Can a moving average act as support or resistance?
"It bounced off the 200-day." You'll hear it on every market channel, and the chart usually seems to agree. It agrees for a reason that has nothing to do with the line pushing back.
The mechanical reason bounces appear
In an uptrend, a rising moving average generally sits below price, because it averages older, lower closes. So a pullback shallow enough to leave the trend intact tends to run out of room somewhere near that line — not because the line pushed back, but because that is simply where a shallow pullback ends up. Allow a couple of percent of visual slack for "near", and a large share of ordinary pullbacks will look like moving-average bounces without any special force being involved. Deeper ones cut clean through a still-rising average and recover anyway, which is exactly the case the next section is about.
The fuzziness that makes it unfalsifiable
An index trades at 4,000 with its 200-day average at 3,900.
- A pullback stalls at 3,920 — 0.5% above the line. Reported as "held the 200-day."
- A later pullback dips to 3,860 — 1% below — then recovers. Reported as "tested the 200-day."
Both count. When the zone that qualifies as "the 200-day" is elastic enough to include misses in either direction, the claim can no longer fail — and a claim that cannot fail carries no information. The fix is not cynicism; it is a definition written before you look.
Three approaches to one steadily rising line, drawn to the same scale. Decide which of them "held" before you read the labels.
How to count honestly
Decide the rules in advance: a touch is any bar whose low comes within 1% of the 200-day average; a hold is a close higher ten sessions later. Then count every touch in the sample, including the ones where price sliced straight through and kept going. The results will be interesting either way — and they will be nothing like the highlight reel, because hindsight labels only the bounces and leaves the failures unmarked. That asymmetry, not the indicator, is what makes the folklore feel true.
Count on one price series. A published 200-day average is usually computed on dividend-adjusted closes, while the low printed on a candle is the price that traded, and the two drift apart by every dividend paid since. On 3 September 2021 the S&P 500 fund traded as low as 451.55. Its 200-day average of adjusted closes for that day reads 376.51 today; the same average of traded closes is 405.44. The same bar sits 19.9% above one line and 11.4% above the other (measured 29 September 2026; the first figure falls a little with every new dividend). A 1% touch rule applied across that mismatch counts nothing at the far end of a five-year window. Compare like with like: traded lows against an average of traded closes, or everything adjusted, as on the line chart below, where the price and its average are both drawn from adjusted closes.
The one real mechanism
There is a genuine partial effect, and it isn't mystical. Enough participants watch the 50-day and 200-day lines that resting orders and automated rules cluster near them, so the level can matter because it is watched. That's reflexivity — a crowd-behaviour effect. It is strongest in heavily followed, liquid instruments and fades to nothing where nobody is looking. It also decays: everyone watching the same level makes the level a target as easily as a floor.
The honest phrasing
Instead of "the 200-day is support", say: "price has stalled within 1% of the 200-day average on four of the last seven approaches." One sentence is a measurement; the other is a prophecy wearing a measurement's clothes.
Try it now
- Write your definition of a touch and a hold down first — a percentage and a horizon, both numbers. Then apply it to the three marked approaches on the schematic above and score them. Most people's first definition scores all three the same, which is the problem.
- Now run it where you cannot see the answer first. On the five years of the S&P 500 fund below, with its 200-day average drawn on the same adjusted closes as the line, count every instance your definition catches, the passes-through included. A line chart has no lows, so apply the touch rule to the closes and say so beside your tally. Write the tally down before you interpret it.
- Compare your tally with the confident version you've heard elsewhere. Next unit: the oscillators, where the same discipline is needed twice as badly.