CANSLIM: What a Chart Screenshot Can (and Can't) Confirm

CANSLIM is one of those systems everyone has heard of and fewer people have actually checked letter by letter. It's seven criteria, and the part the summaries tend to skip is that they come from two completely different places. Some of the letters live in earnings reports and fund filings. The rest live on the chart. I build a tool that reads chart screenshots, so I care a lot about where that line falls, and this post is mostly about drawing it honestly.

Quick Answer

The rest of this covers each letter with the thresholds people actually cite (they don't agree), the 7-8% loss rule, what the breakout looks like on a chart, why a screenshot grader is not a CANSLIM screener, and the uncomfortable question of whether the thing still works.

What Is CANSLIM?

CANSLIM is a stock selection system for growth stocks, created by William O'Neil, the founder of Investor's Business Daily. He laid it out in How to Make Money in Stocks, first published in 1988, after studying the biggest stock market winners and looking for what they had in common before their big runs. The acronym is the answer he came up with: seven traits, checked together.

What makes it different from a pure chart method is the blend. Roughly half of CANSLIM is fundamental analysis (is the company growing earnings fast, and are big funds buying it) and the other half is technical and market timing (is the stock breaking out, is it acting like a leader, is the overall market healthy enough to buy anything). The Wikipedia entry on CAN SLIM notes that the American Association of Individual Investors (AAII) ranked it the top-performing strategy in its tracking over 1998 to 2009, which is a big part of why the name stuck around. Hold that thought for the section on whether it still works, because the track record is more contested than that one stat suggests.

The CANSLIM meaning, in plain terms: find a company with accelerating earnings and something new going on, wait for its stock to break out of a base to new highs on heavy volume, make sure it's a leader and that institutions are piling in, and only do it while the market is in an uptrend. Then cut the loss fast if you are wrong.

The Seven CANSLIM Criteria, Letter by Letter

One thing before the list. The exact numbers vary depending on which summary you read, and I'm not going to pretend there's a single official cutoff for each letter. Where sources disagree, you get the range and the figure that shows up most often.

C: Current quarterly earnings

Earnings per share for the latest quarter compared with the same quarter a year earlier, so seasonality doesn't fool you. The commonly cited threshold is 25% growth or more, and that's the number CFI's CAN SLIM overview and several others use. Other write-ups go as low as 18-20% or talk about 20-50%. Faster is the point.

A: Annual earnings growth

One good quarter can be a fluke, so A checks the longer record: annual EPS growth of around 25% over the last three years is the common version, and some sources want to see earnings rising for five years straight. Several also add return on equity of 17% or more as part of this letter.

N: New product, new management, new high

Something new is driving the business: a product, a service, a change in management, an industry shift. According to Macro Ops' walkthrough of O'Neil's original study, about 95% of the big winners he examined had something new behind them. The other half of N is price: the stock emerging from a base to new highs. O'Neil wanted you buying near highs, which runs against the buy-low instinct most people start with.

S: Supply and demand

Supply is how many shares exist. In O'Neil's study, most of the biggest gainers had fewer than 25 million shares outstanding, because a smaller supply moves further when demand shows up. Demand shows up in volume: heavy volume on up days, especially the breakout day.

L: Leader or laggard

Buy the leading stock in a leading industry, not the cheap laggard you hope catches up. The usual yardstick is a relative strength rating of 80 or higher on a 1-99 scale, meaning the stock outperformed about 80% of the market. Some sources accept 70 as a floor.

I: Institutional sponsorship

Big moves need big buyers, so I asks for a rising number of institutional holders (funds, pensions, and so on). But not excessive ownership, because a stock every fund already holds has few buyers left and a lot of potential sellers. Specific cutoffs vary a lot from source to source.

M: Market direction

Only buy when the general market is in a confirmed uptrend. The reasoning is that roughly three out of four stocks follow the market's direction, so a perfect CANSLIM stock in a falling market is still fighting the tape. This is the letter people ignore most and regret most.

Which CANSLIM Letters Can a Chart Screenshot Actually Show?

Here's the whole checklist in one table, with the question I actually care about in the last two columns. If you upload a stock chart to any AI tool, which of these can it honestly confirm from the pixels?

Three flat noes. C, A and I are out of reach for any chart reader, full stop. Earnings growth and fund ownership are numbers from filings, and no amount of clever pattern recognition pulls them out of candles. If a tool tells you it confirmed a stock's 25% quarterly EPS growth from a chart image, it made that up.

L is the sneaky one. The RS Rating most CANSLIM material refers to is a proprietary IBD number, so it's out of reach too unless your platform happens to print it on the chart. What a chart does show is the thing the rating is trying to capture: is the stock trending up cleanly, holding up while the market wobbles, making new highs while other names don't. That's a visual read of relative strength, and it's useful, but it isn't the number.

That leaves N (the base and the breakout to new highs), S (the volume pattern through the base and on the breakout day, minus the share count part, which is also not on the chart), and M (if you screenshot the index itself). Those are the letters where reading a chart well actually changes the decision, and they're where the rest of this post spends its time.

What Does a CANSLIM Breakout Look Like on the Chart?

The buy point in CANSLIM is a breakout from a base. A base is a stretch of sideways-to-down consolidation, usually several weeks long on a daily chart, where the stock digests a prior run. The shapes O'Neil described include the cup with handle (the famous one, and the cup and handle breakdown on this blog goes through its proportions), the flat base, and a few others. Our broader chart pattern library covers the rest of the family.

