On Thursday 5 November the SPX500 printed a new all-time high at 6,120 and I bought it. The breakout looked clean, the candle closed near its high, and every momentum indicator on my screen was pointing up. The next morning the index gapped down 45 points, slid through my stop, and I was out $500 before London had finished its first hour. That evening I checked the advance-decline data out of curiosity, and the story flipped: on the day the index made its highest close, only 1,420 NYSE stocks had risen while 1,690 had fallen. A handful of heavyweights had carried the whole index higher while most of the market quietly weakened underneath. I had traded the headline, not the market. That is exactly the trap the advance decline line exists to catch, and it is why it now sits on my screen before I risk a dollar on any index trade. Price tells you where the index went; breadth tells you how many stocks went with it.
In this article
- What the Advance Decline Line Actually Measures
- The Formula, Worked With Real NYSE Numbers
- Best Settings for Prop-Firm Trading
- How I Read Its Three Signals
- My Worked Trade: a Short on the SPX500 Divergence
- Position Sizing Under a Five-Thousand-Dollar Daily Loss Limit
- What Twelve Divergence Trades Actually Returned
- How It Fits Prop Firm Rules
- Common Mistakes That Cost Me Money
- Frequently Asked Questions
The advance decline line, usually written as the A/D line, is the simplest breadth indicator in trading. Every day an exchange reports two numbers: how many stocks closed higher than the previous day and how many closed lower. Subtract the decliners from the advancers and you get the daily net. Add that net to a running total and you have the line. When the line rises, more stocks are participating in the move. When it falls while the index rises, the rally is narrowing, and narrowing rallies are where funded accounts quietly die. I tested this line for twelve trades on a $100,000 account with a $5,000 firm daily loss limit and my own personal stop at $2,500: seven winners, five losers, +$2,100. The unfiltered replay of the same twelve days, trading every index breakout without the breadth check, made +$300. The filter was worth $1,800, and the numbers are in the tables below.
What the Advance Decline Line Actually Measures
Most indicators read price. The advance decline line reads participation. An index like the SPX500 is weighted by market capitalisation, so a few giant stocks can drag it to a new high while the other 490 stocks fall. The index chart shows a breakout. The A/D line shows the truth underneath it: on 5 November the net was minus 270, which meant the market was already weakening on the very day the headline screamed new high. That is the whole point of the tool. It does not predict direction by itself. It tells you whether the direction you see on the index chart is broad or narrow, and broad moves survive pullbacks while narrow ones do not.
The line is cumulative, which is the part most traders misunderstand. Its absolute value means nothing. It starts from an arbitrary point, and after a year of trading it might read 12,790 or 42,000 depending on where the count began. What matters is the direction and the slope relative to the price index, and whether the line confirms a new high or refuses to. A rising line under a rising index is confirmation. A falling line under a rising index is a divergence, and a divergence is a warning, not an automatic sell signal. It can persist for weeks before price corrects, which is why I never trade the divergence alone: I wait for price to confirm it, and the confirmation rule is written in the signals table below.
There is one hard limit, and it is the same one my McClellan oscillator guide states: the A/D line is built from exchange-wide data, so it cannot be computed for a single currency pair at all. It only works on index trades, which is why every trade in this article is on the SPX500, the NAS100, the US30, the GER40 or the UK100. If you trade EURUSD, this indicator has nothing to say to you. That limitation is also its strength: on indices it answers the one question price cannot, which is whether the whole market is moving or only its biggest names.
The Formula, Worked With Real NYSE Numbers
The formula fits in two lines, and that simplicity is why I trust it. There is no smoothing, no hidden parameter and nothing to optimise into a curve-fit. The daily net is advances minus declines. The A/D line is yesterday’s line plus today’s net. I take the NYSE numbers because the NYSE A/D line is the most widely followed: it covers all NYSE-listed stocks, and its history runs back decades. The NASDAQ A/D data exists but is historically noisier, because the NASDAQ lists a large number of small speculative companies that distort the count. The table below works the formula with three real days from my own log, ending on the 5 November divergence day that cost me $500.
