Pattern Recognition for Penny Stocks: What 90 Days of Scanner Data Shows

Ninety days of pattern data tells you one thing louder than any single ticker: the setups repeat, but the volume signature under them decides whether you get a clean run or a trap. Over the trailing five sessions (Aug 18-24), five small-caps ran between +135.9% and +365.8% close-to-close, and the difference between the ones that held and the one that punished late buyers was written in the tape before the fade.

This is not financial advice. It is a data-verified breakdown of what the scanner recorded and how to structure a plan around it.

TLDR

• Five featured runners over Aug 18-24: RFAI +365.8% ($11.03 to $51.38), BTCT +285.6% ($0.45 to $1.74), DAIC +216.7% ($0.54 to $1.72), ZSTK +165.3% ($1.90 to $5.04), CYPH +135.9% ($0.71 to $1.68).
• ZSTK printed the best intraday window of the group: +715.8% true MFE on Aug 19, low $1.52 to high $12.40, before closing the regular session at $4.88. That fade is the risk lesson.
• Pattern follow-through held at 100%: 131 high-volume breakout setups (100M+ shares intraday) triggered over the last 30 days and all 131 hit target; 283 intraday-doubling setups fired and all 283 completed.
• This week's pattern count ran below normal: 188 patterns detected in the past 7 days versus the 90-day weekly average of 202.8.
• Volume signature separates the two archetypes: thin-float grinders (RFAI, 3,526,351 total shares over five sessions) versus high-turnover blowoffs (BTCT, 498,824,854 total shares). Same percentage, different trade.

What 90 Days of Scanner Data Actually Shows

The repeatable small-cap patterns are three: stocks that trade 100M+ shares intraday, stocks that double from their session low to their session high, and liquidity tests where market makers sweep a price level to probe supply before the real move. Over the past 30 days, all three carried a 100% follow-through rate in the data: 131 high-volume breakout setups triggered and every one hit its target, and 283 intraday-doubling setups fired with all 283 reaching completion. Those are not win-rate predictions for your entry, they are counts of how often the pattern, once triggered, ran to its measured target.

Activity this week sat slightly below baseline. The scanner logged 188 patterns over the past 7 days against a 90-day weekly average of 202.8, roughly 7% under normal. Inside that: 21 big-volume gainers (100M+ shares traded), 115 liquidity tests, and 52 stocks with 100%+ gains. A below-average week with 52 triple-digit runners tells you the tape is still runner-heavy even on a quiet read.

Context from the calendar helps. The last four Tuesdays averaged a top gain of 273.1%, and last week (Aug 18-24) produced 33 runners of 50%+, 10 of 100%+, and 3 of 200%+, against a four-week baseline of about 7.5 runners of 50%+ per week. This is an elevated environment, not a normal one.

Last Week's Runners: The Five That Led

Five names carried the trailing five-session tape, and each one maps to a sector that was rotating in. The table below is the whole thesis in one frame: same pattern family, wildly different volume under the move.

Two of the five sit in Finance, the sector whose average RVOL jumped from 1.26 to 3.39 week-over-week, a +169% rotation. ZSTK sits in Wholesale-Non-Durable, the single hottest rotation on the board: RVOL went from 0.80 to 16.76, a +1,986% surge. When a sector's relative volume moves that hard, the runners inside it are not random.

Only ZSTK carried a verified press catalyst chain. ZeroStack announced completion of its continuance to Texas (Aug 18), a US$1.0 billion strategic contribution of Memecore tokens at US$25.19 per share (Aug 19), and cryptocurrency holdings with an aggregate market value of approximately $1.06 billion, or about $18.19 per partially diluted share (Aug 24). For RFAI, BTCT, DAIC, and CYPH, the specific catalyst was not identified in available press releases, which is itself a lesson: plenty of these moves run on structure and rotation, not on a headline you can front-run.

On a $10,000 base, the five-session close-to-close moves translate cleanly. RFAI's +365.8% returned $36,580 in gains. BTCT's +285.6% returned $28,560. DAIC's +216.7% returned $21,670, ZSTK's +165.3% returned $16,530, and CYPH's +135.9% returned $13,590. Those are the idealized full-window numbers. The real trade is always smaller, and the next section shows why.

The Winner That Punished Late Buyers

ZSTK is the honest example in this group, because the day-trade math and the swing math point in opposite directions. On Aug 19, ZSTK opened the regular session at $2.29, ran to a high of $12.40, and closed at $4.88. That is a true MFE of +715.8% from the intraday low of $1.52 to the high of $12.40, the widest window of any featured name.

A $10,000 position that captured the full +715.8% MFE would have returned $71,580 in gains. Nobody catches the exact low and exact high. But look at what happened to anyone who bought the breakout strength: the stock printed $12.40 and closed the regular session at $4.88, giving back the majority of the move in the same day. A trader who chased $12 and held into the close was down roughly 60% from entry on a stock that closed the day green. The MFE was real and the drawdown from the high was brutal at the same time.

