Footprint lesson · Tue Oct 20 2026 · ES DEC26 · 5-min Volumetric (delta view)
Reading the Big Print
Why −901 and −956 were the same size and meant opposite things, how +1,101 and +1,104 could be seen coming, and the three questions that replace fear with a checklist.
The morning, as it printed
Each number is the net delta at one price in one 5-minute bar: aggressive buys minus aggressive sells, in contracts. Green = buyers hit the ask, red = sellers hit the bid. The bar's color is price; the numbers are effort.
A print tells you how hard one side pushed. The next bar tells you whether it worked. You need both, and the second one is the one that matters.
Fear comes from reading the first part alone. A −956 looks like a wall of selling and the instinct is to get out of the way. But a wall of selling that doesn't move price isn't a wall, it's a trap. The buyer who absorbed it is bigger than the seller, and he's now sitting on 956 contracts of trapped shorts who will have to buy back.
Four prints, same size, opposite outcomes
9:30 → 10:00. Left to right: −995 (9:30), −901 (9:40), the 9:50 recovery bar, −956 with +1,101 directly below it (9:55), and +1,104 (10:00, top right).
| Time | Print | Where | Next bar | Answered? | Verdict |
|---|
| 1 | 9:30 | −995 @ 7724 | Just under the 7727–29 edge | Fell 2 pts, then stalled | +539 / +222 right below | Absorbed. First clue the sellers were being fed |
| 2 | 9:40 | −901 @ 7721 | Mid-air, below the edge | Fell 10 pts to 7711 | No | Push-through. Real selling. You were right to fear this one |
| 3 | 9:55 | −956 @ 7729 | Retest of the broken edge from below. The textbook short location | Fell zero. A 2-tick bar that closed flat, above 7727 | +1,101 one tick below, same bar | Absorbed. The obvious short got filled and went nowhere |
| 4 | 10:00 | +1,104 @ 7733 | Above the edge, price already reclaimed | Rose 5 pts, closed on its high | −377 tried, lost | Push-through. Trapped shorts covering plus new buyers |
Look at 2 and 3 together. −901 and −956 are the same number. One moved price ten points, the other moved it none. That's the entire difference between "sellers winning" and "sellers trapped," and it is invisible in the print itself. It only shows in what came next.
Why +1,101 was not a surprise
Walk the sequence as the market would have seen it:
- 9:45–9:50 the low held twice at 7710.50, and the pre-open buyer at 7704–7712 (the +889 / +988 prints at 8:30) had already shown that a real bid lived down there. The second hold is the spring.
- 9:50 recovered 20 points on moderate green prints. Price up with buying = honest, not a short squeeze yet.
- 9:55 arrived at 7729 — the underside of the broken 7727–29 edge. This is where every short on the planet sells: "broken support becomes resistance." So a big red print here is expected. The information is not that it appeared; it's whether it worked.
- It didn't. The bar had a two-tick range. −956 of selling moved price nothing. Somebody took all of it at 7728.50, and the footprint shows him: +1,101.
- 10:00 confirmed it: +1,104 at 7733 with price rising through it, and the −377 attempt one tick above was overrun. The trapped shorts from 7729 are now buying back, and the market goes looking for the next seller. He wasn't there until 7786.
You didn't need to predict the +1,101. You needed to see that the −956 failed, which the bar close told you at 9:59:59, and then buy the 10:00 bar's close with the +1,104 on it. The entry was a reaction, not a forecast.
The three questions, in order
1Where is it?
At a level, or in the middle of nowhere? A big print at a level is expected traffic; its failure is information. A big print in mid-air with nothing around it usually just means what it says.
2What did price do in the next 1–2 bars?
More than ~4 pts in the print's direction = push-through, go with it. Less = absorbed, lean the other way. This is the spring-vs-trap magnitude rule in a different coat.
3Did the other side answer?
An opposite print of at least half the size, within two ticks, in the same or next bar. That's the absorber showing himself. −956 → +1,101 one tick below is as clear as it ever gets.
Two yes's out of three is the signal. You rarely get all three, and you never need all three.
What it looks like as a picture
Same-sized red print, two different next bars. The next bar is the verdict.
The same rule later in the day
10:30 → 11:15. The −948 at 7786 inside the 7791–7803 stack, the stall, and the 11:15 bar that fell 34 points on −437 / −418.
| Time | Print | Where | Next bars | Verdict |
|---|
| 10:35 | −948 @ 7786 | Top of the 7791–7803 ★★★ stack, on a rally where price (+72) had run far ahead of CVD (+4.7k) — short covering, not fresh buying | Price rose 8 more points, then stalled for 35 minutes, then made a lower high at 7792 | Absorbed first, then won. A seller at a ★★★ stack who gets absorbed once and then holds the lower high is the fade; the trigger was the lower high, not the print |
| 11:15 | −437 / −418 @ 7773–75 | At the broken Fri singles, which should have been support | Fell 34 points in one bar | Push-through. Real. Anyone who bought "pullback #1" at 7770–73 was run over |
Notice the −948 case is the hard one. The print was absorbed for one bar, which by question 2 says "lean long." But question 1 says it sat at a ★★★ stack on a covering rally, and the answer that came back (+658 above it) didn't produce a new high that held. When the three questions disagree, wait for structure: the lower high at 11:05 settled it.