The chart-visible checklist for N and S looks like this. The base has a clear ceiling, the pivot, and you draw that line before the breakout, not after. Volume through the base tends to dry up, especially in the handle or the right side of a flat base, which says sellers are running out. Then the breakout day closes above the pivot on volume well above the recent average. That volume surge is S doing its job: demand overwhelming a limited supply.

If you want the mechanics of reading that ceiling, it's the same skill as marking horizontal resistance on any chart. And if you want a second read on whether volume is confirming or quietly disagreeing with price through the base, an accumulation line like on-balance volume is one option people plot under the chart.

The failure modes are all visible too, which is the good news. A breakout on light volume. A base that is loose and sloppy, with wide swings, instead of tight. A stock that already ran well past the pivot before you noticed, so buying it means paying up with nowhere sensible to put a stop. How AI tools separate a real breakout from these lookalikes is covered in more depth in the post on AI breakout detection.

The 7-8% Loss Rule and Reading the M Letter

The single most quoted risk rule in CANSLIM comes straight from O'Neil's book: cut every loss when a stock falls 7% to 8% below your buy price, with no exceptions. No averaging down, no waiting for it to come back. The logic is that a proper breakout from a sound base shouldn't fall that far below the pivot, so if it does, the setup was wrong and you want to find out while the loss is still small.

On a chart, that rule turns into a line. If you bought at the pivot, 7-8% below is your hard ceiling on risk. In practice the chart often gives you a closer, more logical stop (the handle low, the bottom of a tight flat base), and when it does, that's usually the better choice. The 7-8% rule is the maximum, not a target. If the only sensible stop on the chart sits 12% below the entry, that doesn't mean you stretch the rule. It means you wait for a better entry. More on picking stop levels from structure in our guide to AI stop-loss placement.

M lives on a different chart

M is the one letter that has nothing to do with your stock's chart. It's a read of the index. The Wikipedia summary describes O'Neil's three market states as a confirmed uptrend, an uptrend under pressure, and a market in correction, and the instruction is to be buying only in the first. You can read that off a chart of the S&P 500 or Nasdaq: is price above rising averages, are the lows getting higher, is the index getting hit with heavy volume on down days.

The practical point: if you want M covered in a screenshot workflow, you have to screenshot the index separately. A grade on your stock's chart is a grade on your stock's chart. It says nothing about the market it's trading in, and a great setup in a correcting market is still a coin flip with worse odds. Stan Weinstein's four-stage framework is a decent way to classify the index trend if you want something more structured than eyeballing it.

Is a Chart Grader a CANSLIM Screener?

No, and it's worth being clear about because "canslim screener" is one of the most searched phrases around this topic. A CANSLIM stock screener runs through thousands of tickers and filters them on data: quarterly EPS growth above X, annual growth above Y, RS rating above 80, institutional holders rising, price near a 52-week high. It needs a fundamentals database to do that. That's how you go from the whole market to a short list of CANSLIM stocks.

A screenshot grader does something different and narrower. You upload one chart. It reads that one image. It does not scan the market, it does not monitor your watchlist, it does not pull earnings or ownership, and it does not output a CANSLIM score. What it can do is look at the chart you already picked and tell you whether the base and the breakout are any good: trend, structure, volume behavior, where the levels are, and what a sensible entry and stop look like. That's the scope described on the AI chart analysis page, and the general idea of how AI reads a chart image is laid out in the AI trading overview.

So the realistic workflow has two halves. The screener (or your own digging through earnings reports) handles C, A, I and the RS Rating. The chart read handles N, S and M. Neither replaces the other, and anyone telling you one tool does all seven from a picture is overselling.

• Fundamentals first, chart second
  Run C, A and I through a screener or the filings before you look at the chart. A perfect base on a company with shrinking earnings is not a CANSLIM setup, no matter how clean it looks.
• Draw the pivot before the breakout
  The ceiling of the base has to exist before the breakout bar does. A line drawn around a bar you already like will always confirm.
• Volume has to show up on the breakout day
  Judge it against the recent average on the same chart. A breakout on below-average volume is the most common way a nice-looking base fails.
• Check the index on its own chart
  M is not visible on your stock's chart. Screenshot the index separately and be honest about whether it's in an uptrend.
• 7-8% is the maximum loss, not the stop
  Use the closest structural level the chart gives you. If that level is further than 7-8% away, the entry is the problem.

Does CANSLIM Still Work?

Depends who you ask, and the evidence is messier than the fan write-ups admit. On the positive side, there's the AAII ranking over 1998 to 2009, and at least one 2013 academic study found outperformance. IBD has promoted a 20.7% average annual return figure for the method, but both WallStreetMojo and Liberated Stock Trader's independent test of CANSLIM point out that figure is unaudited. The same review notes that funds and ETFs built around the approach have lagged the S&P 500, and the Wikipedia entry mentions that O'Neil's own funds struggled at times.

Then there's the strictness problem. Portfolio123 ran a strict CANSLIM screen and got just 58 passing stocks across 15 years, with an average 6-month return of about -6%. That's one quant implementation, not the final word, but it shows how much the result depends on how literally you apply the rules and how much judgment you layer on top.

The limitations that show up across most honest reviews:

My read, as someone who mostly trades much shorter timeframes: the strict seven-letter screen is less useful as a black-box system than as a way of thinking. Trade leaders, not laggards. Buy strength breaking out of a quiet base, not weakness you hope turns around. Respect the market direction. Cut losers fast. Those ideas show up in almost every serious momentum trading strategy, just compressed onto an intraday clock. And if you're coming at this from the Minervini side, his Trend Template and SEPA rules are the close cousin worth comparing (see the FAQ below for how they differ).