A 90% up day is bullish & the stats support a year-end rally
click for giant table
90 percent up
Source Merrill Lynch BA



On Thursday. we had a huge up day, with US markets gaining ~2%. Some folks credited the possibility of a debt ceiling deal, while others called it a low volume short covering rally. Regardless, it was substantial, and should not be ignored.

As the table above showed, these 90/90% up days– when 90% of stocks on the NYSE are up and 90% of the shares traded are to the upside — the tendency is towards higher future returns in the ensuing weeks and months.

As you can see in the table above, 20, 30 and even 65 days after a 90/90 day, markets are higher at least 70% of the time. Under normal circumstances, that range would be low 60% range. In other words, the odds of gains improves some 10% over normal following a 90/90 day in our time period (20, 30 or 65 days from 2007-present) .

Here’s Merrill’s Stephen Suttmeier:

“That big buying the US equity market opened strong with a 90% up day (at least 90% of stocks on the NYSE up on at least 90% up volume) and the equity market maintained this 90% up day into the close.

This is bullish and was the first 90% up day since the pair of 90% up days from 12/31/12 and 1/2/13. We have data on 90% days going back to January 2006. Since then 90% up days have occurred only 3.3% of the time, but after a 90% up day, the S&P 500 has well above average 10, 20, 30, and 65 day returns and this supports the case for a year-end rally. For example, the 65-day S&P 500 return after a 90% up day is 4.9% vs. an average 65-day return of 1.4%. Interestingly, the 1, 2, and 5-day returns for a 90% up day are below average and negative. This suggests buying into a short-term dip after a 90% up day.”


That 14% edge over 65 days isn’t a sure thing, but it is quite statistically significant . . .

A 90% up day is bullish & the stats support a year-end rally
Merrill Lynch Bank America, October 11 2013
Stephen Suttmeier

Category: Markets, Technical Analysis

Please use the comments to demonstrate your own ignorance, unfamiliarity with empirical data and lack of respect for scientific knowledge. Be sure to create straw men and argue against things I have neither said nor implied. If you could repeat previously discredited memes or steer the conversation into irrelevant, off topic discussions, it would be appreciated. Lastly, kindly forgo all civility in your discourse . . . you are, after all, anonymous.

13 Responses to “90% Up Days Are Bullish”

  1. Tacomaman says:

    Thanks for the heads up. I have not made too many moves for fear of being whipsawed.

    By the way, what happened the the “ginormous” charts reference? It’s sounds much better than giant.

    Thanks for subscribing me to your newsletter.

  2. Concerned Neighbour says:

    Of course this small sample is heavily biased by central bank intervention in the “markets”. I would be curious how many of these non-stop up days occurred on average prior to this period of “temporary” stock market stimulus.

    Not that I disagree, mind you. It would not surprise me if central banks never let stock markets go down again. Price discovery is quite dead and buried.

      • Concerned Neighbour says:

        I said the study is heavily biased, not totally biased. When more than half your sample is biased by pretty transparent and overwhelming market manipulation, the value of said sample is diminished. I’ve been arguing on this and other sites for a while that the value of this kind of historical comparative analysis is considerably less than it otherwise would be due to the extraordinary manipulative forces at play in asset markets.

        I do apologize for any perceived snarkiness. It does at times get the better of me, especially on the topic of QE.

  3. BenGraham says:

    “We have data on 90% days going back to January 2006.”

    Seriously? 7 years of observations is considered tradeable “data”? I call it anecdotal garbage. For something useful- get data back several decades and then divide into two groups- those following sell-offs and those following rallies (however defined). I bet the outcomes are different.

  4. RW says:

    Oh I’m bullish all right …but I still expect a major buying opportunity coming soon to a venue near y’alls.

    Chaos Ensues As Conservatives Tank House GOP’s New Debt Plan

  5. peterkrause says:

    Martin Zweig in 1986 published “Winning on Wall Street.” He mentioned the significance of 90% up days so for those of you looking for beefier data sets, there you go. BR, if you’re familiar, can you comment on whether you agree the signal needs a followup 90% day to be really actionable? I would appreciate your comment.

  6. GregP says:

    “it is quite statistically significant ”

    Given the data and analysis, there is no way to tell.
    That’s b/c there may be a high level of covariance among 90% days.
    A group of closely spaced 90% days (2009?) has high covariance, and the measures of subsequent performance are not independent.

  7. clipb says:

    a question that seems to be mostly unanswered is this: to what extent is the market (here and elsewhere)
    rallies dependent on the fed and other central bank”easing” in various ways? and should a relatively restrictive data set be evaluated on other criteria/metrics that don’t incorporate this stuff? and will the central bank “liquidity” program continue ad infinitum? and if they don’t?….

  8. clipb says:

    narrow database excerpts= data mining. what about the central bank”easing’ programs? will they continue forever? will the fed keep subsidizing low rates ad infinitum? 5 trn? 7 in a couple of years? and what will/might be the consequences of termination, let alone liquidation? is this gemaine to 90% extrapolations? get real!