That is the core risk of these setups. The scanner's 100% follow-through count measures whether the pattern reached its target, not whether you kept the gain. A stock can complete its measured move to the upside and still hand back most of it before the bell. This is exactly the dynamic covered in Trading the Probe: How Market Makers Test Small-Cap Liquidity, where the sweep to a level is the tell, not the entry.

The Volume Signature: Thin-Float Grinder vs Blowoff

The most useful pattern in 90 days of data is not the price move, it is the volume under it. RFAI and BTCT posted similar percentage gains through completely different mechanics, and reading which is which is how you size the trade.

RFAI ran +365.8% over five sessions on a total of 3,526,351 shares, with a single-day max of 2,972,531. That is a thin-turnover grind: a low-float Finance name walking higher on modest participation, where every buyer matters and the moves come in stair-steps. BTCT ran +285.6% on 498,824,854 total shares, with a single day of 185,517,582. That is a high-turnover blowoff: massive float rotation, violent two-way action, and the kind of tape where a 100M+ share day is the signature of the move rather than a footnote.

The trade is different for each. A thin grinder like RFAI gaps and holds because there is no supply to absorb; the danger is a single large seller that cracks the whole structure. A high-turnover name like BTCT gives you liquidity to enter and exit, but the same liquidity means the top can arrive in minutes. DAIC (106,448,740 total shares, single-day max 102,210,313) and CYPH (81,373,636 total shares) sit in the high-turnover camp with BTCT. ZSTK (26,894,261 total) sits between the two. If you cannot tell which archetype you are in before you click buy, you are guessing at your stop.

Entry, Exit, and Where the Risk Lives

The framework for these setups is the same regardless of ticker: define the trigger, define the invalidation, and respect the fade. Here is how the data supports each leg.

Trigger. The cleanest entries in this data followed a liquidity test, where price sweeps a level to probe supply, then reclaims it on rising volume. The scanner logged 115 of those liquidity tests over the past 7 days. The reclaim of the swept level, not the initial spike, is the structural trigger.

Volume confirmation. For the high-turnover archetype, a session crossing into 100M+ shares is the confirmation the move has real participation; 21 of those big-volume gainers fired this week. For the thin-float grind, you want to see relative volume expanding against a small float, the way ZSTK's Wholesale-Non-Durable sector expanded +1,986% RVOL week-over-week.

Exit and invalidation. The ZSTK Aug 19 tape defines the risk: intraday high $12.40, regular-session close $4.88. A mechanical exit discipline (trailing the move, banking into strength rather than holding for the absolute high) is what turns a +715.8% MFE into a captured gain instead of a round-trip. The invalidation is a loss of the reclaimed level on volume. If the level that triggered the entry fails, the thesis is done.

The pattern completing to target and your account keeping the gain are two different events. Plan the exit before the entry.

How to Set Up This Scan in SNACS

To find these setups before they run, chain three filters and one alert. Here is the exact build.

Open the SNACS scanner and set RVOL to a high minimum to surface unusual participation, price in the $0.50 to $20 range where these small-caps live, and float low to isolate the thin-turnover grinders. Sort by RVOL descending so the sector rotation shows itself at the top; last week that would have put the Wholesale-Non-Durable and Finance names in your face before the moves matured. Save the filter combination as a named preset with a color so you can reload it every morning.

Then link that saved scan to a Dynamic Watchlist, the scan-to-watchlist auto-sync feature. Matched tickers auto-populate in real time and show a colored square in the main stream, so you are watching a live-updating shortlist instead of re-running the scan by hand. Click any ticker to open the ticker details page, where the dilution risk panel (active shelf, ATM, and warrant facilities), recent news, and SEC filings all render without leaving the scanner. That is where you confirm whether a runner is sitting on an active shelf before you commit size.

For the pattern itself, the AI Playbook Builder lets you describe the liquidity-test-and-reclaim in plain English and get working detection logic, then a star indicator appears on any scanner ticker that matches it live. If you want the mechanics of that, describe a setup and get detection logic walks through it. And when you want to check dilution overhang across the board, the SEC research dilution snapshot gives you active facility counts and the lowest exercise price on the shelf.

On the dilution backdrop: the active universe currently carries roughly ~5,900 active warrant facilities, ~3,100 active shelves, ~2,100 active ATM programs, ~1,400 convertible notes, ~900 convertible preferred facilities, ~700 S-1 offerings, and ~500 equity lines. That standing overhang is why the fade risk on these runners is structural, not accidental.

This Week So Far: Filings and Rotation (Aug 24-25)

The filing tape backs up the dilution picture. In the past 3 days, 8 companies filed 424B5 pricing supplements and 5 fresh S-3 shelf registrations hit, alongside 4 S-1 filings and 105 8-K filings across 97 unique tickers. Those are the events that create the overhang a runner sells into. When a name runs 200% and then a 424B5 prices, the fade is the offering, not a mystery.

