Reviews

CAN SLIM: The Seven Criteria, and Which Ones a Screener Can Test

Disclosure: some links in this article are affiliate links, if you subscribe through one we earn a commission, at no extra cost to you. It does not change what we recommend, and we say where a product is the wrong choice.

CAN SLIM is the best known growth stock checklist in the world, and it is unusual among trading methods in being an acronym that is also a specification: seven named conditions, most of them with a number attached. William J. O'Neil set it out in How to Make Money in Stocks (1988) after studying the characteristics of the biggest winning stocks of previous decades, and the framework has been rebuilt in more screeners than any other.

This guide takes it letter by letter, gives the thresholds O'Neil published, and is straight about which criteria a scanner can evaluate and which two need something a scanner does not have.

A note on integrity before we go further: InvestingPaths is an independent research site and is not affiliated with, endorsed by, or connected to William O'Neil + Co, Investor's Business Daily, or IBD. CAN SLIM® is their registered trademark. Everything below is our summary of publicly documented criteria, attributed as such. Trading involves substantial risk of loss, and nothing here is financial advice or a promise of any result.

The seven letters

LetterStands forThe published threshold
CCurrent quarterly earningsEPS up at least 25% year on year in the most recent quarter, and accelerating
AAnnual earnings growthAnnual EPS growth of at least 25% over the last three to five years
NNew product, service, management, or price highSomething genuinely new, plus the stock breaking out to a new high
SSupply and demandA smaller share count is preferred; volume expansion on the breakout
LLeader or laggardRelative Strength Rating of 80 or higher, that is, in the top 20% of all stocks over twelve months
IInstitutional sponsorshipOwned by a growing number of institutions, with quality of ownership counting more than quantity
MMarket directionTrade only in a confirmed market uptrend, and stand aside when it is not

The 25% figures are the two most often quoted, and they are the ones most often loosened. It is worth being clear about the direction of that: a screen run at 15% is a different screen, and it will hand you a much longer and much weaker list.

What the letters are really doing

Read the acronym as three separate jobs rather than seven items.

C, A, N and S are the business. They describe a company whose earnings are already accelerating, for a reason you can name, and whose share count is not diluting the result away. The S is the one people skip, and it is the least intuitive: a company issuing shares is spreading the same earnings over a larger base, which is why buybacks read as a positive here.

L is the market's verdict. O'Neil's position, and it is the one that most annoys value investors, is that you should buy stocks that are already outperforming, not the ones that look cheap. He bought nothing rated below 80 on relative strength.

M is the veto. The last letter overrides the other six. In a market downtrend the correct number of CAN SLIM positions is close to zero, no matter how good the list looks. Most people who report that the system stopped working for them were, on their own account, ignoring M.

Which criteria a screener can actually evaluate

Five of the seven are computable from public data. Two are not, and knowing which is which saves you from buying a tool that promises the whole thing.

CriterionScreenable?Why
C, quarterly EPS growthYesReported figures, though "accelerating" needs several quarters, not one
A, annual EPS growthYesReported figures
N, new productNoRequires reading about the company. The price-high half of N is screenable; the "new" half is not
N, new price highYesA 52-week high condition
S, share countYesShares outstanding, quarter on quarter
S, volume on breakoutYesVolume against its own average
L, RS Rating ≥ 80ApproximatelyThe rating itself is IBD's proprietary percentile. Any other tool is approximating it
I, institutional sponsorshipPartlyHoldings data exists; "quality of ownership" is a judgement
M, market directionYesAn index against its own moving average

The L problem is the same one the Minervini Trend Template runs into with its eighth criterion, and for the same reason: a percentile rank against the entire market is a different object from a return comparison. A stock that beat the S&P 500 by 20% might be in the top 5% of the market or the top 40%, depending on the year. Vendors who advertise an RS Rating outside IBD have computed their own, which is legitimate as long as nobody pretends the numbers are interchangeable.