Why the top took 30 minutes and the bottom took 5
Same-sized prints, opposite speeds. At 9:55 the −956 was one seller against a trapped crowd: the shorts who sold the 7727 break had stops above, so once the level held they were forced to buy. At 10:30 the −948 was one seller against a comfortable crowd: a 72-point drive with CVD +4.7k and VIX collapsing. Nobody was forced to do anything until the longs who bought the last high were underwater.
The bottom, bar by bar
Forced covering: the reversal is one bar wide.
| Bar | Biggest green | Biggest red | Low | Read |
|---|
| 9:40 | — | −901 | 7711 | Push-through. Real selling, no answer |
| 9:45 | +132 | −206 | 7710.50 | First hold, small prints |
| 9:50 | +392 | −174 | 7710.50 | Double bottom. Same low, buyers bigger than sellers |
| 9:55 | +1,101 | −956 | 7726 | Seller sells the broken-edge retest, gets eaten one tick below. Bar closes above 7727 |
| 10:00 | +1,104 | −377 | 7729 | Trapped shorts cover. Buy this close |
The top, bar by bar
Voluntary distribution: the seller works every push a little lower while the green numbers shrink.
| Bar | Biggest green | Biggest red | High | Read |
|---|
| 10:30 | +658 | −948 | 7791 | Seller appears, gets absorbed. Warning only |
| 10:40 | +511 | −477 / −330 | 7787 | Buyers weaker, seller still there |
| 10:50 | +265 / +212 | −210 / −202 | 7786 | Both sides shrinking. Stall |
| 11:00 | +87 / +80 | — | 7794 | New high on tiny delta: the last buyers. A high made with no effort behind it is the weakest kind |
| 11:05 | +461 | −281 / −263 / −251 / −383 | 7792 | Lower high. Reds now outnumber greens |
| 11:10 | +217 | −265 / −100 | 7782 | Closes below 7786, the price of the first big red print. Short this close |
| 11:15 | — | −437 / −418 | — | Push-through. −34 in one bar |
The tell at a top is never one print. It is the green deltas shrinking on each push (658 → 511 → 265 → 87) while the reds grow. The 11:00 high is the clearest version of it: price at the high of the day, and only +87 behind it. Effort to go up had run out before price did.
The two rules, side by side
BOTTOMFast — act on the first confirming bar
A big green print at a ★★ or better support is the warning. The trade is two of these three, usually within 1–2 bars:
- Undercut stays shallow: under the level by less than the pad (~4 pts). Deeper is a break.
- Price stops making new lows: a second hold at the same price, or a higher low.
- A 5-min close back above the broken edge, or above the price of the last big red print.
Buy the close of that bar. Stop under the low plus the pad. Target the first magnet above.
Fails if price keeps making lows 5+ pts per bar on green prints. That is the trap, not the spring.
TOPSlow — wait for structure
A big red print at a ★★★ stack is the warning; it will usually be absorbed first. The trade is all three:
- A lower high.
- Green deltas shrinking push by push while reds grow.
- A 5-min close below the price of the first big red print.
Short the close of that bar. Stop above the lower high plus the pad. Target the broken singles.
Fails if a new high prints on a growing green print. Buyers are not exhausted yet.
| Bottom | Top |
|---|
| Why the speed differs | Shorts who sold the break are forced to cover once the level holds | Longs who bought the high are not forced until the lower high prints |
| Typical time to trigger | 1–2 bars | 20–40 min |
| Entry style | Responsive: first reversal bar, don't wait for more | Structural: lower high + close below the print |
| Proof | Oct 19 9:55–10:15 (7728.75 → 7772) · Oct 20 9:45–10:00 (7710.50 → 7794) | Oct 20 10:30–11:15 (7794 → 7747) |
ES bottoms are usually V-shaped and tops are usually rounded for exactly this reason: capitulation is forced, distribution is voluntary.
How to use this tomorrow
Before you look at the number, look at the price axis
Is this print at a level from the card? If yes, it's expected traffic and you're watching for failure. If no, it's probably just what it says.
Let the bar close
Never act on a print mid-bar. The close tells you whether the print worked. A red print inside a bar that closes flat or green is the mismatch, and the mismatch is the setup.
Enter on the next bar, not the one with the print
Absorbed red print → buy the following bar's close if it's green with a green delta. Stop goes under the absorbed bar's low plus the 6% ATR pad (about 4 points). Today: buy ≈7733 at 10:00, stop 7722.
Check the two referees
CVD should be turning the same way (it was rising from −2,064 at 9:55). VIX should not be rising against you (it was slipping from 18.78). If either disagrees, size down or wait one more bar.
Keep a tally of prints that failed
Two failed red prints at the same level in a row (−995 absorbed, then −956 absorbed) is a bigger tell than either one alone. The sellers are repeatedly trying the same door and it keeps not opening.
Stop asking "is this print big?" Ask "did it work?" The footprint answers that one bar later, every time.