
A fictional 5-minute chart with the session VWAP line. Shaded bars mark a pullback to VWAP. A relative-volume panel underneath shows those bars below 1.0, then one bar at 1.4. A bracket below the entry is labeled “half an ATR below the low: the stop distance”. A caption reads “Context, condition, distance: one job per tool.”
What this does not tell you about combining tools
- That more tools make a plan better. Extra conditions shrink the sample and add more chances to overfit.
- That agreement between tools built from the same data is independent evidence.
- How often this fictional plan, or any real one, works.
- Nothing here suggests buying, selling or holding at any VWAP, volume or ATR reading.
Check yourself
Reflect on these; there is no score.
- In John’s plan, which tool gives context, which gives the condition, and which sets the distance? Hint: VWAP, relative volume, ATR.
- Why can agreement between RSI and MACD be partly arithmetic? Hint: what are both built from?
- In the invented count, what happened to the sample as each condition was added? Hint: 31, 14, 6, 2.
- In Replay A, how was the stop level found, and why was it rounded down? Hint: 100.66 − 0.30 = 100.36, and the tick size is 0.05.
Reflect on your own process
- For each indicator on your chart, can you name the one job it does in your plan?
- Do any two of your tools use the same input data, and if so, what does the second one add?
- How many trades did your last test count, and how many conditions did the plan have?