Rally Base Drop: The Four Zone Types, Counted on 33 Years of SPY
Rally base drop (RBD) is a supply zone: price rallies, pauses in a few small candles called the base, then drops away. Its three siblings are rally base rally, drop base rally and drop base drop, named for the move into the base and the move out of it.
Rally base drop, usually shortened to RBD, is a supply zone drawn around a short pause between a rise and a fall. It belongs to a family of four labels used in supply and demand trading, and each one names the move into the pause and the move out of it. This page sets out the four types, turns them into a rule a computer can count, and measures how often each one held on SPY from 1993 to September 2026.
How it forms
Three pieces, in order. A leg in, a base, and a leg out. The legs are large candles with most of their range in the body, so price clearly traveled. The base is one or a few small candles where price went nowhere. The zone is drawn around the base.
The name is just the two legs. Up into the base and up out of it is a rally base rally (RBR). Down in and up out is a drop base rally (DBR). Up in and down out is a rally base drop (RBD). Down in and down out is a drop base drop (DBD).
The leg out decides the side. If price left the base going up, the base becomes a demand zone and traders look to buy when price comes back down to it. If price left going down, it becomes a supply zone and traders look to sell on the way back up. So RBR and DBR are demand, and RBD and DBD are supply.
The leg in decides the story. RBR and DBD are continuation shapes: the market paused and carried on. DBR and RBD are reversal shapes: the market paused and turned. The wider framework, and why the base is where the zone gets drawn, is on the supply and demand page.
A rule precise enough to count
Most descriptions of these zones say “strong” legs and a “tight” base without a number, which makes them impossible to test. The count below uses one fixed rule on SPY daily bars.
Average true range. The mean true range of the 14 sessions before the candle being judged, so the yardstick never includes the candle itself. The average true range page covers the measure.
Leg candle. High minus low at least one average true range, and a body (open to close) at least half of that range. A green one is a rally leg and a red one is a drop leg.
Base candle. High minus low under one average true range, and a body under half of that range.
The pattern. A leg candle, then one to three base candles in a row, then a leg candle whose close lands outside the base: above its high for a rally out, below its low for a drop out.
The zone. From the lowest low of the base to its highest high.
A return. The first session within 60 sessions of the leg out that trades back into the zone.
Held or broke. After the return day, the zone held if price traveled one average true range (as measured on the return day) beyond the near edge of the zone before any close beyond the far edge. It broke if a close beyond the far edge came first, the return day included, and a session that did both counted as a break. Twenty sessions, the return day among them, were allowed for either to happen.
A worked example
A drop base rally on SPY, March 2024.
- Leg in, Friday 8 March. SPY opened at $515.46 and closed at $511.72. The day’s range was $7.09 against an average true range of $4.63, and the $3.74 body was 52.8% of that range. A drop leg.
- Base, Monday 11 March. High $511.88, low $508.50, so a $3.38 range, well under the $4.84 average, with a body of $0.80. One base candle.
- Leg out, Tuesday 12 March. SPY opened at $513.45 and closed at $516.78, above the base high of $511.88. Range $6.52 against $4.72, body 51.1% of the range. A rally leg, so the base is a demand zone from $508.50 to $511.88.
- The return, Thursday 14 March. The low was $511.82, six cents inside the top of the zone. The average true range that day was $4.31, so the zone would count as held once price reached $511.88 + $4.31 = $516.19.
It nearly broke first. On Friday 15 March the low reached $508.12, below the bottom of the zone, but the close was $509.83, still inside it, so under the close rule the zone was still alive. A buyer at the top edge would have been $3.76 a share against them at that low.
Then it held. Highs of $515.48 on 18 March and $516.00 on 19 March fell short, and on Wednesday 20 March SPY reached $520.62 and closed at $520.48. The zone counts as held, four sessions after the return.
The original data
The worked example, drawn. The chart below shows the March 2024 zone from the example above, with every candle taken from the same SPY file.
