Squeeze Performance Tracker

What the Squeeze Scanner's alerts actually did afterwards, posted after every session.

Posts in #squeeze-reports

What it is

The Squeeze Scanner reporting back on itself. Once a trading day settles, the channel posts what every alerted ticker went on to do — how many reached +5%, +10%, +20% and +50%, the typical run, the typical drawdown against you, and how long the moves took — split by whether the tape was fully covered or the alert landed in extended hours. Weekly roll-ups follow the same shape. Losses are reported as prominently as gains, and every figure carries the sample size behind it.

What it looks like

Squeeze Scanner — Thu 17 Sep Example

Alerts

Total: 3
Complete: 3
Insufficient: 0

Coverage

Full: 1
Premarket: 2
After-hours: 0

Data quality

Insufficient: 0
Incomplete hist: 0

Full-coverage follow-through · n=1

Reached: +5% 100% (1/1) · +10% 0% (0/1) · +20% 0% (0/1) · +50% 0% (0/1)
Max run: median +5.3% · p25 +5.3% · p75 +5.3% · avg +5.3%
Adverse: median −0.8% · p25 −0.8% · p75 −0.8% · peak→trough −2.1%
Timing: →+5% 1777m · →+10% — · →peak 1777m
Peak: same-session 0% (0/1) · next 100% (1/1)
Shapes: immediate 0 · delayed 0 · dump→run 0 · no-follow 1

Top

$XHLD: +5.3%

Weakest

$XHLD: +5.3%

Premarket · n=2

Extended: scanner-observed (n=2) · insufficient sample
Regular follow (n=2): open→high +0.5% · +10% from open 0% (0/2) · open→close −0.9%

Performance reflects observable post-alert market activity. Extended-hours scanner observations are discrete and may not capture every price move.

The weekly roll-up, the same shape over a full week

Squeeze Scanner — Week 2026-W35 Example

Alerts

Total: 89
Complete: 72
Insufficient: 17

Coverage

Full: 77
Premarket: 0
After-hours: 12

Data quality

Insufficient: 17
Incomplete hist: 59

Full-coverage follow-through · n=63 (+14 insufficient)

Reached: +5% 35% (22/63) · +10% 21% (13/63) · +20% 14% (9/63) · +50% 10% (6/63)
Max run: median +3.5% · p25 +0.6% · p75 +7.2% · avg +55.5%
Adverse: median −5.4% · p25 −9.7% · p75 −1.8% · peak→trough −7.7%
Timing: →+5% 194.5m · →+10% 196m · →peak 288m
Peak: same-session 56% (35/63) · next 44% (28/63)
Shapes: immediate 3 · delayed 9 · dump→run 1 · no-follow 50

Top

$VBIO: +2713.7%
$WHLR: +275.7%
$BTCT: +71.9%

Weakest

$RETO: −11.5%
$KXIN: −5.5%
$WLDS: −3.5%

After-hours · n=12

Extended: scanner-observed (n=0) · insufficient sample
Regular follow (n=9): open→high +1.5% · +10% from open 11% (1/9) · open→close +0.0%

Broken down by what the alert looked like

squeezeFactor

70–100: +10% 21% (11/52) · med +3.3%
40–69: +10% 11% (1/9) · med +3.8%

borrowFeePct

50–150%: +10% 16% (4/25) · med +3.3%
≥400%: +10% 28% (5/18) · med +5.0%
150–400%: +10% 20% (4/20) · med +2.4%

shortPct

<20%: +10% 9% (1/11) · med +1.6%
≥50%: +10% 21% (6/28) · med +3.3%
20–50%: +10% 25% (6/24) · med +4.2%

floatShares

1–5M: +10% 33% (4/12) · med +4.3%
<1M: +10% 19% (8/43) · med +3.2%
≥5M: +10% 13% (1/8) · med +4.4%

mktCap

≥20M: +10% 30% (6/20) · med +3.6%
5–20M: +10% 0% (0/20) · med +3.2%
<5M: +10% 30% (7/23) · med +3.8%

relVol

<1: +10% 19% (11/59) · med +3.4%

price

$3–10: +10% 29% (5/17) · med +2.7%
<$1: +10% 24% (4/17) · med +3.6%
$1–3: +10% 17% (4/24) · med +3.6%

timeOfDay

regular: +10% 21% (13/63) · med +3.5%

direction

up: +10% 20% (12/59) · med +3.3%

confirmations

isolated (0): +10% 6% (1/16) · med +3.1%

Performance reflects observable post-alert market activity. Extended-hours scanner observations are discrete and may not capture every price move.

Rebuilt in TickerRoom's own components from past output, to show the format. Illustrative, not live market data.

What each part shows

Alerts, coverage and data quality: How many alerts the period produced, how many had a complete enough tape behind them to measure, and how many did not. The report leads with what it could not measure rather than burying it.
Reached: The share of measured alerts that touched +5%, +10%, +20% and +50% at any point afterwards, each with the count behind it. Touching a level is not the same as closing there.
Max run: The best move after the alert — median, quartiles and mean. Read the median. When the mean sits far above it, one or two extreme alerts are carrying the average and the typical alert did nothing like that.
Adverse: How far price went against you before it went anywhere, on the same scale. This is the line that says what holding the alert would have felt like, and it is printed as prominently as the upside.
Timing: How long moves took to reach each level, and to reach their peak, in minutes. A level reached after 1777 minutes is a different proposition from one reached in twenty.
Peak and shapes: Whether the high landed in the same session or the next, and how the moves behaved — immediate, delayed, dump then run, or no follow-through at all. The no-follow count is usually the largest of the four.
Top and weakest: The best and worst alerts of the period, both shown. The weakest column exists so the top column cannot be read on its own.
Extended-hours split: Premarket and after-hours alerts are reported separately, because the tape there is thinner and the scanner observes it in discrete samples. Where the sample is too small to say anything, it says so instead of printing a number.
Factor breakdowns: On the weekly report, the same follow-through figures cut by squeeze factor, borrow fee, short percent, float, market cap, relative volume, price, time of day, direction and confirmations — so you can see which kind of squeeze alert the week actually rewarded.

What it is for

It tells you what a scanner is worth rather than asking you to take it on trust. The same figures get posted whether the week went well or badly.
It shows the downside beside the upside. Adverse move, peak-to-trough and the no-follow count sit in the same report as the hit rates.
It puts a number on what normal looks like for this scanner, so an alert running far past the median is visibly unusual rather than merely exciting.
The factor breakdowns show which conditions the follow-through clustered in over the week — borrow-fee bands, float sizes, price ranges.