Building the screen in TC2000

TC2000 does not ship a CAN SLIM screen, and it is the wrong instrument for the fundamental half, but it handles N, S and M cleanly, which is the technical spine:

CriterionCondition
N, new price highBuilt in 52-week high scan, or C >= MAXH252
N, near the highC / MAXH252 >= .95 for stocks consolidating just under one
S, volume expansionV > 1.5 * AVGV50
M, market uptrendAn index chart with a 50 and 200-day moving average, checked before you trade the list
LiquidityV > 100000, plus a minimum price

Custom formula conditions need the Premium tier; the 52-week high scan does not. Our pricing breakdown covers the tiers, and the EasyScan guide covers how to assemble conditions into a saved scan.

For the earnings half you want a fundamental screener rather than a charting platform, which is the split the whole TC2000 versus Deepvue comparison is about: Deepvue precomputes CAN SLIM style ratings, TC2000 gives you formula control over the chart.

Testing the parts that are testable

Most of CAN SLIM is expressible in our free backtester, which is unusual for a named system, because the criteria happen to line up with rules we already have:

CriterionOur rule
C and A, earnings growthEPS growth and EPS CAGR
S, share countShare count change
L, relative strengthRelative strength vs the index, an approximation, see above
N, new highDistance from the 52-week high or low
S, volumeVolume vs its own average
M, market directionS&P 500 vs its moving average, which is M almost exactly

The M rule is the interesting one to test, because it is the criterion people drop and the easiest to isolate: build the screen, run it, then run the identical screen with the market filter switched off and compare. That is a question with a real answer, and it takes about a minute.

Two limits to state before you trust any figure that comes back. The N narrative and the I sponsorship quality are not in the test at all, so what you are testing is a CAN SLIM shaped filter rather than CAN SLIM. And our fundamental data is current rather than as-reported at the time, so a fundamental screen answers "what does this look like today", while the backtest's chart rules run bar by bar. Where we do have point-in-time filings, the value screener uses them; it is a different instrument aimed at cheapness rather than growth.

An honest look at the limits

  • The thresholds are from a study of past winners. Selecting the biggest winners of previous decades and then describing what they had in common is a survivorship-shaped exercise. It tells you what winners looked like; it does not, on its own, tell you how many losers looked the same.
  • It is a bull market system by construction. The M criterion admits this openly. A method that stands aside in downtrends will spend long periods doing nothing, and that is the design working, not failing.
  • Two of seven need a subscription. The RS Rating and the sponsorship data are IBD products. You can approximate both, and you should know you are approximating.
  • The 25% thresholds get quietly relaxed. In a market with few qualifiers, the temptation is to lower the bar rather than hold fewer positions. That converts a demanding filter into an ordinary growth screen.

Frequently asked questions

What does CAN SLIM stand for? Current quarterly earnings, Annual earnings growth, New product or price high, Supply and demand, Leader or laggard, Institutional sponsorship, Market direction.

What are the CAN SLIM criteria numbers? The two published thresholds most often quoted are 25% growth in current quarterly EPS year on year, 25% annual EPS growth over three to five years, and a Relative Strength Rating of at least 80 for the L criterion.

Can I run a CAN SLIM screener for free? You can screen most of it for free. The earnings, share count, 52-week high and market direction conditions are all available in our backtester with no account. The IBD Relative Strength Rating specifically is proprietary and is approximated elsewhere rather than reproduced.

Is CAN SLIM fundamental or technical? Both, deliberately. Four letters describe the business and three describe the price and the market. Running only the fundamental half is a growth screen; running only the technical half is a breakout screen. O'Neil's argument is that the combination is what mattered.

How is CAN SLIM different from the Minervini Trend Template? The Trend Template is purely technical, eight pass or fail chart conditions, and is a filter that runs before any fundamental work. CAN SLIM builds the fundamental requirements into the checklist itself. They share the relative strength idea, and both hit the same wall at IBD's proprietary rating.

Build the screenable part and see what it leaves

Most of the checklist is testable today. The backtester covers the earnings, share count, 52-week high, volume and market direction conditions, free and with no account, and switching the market filter on and off is the cheapest experiment in this whole article.

Test it before you trust itEvery rule in this article can be backtested on real daily prices in seconds, free, no signup.Open the backtester

Backtest figures in our articles are computed by our own engine over real price data: fees and slippage included, shown against buy-and-hold, and live embeds refresh as new data lands.

One useful email a weekWhat our backtests, screens and congress feed surfaced this week, no hype, one click to unsubscribe, nothing else sent.