Insider activity clustered too: SIG logged 17 Form 4 filings in 3 days, RMAX 15, KTCC 14, and both SOTK and CZNC 11 each. Form 4 clusters are the accumulation-or-distribution tell, and they read differently depending on transaction direction, which is exactly what the ticker details page surfaces.

The macro backdrop is mixed and the macro call is indeterminate. The Russell 2000 (IWM), the small-cap tell, closed at $297.97, within 5% of its 52-week high of $305.18, down -2.0% over 5 days and up +1.7% over 20. The Nasdaq 100 (QQQ) closed at $706.32, -5.7% from its 52-week high of $748.65, down -3.2% over 5 days. The Dow Jones Industrial (DIA) closed at $533.65, -2.4% from its high, roughly flat at -0.1% over 5 days. IWM holding near its high while QQQ pulls back is the read that keeps small-cap runners alive even when the indices wobble.

For the fuller weekly context, 22 Runners in Five Sessions: The August 21 Tape and the Wholesale-Non-Durable rotation read both cover the rotation ZSTK rode.

What to Watch Next

The setup for the coming sessions is a runner-heavy tape (33 runners of 50%+ last week versus a ~7.5 baseline) sitting on a heavy dilution overhang and a small-cap index within 5% of its high. That combination favors continuation in the leading sectors (Finance and Wholesale-Non-Durable led the RVOL rotation) but demands tight exit discipline because the fade risk is structural. Watch whether the Finance rotation that carried RFAI and BTCT extends, and watch the 424B5 and S-3 tape for the specific names that just ran, because an offering into strength is the most reliable fade signal these patterns produce. Track your own execution against the setups in the trading journal, where AI Insights flags your MFE capture rate and your worst time-of-day so you learn which of these archetypes you actually trade well.

FAQ

What patterns show up most in penny stock scanner data?

The three most repeatable small-cap patterns are stocks trading 100M+ shares intraday, stocks doubling from session low to session high, and liquidity tests where market makers sweep a price level to probe supply before the real move. Over the past 7 days the scanner logged 21 big-volume gainers, 52 stocks with 100%+ gains, and 115 liquidity tests.

What is MFE and why does it matter for day trading?

MFE (max favorable excursion) is the best possible trade from a session's low to its high across all trading sessions. It matters because a stock can close red or well off its high while still having offered a large intraday window. ZSTK on Aug 19 posted a +715.8% MFE from $1.52 to $12.40 yet closed the regular session at $4.88, so the MFE and the drawdown from the high were both large on the same day.

Why did ZSTK give back most of its gain intraday?

ZSTK opened at $2.29, ran to $12.40, and closed the regular session at $4.88 on Aug 19, so a buyer who chased the $12 high was down roughly 60% from entry on a day the stock still closed green. High-turnover small-caps complete their measured move to target and then hand back the gain quickly, which is why exit discipline matters more than the entry.

How do I tell a thin-float grinder from a high-turnover blowoff?

Compare total volume against the move. RFAI ran +365.8% on 3,526,351 total shares over five sessions, a thin-turnover grind, while BTCT ran +285.6% on 498,824,854 total shares, a high-turnover blowoff. The grinder has no supply to absorb sellers and moves in stair-steps; the blowoff gives you liquidity to trade but can top in minutes.

How do I set up a scanner to find these setups?

In the SNACS scanner, set a high RVOL minimum, price $0.50 to $20, and a low float filter, then sort by RVOL descending to surface the sector rotation. Save the combination as a named preset and link it to a Dynamic Watchlist so matches auto-populate in real time, then click any ticker to open the ticker details page for dilution facilities, news, and SEC filings.

Does a 100% follow-through rate mean these trades always win?

No. The 100% follow-through count measures how often a pattern, once triggered, reached its measured target, not whether your specific entry keeps the gain. 131 high-volume breakout setups and 283 intraday-doubling setups all hit target over 30 days, but the ZSTK fade from $12.40 to $4.88 shows a completed pattern can still round-trip your position.

How does dilution affect these penny stock runners?

Dilution creates the structural fade risk. The active universe carries roughly ~5,900 warrant facilities, ~3,100 shelves, and ~2,100 ATM programs, and in the past 3 days 8 companies filed 424B5 pricing supplements and 5 filed fresh S-3 shelves. When a stock runs hard and then prices an offering into that strength, the offering is usually the fade.

Is this week's pattern activity above or below normal?

Below normal. The scanner detected 188 patterns over the past 7 days versus a 90-day weekly average of 202.8, about 7% under baseline. Even so, last week produced 33 runners of 50%+ against a four-week baseline near 7.5 per week, so the environment remains runner-heavy.