223 zones in 33 years. Across 8,472 SPY daily bars from 29 January 1993 to 25 September 2026, the rule found 46 rally base rally, 63 drop base rally, 54 rally base drop and 60 drop base drop zones. 154 had a single base candle, 55 had two and 14 had three. Every zone, its dates and its result are in the SPY zone file.
Almost all of them were revisited, and quickly. Price came back into 42 of the 46 RBR zones (91.3%), 54 of 63 DBR (85.7%), 51 of 54 RBD (94.4%) and 59 of 60 DBD (98.3%) within 60 sessions. The median first return came 2 sessions after the leg out.
Once revisited, the zones held about half the time. RBR held on 22 of 42 returns (52.4%), DBR on 29 of 54 (53.7%), RBD on 23 of 51 (45.1%) and DBD on 24 of 59 (40.7%). Every return was settled one way or the other within the 20 sessions allowed.
The comparison that matters is a plain box. To see whether the leg-base-leg shape adds anything, the same return test was run on every session’s own high-to-low range, once the next session closed outside it. There were 2,663 such demand boxes and 1,964 supply boxes. Demand boxes held on 1,310 of 2,359 returns (55.5%) and supply boxes on 841 of 1,882 (44.7%). The base-rate boxes are in the plain box file.
No zone type did measurably better than its box. Three of the four held less often than their box. Rally base drop held 45.1% against 44.7%, a gap of 0.4 points. Against those base rates, the exact binomial test gives p = 0.757 for RBR, 0.786 for DBR, 1.0 for RBD and 0.601 for DBD. Pooled, demand zones held 51 of 96 returns (53.1%, p = 0.682) and supply zones 47 of 110 (42.7%, p = 0.702). On this rule and this market, the label did not separate the zones from any box price had just left.
Why demand held more than supply
The gap follows the side, not the shape. Both demand types held more often than both supply types, and plain demand boxes beat plain supply boxes by a similar margin. A likely reason is that SPY rose over the period, so a level below the price had the trend behind it and a level above had the trend against it.
Continuation and reversal came out close. Adding the two continuation shapes gives 46 holds in 101 returns (45.5%), and the two reversal shapes 52 in 105 (49.5%). That mixes demand and supply, and the side difference is larger than the continuation-reversal one.
Samples are small. Forty-odd returns per type can only rule out large differences. A zone type that truly held five points more often than a plain box would not reliably show up in counts this size.
When it fails
A drop base drop that broke, July 2026. SPY fell on Monday 27 July, opening at $744.91 and closing at $739.09. On 28 July it made a small base, high $742.79 and low $735.98. On 29 July it dropped out, closing at $729.46. That marked a supply zone from $735.98 to $742.79.
It lasted two sessions. SPY opened back inside the zone at $736.05 on 30 July and closed at $741.69. On Friday 31 July it closed at $747.03, above the top of the zone, which counts as a break. By 4 August SPY had closed at $771.33, $28.54 above the zone.
Fast returns cut both ways. The median return came two sessions after the leg out, so most zones are tested while the move that created them is still fresh. That is when a strong market is most likely to run straight through.
Loose definitions hide the failures. Without numbers for “strong” and “tight”, any pause can be called a base after the fact, and the zones that broke are easy to redraw or forget. The fixed rule above is what made the 108 breaks countable.
One market, daily bars. These results are SPY only, on one definition. A different average true range length, a different target or a single stock could change the counts. They have to be measured there before being trusted.
Related
The supply and demand page is the parent idea and explains why the base is where unfilled orders are thought to sit. The demand zone page covers the buying side on its own and a stricter way to draw it. The steps for trading a zone, from marking it to placing the stop, are in how to trade a supply and demand zone. And the average true range page explains the yardstick this page used to tell a leg from a base.
Write down what a leg candle and a base candle are before you mark a single zone, and then count the ones that failed as well as the ones that held. If the boxes on your chart do no better than a box drawn around any quiet day, the label is not doing the work.
— Michael Whitman
This page is educational, not financial advice. Test every idea on your own charts before risking money.