One Squeeze Scanner for Over 600 Stocks and 10 Timeframes

Key takeaways

• Flagging a squeeze is trivial. Almost any tool can print a list of names that are compressed right now, and that list on its own is close to noise.
• A bare squeeze carried no tradeable edge in our 22-year backtest. The edge came from the setup behind it: momentum, trend, and a positive per-symbol record.
• A scanner earns its keep when it ranks rather than just flags, sorting over 600 stocks across 10 timeframes by the three things that actually decided the trade.
• Expectancy, not win rate, is the number to sort by. The short Slingshot won 60.3% of its trades and still lost money, so a high hit rate alone is not an edge.

Open any charting package, set a scan for stocks where the Bollinger Bands have slipped inside the Keltner Channels, and you get a list of names in a squeeze. It feels like the hard part is done. It is not. That list tells you which stocks are compressed and nothing about which compression is worth trading. In our own historical backtest a bare squeeze, taken on its own, was close to a coin flip. The work a real squeeze scanner has to do is the part that list skips: rank the names by whether the setup behind the squeeze has actually paid.

Why a list of squeezes is not enough

A squeeze is a measurement of volatility compression. It says a stock has gone quiet and is likely to expand. It does not say which way, and it does not say whether this particular name rewards you when the expansion comes. We tested that directly, and the finding was blunt: a bare squeeze has no edge on its own. Buying compression because it is compressed, with nothing else behind it, produced no reliable gain across thousands of trades and 22 years.

So a scanner that stops at “here are the stocks in a squeeze” hands you the least useful version of the information. You still have to answer, name by name, whether the squeeze is attached to a setup with a record. Do that by hand across a few hundred charts and the morning is gone before you place a trade.

What a real squeeze scanner should show

The three things that separated a winning squeeze from a losing one in our data are the three columns worth scanning for. We call them the pillars.

• Edge. The stock’s own 22-year record on the squeeze. Not the market average, this specific symbol. Some names carry a strong multi-year record, others have none at all.
• Confluence. How many timeframes are in a squeeze at the same time. One name compressed on the daily and the weekly together is a different picture than one squeezing on the 5-minute alone.
• Trend. Whether the moving averages are stacked in the direction of the setup. Compression inside a clean trend is not the same trade as compression in the middle of chop.

A flat list of squeezes shows you none of these. A board that reads all three, on every name, in one view is what turns a raw scan into a decision. You can see how we score the three pillars in how to read the Squeeze Board.

Ranking, not just flagging

Once you have the record for each name, the scan becomes a sort. Instead of a flat list you rank by the number that decided the outcome, and the honest number there is expectancy, not win rate.

The short side proves why. The short Slingshot Squeeze won 60.3% of its trades, which sounds like plenty. It still lost money, because the average trade came in near -2.0%: the losses were large enough that winning three times out of five did not cover them. A setup can win more often than it loses and lose money overall. Sort a scanner by win rate and that setup floats to the top. Sort by expectancy and it drops where it belongs.

This is also why the same setup belongs on one stock and not another. The daily Slingshot Squeeze averaged 69.1% across our 614-stock universe over 5,852 trades, and that average held in a tight band in every era since 2004. But it is an average built from names that did much better and names that did much worse. A scanner that ranks by each symbol’s own record lets you take the strong responders and skip the weak ones. We go deeper on that in not every stock squeezes well, and you can pull any symbol’s full history in the backtester.

Every timeframe, from 5-minute to weekly

Most scanners read one timeframe, usually the daily. That throws away most of the picture, because the squeeze behaves differently depending on the clock you read it on. In our backtest the slower timeframes carried the real per-trade economics and the fastest ones did not:

• Daily: +0.93% expectancy per trade
• 4-hour: +0.56%
• 30-minute: +0.51%
• 195-minute: +0.44%
• 5-minute: roughly break-even
• 15-minute: slightly negative

Reading all 10 timeframes at once, from the 5-minute to the weekly, does two things. It keeps you on the clocks that actually paid, and it shows you when several of them line up on the same name. That overlap is the confluence pillar, and we cover what it does and does not do in multi-timeframe squeeze confluence. A single-timeframe scan cannot show it at all.

Live fires, and what is closest to firing

A scanner also has to handle the moment the squeeze releases. On our board that shows up as a green dot and a fired status. One thing to keep straight: the fire is the exit on our setups, not the entry. Both the Slingshot Squeeze and the Squeeze Signal are entered while the squeeze is still on, on the trigger, so the names you care about most are the ones that are still compressed and close to releasing, not the ones that already have.

You cannot name the exact bar a squeeze fires, and no honest scanner claims to. What it can do is sort by proximity: how tight the compression has become and how long it has lasted. That gives you a group of names closest to firing to watch, which is a different job from the already-fired list. We walk through reading proximity without false precision in how to know when a squeeze is about to fire.

Over 600 stocks, every session

The scale is the last piece. Our board scans more than 600 stocks across all 10 timeframes every session, reconstructs the squeeze dots from price, attaches each name’s 22-year record, and ranks the whole universe by edge, confluence, and trend. One board instead of several hundred charts. You can see the full list of names we cover on the supported symbols page and watch the ranking update live on the Squeeze Board.

None of this is a promise about the next trade. Roughly one setup in three still lost even on our best-ranked names, and every number here is our own historical record rather than a forecast. What a ranked scan buys you is not certainty. It is the ability to spend your attention on the handful of names where the squeeze has actually paid and to ignore the rest of the list. Visit our website to see it live, squeezesetups.com.

Squeeze Setups is an educational research tool, not financial advice.

Related: Why a bare squeeze has no edge, Not every stock squeezes well, How to know when a squeeze is about to fire