We replayed every quarter from 2014 to mid-2026 for every company that met our coverage rules at the time: about 4,200 over those years, including companies that have since been acquired, delisted or grown too small. Today we cover about 2,200. Each quarter was scored using only the information available at that time, the way the site scores companies today. Then we checked what actually happened next.
For stock prices we used the closing price on the last trading day of each quarter, the same way index and fund returns are reported. Each stock was compared with SPY, the fund that tracks the S&P 500, at its closing price on the same days. Prices are adjusted for stock splits but don't include dividends, for the stocks and SPY alike. Results after publication start from the first quarter-end close after the results were filed.
As a group, stocks of companies whose business improved sharply outperformed the S&P 500 in about 4 of every 5 quarters. Stocks of companies whose business weakened sharply trailed the S&P 500 in 5 of every 6 quarters. Both held in every year since 2016.
How the stock did against the S&P 500 while a company was in each phase, by where that phase led:
Measured from the quarter before each phase began to its last quarter, against SPY over the same dates. A year counts when the average stock in that step did better than SPY that year. Where a phase leads is only known once it ends, so these describe the past.
A Strong reading means the company's numbers agree with each other, the move has lasted more than a quarter or two, and the phase has held for a while. Readings like that tend to keep going for about three quarters. After one quarter, 9 in 10 were still heading the same way, still improving or still weakening. After two quarters, 3 in 4. After three, about half. After a full year, 1 in 3, because by then most moves had run their course.
Weak readings were a coin toss: after two quarters, about half were still heading the same way. Strong readings held up better than Weak ones in every year since 2014.
A company in Recovery went on to Expansion about twice as often as it slipped back into Contraction. A phase lasts about three quarters on average. Recovery and Topping tend to be shorter, and Expansion and Contraction longer.
When institutions were adding to a company's shares, its business improved the following quarter 53% of the time. When they were cutting back, 44%. That held in 11 of the last 12 years.
We start with every institution that files Form 13F, 10,525 of them, and narrow it down:
Tested on the filings that came after. Patterns found in earlier years pointed the same way over the next two years 72% of the time. Chance would give 54%.
Measured against their own usual rate. Plenty of institutions simply buy a lot. A pattern here means an institution behaves differently in one phase than it does overall.
Direction, not degree. The tests support which way an institution leans in a phase, not by how much, so that is all we show.
A move doesn't stop when the results come out. While a company was in Recovery on its way to Expansion, its stock kept outperforming the S&P 500 in the later quarters of that Recovery, not just the first. And stocks with a Strong reading in Recovery or Expansion, from the quarter after it was published, did better than the average company we cover in 7 of 9 years, and better than the S&P 500 in 4 of 9.
These are averages across thousands of companies. Any one company can behave differently, and a phase can end sooner than usual. CycleIQ shows where a business stands and how solid that reading is. It doesn't tell you to buy or sell. Past performance is no guarantee of future results.
Tested October 2026 on filings from 2014 to mid-2026. Quarter-end closing prices, split-adjusted, without dividends. We'll re-run these tests every year and publish the results here, including anything that stops holding.