Do Three Down Days Set Up a Bounce?

Some traders call three red days in a row the sweet spot for a reversal.

We tested whether the edge actually exists, using 55 names.

On an ordinary day, these names close green about 52% of the time. After three consecutive down days, that rate rises to roughly 54%. The improvement is small. 43 of the 55 names showed a higher probability of closing green on the fourth day.

The more interesting finding is how differently the names behave. When sorted by the strength of this effect, the list splits into two distinct groups with very different characteristics.

How much three down days changes the fourth-day bounce odds, in percentage points. Green names bounce more often than usual, red names bounce less.

It works best on the steady giants and the index

The bounce shows up most clearly where you might least expect drama: the biggest, most widely held names, and the index itself.

Name	Green on any day	Green after 3 down days	Edge
S&P 500	55%	59%	+4pp
AMZN	52%	59%	+7pp
META	52%	59%	+7pp
GOOGL	52%	58%	+6pp
MCD	53%	59%	+6pp
AAPL	53%	58%	+5pp
COST	52%	57%	+5pp

On the S&P 500, the fourth day closes green 59% of the time after three red days, up from an everyday 55%. Amazon, Meta and Google each jump about six or seven points. These are enormous, broadly owned companies, and a three-day dip in them is usually met by the buy the dip crowd.

Sometimes the hottest names work in reverse

Now the other camp. These are the volatile names a dip-buyer is most tempted to grab on a pullback, and they are exactly where three down days points the wrong way. Some are young and speculative, some (like Boeing) are working through real trouble, but a run of red days on any of them tends to keep going.

Name	Green on any day	Green after 3 down days	Edge
BA	51%	49%	-2pp
COIN	47%	45%	-2pp
CRWD	53%	49%	-4pp
PLTR	51%	46%	-5pp
HOOD	50%	45%	-5pp
SOFI	49%	44%	-5pp
RIVN	47%	39%	-8pp

Even when it bounces, the payoff is small

There is one more reason to hold this loosely. When the fourth day is green, the average gain runs about 1% to 3%, depending on the name, and when it is red the average loss is about the same size. So even on the names that bounce a little more often, the up days and the down days are roughly matched. You are not paid much extra for the small improvement in odds.