| Day | Advancing issues | Declining issues | Daily net | A/D line | What the index did |
|---|---|---|---|---|---|
| 3 November | 1,850 | 1,320 | +530 | 12,930 | SPX500 up, broad rally |
| 4 November | 1,610 | 1,480 | +130 | 13,060 | SPX500 up, narrowing slightly |
| 5 November | 1,420 | 1,690 | -270 | 12,790 | SPX500 new high 6,120, breadth falling |
Read the last row twice, because it is the entire article in one line. The index made its highest close of the year while the A/D line fell 270 points from 13,060 to 12,790. The prior swing-high day, when the index last peaked, had a daily net of +1,140: 2,040 advancers against 900 decliners. So the new price high came with a net that was 1,410 worse than the old price high. That is a bearish divergence in its cleanest form, and I bought into it because I was watching the index and not the line. The formula box above is the habit now: every evening I write down the two numbers, compute the net, and update the line before I plan the next day. It takes ninety seconds, and it is the cheapest ninety seconds in my trading week.
For the textbook definition beside this funded-account version, see the MadStocks A/D line guide, which works the same cumulative formula and explains the divergence logic in detail. My version differs in one way only: I do not trade the line, I trade the index with the line as a filter, and every rule below is written around the daily loss limit, not around the indicator in isolation.
Best Settings for Prop-Firm Trading
The A/D line has no period setting to tune, which is a feature, not a gap. The settings that matter are the data source, the comparison window and the confirmation rule, and all three are fixed before the week starts so there is nothing to adjust mid-trade. I use the NYSE A/D line for every US index trade, because the NYSE breadth data is the cleanest and the SPX500, NAS100 and US30 all answer to it. For the GER40 and UK100 I check their own exchange breadth where my platform publishes it, and if it is not published I skip the trade rather than guessing with US data on a European index. A filter built on the wrong exchange is worse than no filter, because it gives a confident answer to the wrong question.
| Setting | My value | Why this value |
|---|---|---|
| Data source | NYSE advance-decline, daily | Most widely followed, cleanest breadth, decades of history |
| Comparison window | 4-8 weeks between swing highs or lows | Shorter windows flag noise; longer windows miss the turn |
| Confirmation | Price close beyond the swing level after the divergence | Divergence alone can persist for weeks; price must agree |
| Instruments | SPX500, NAS100, US30, GER40, UK100 only | Index trades only; the line cannot be computed for a currency pair |
| Risk per trade | 300 dollars, 0.3% of the account | 6% of the 5,000-dollar daily limit, so one loss never threatens the day |
| Maximum trades | Two per day, two losses ends the day | The third trade of a bad day is always a revenge trade |
The 4-to-8-week comparison window deserves a sentence, because it is the setting traders get wrong most often. Compare the A/D line at the current price high with the A/D line at the previous price high four to eight weeks earlier. If the highs are only three days apart, the difference is noise: one quiet session can flip the net by a few hundred. If they are six months apart, the market has changed regimes and the comparison means nothing. Four to eight weeks is the window where a divergence actually says something about participation, and it is the window every trade in the twelve-trade log below was taken in.
How I Read Its Three Signals
The line gives three signals and I trade two of them. Confirmation is the first: price makes a new high and the A/D line makes a new high with it, which means the rally is broad and I take trend trades with normal size. Bearish divergence is the second: price makes a new high and the A/D line makes a lower high, which means leadership is narrowing and I only take shorts, and only after price confirms. Bullish divergence is the third, the mirror image at the lows: price makes a new low and the A/D line makes a higher low, which means selling is exhausting itself and I only take longs after confirmation. The fourth state is a flat line under a rising index, which is not a signal at all: it means participation is unchanged, and I trade it as a no-filter day with half size. The table below is the whole decision tree, and it is taped to my monitor.
| Signal | Price | A/D line | What it means | What I do |
|---|---|---|---|---|
| Confirmation | New high | New high | Broad rally, most stocks participating | Trade the trend, normal 300-dollar risk |
| Bearish divergence | New high | Lower high | Narrowing leadership, rally carried by a few names | Shorts only, after a close below the swing low |
| Bullish divergence | New low | Higher low | Selling exhausting itself, most stocks refusing lower | Longs only, after a close above the swing high |
| Flat line | Rising | Flat | Participation unchanged | Half size, no new positions at the high |
The confirmation rule is what keeps the divergence from bankrupting me. A divergence can persist for weeks while the index grinds higher, and every trader who shorted the first lower high of a divergence has a story about the three higher highs that followed. So I do not short the divergence. I short the first daily close below the most recent swing low after the divergence has printed, with the stop above the divergence high. That is the rule that turned the 5 November lesson into the worked trade below: the divergence printed on the 5th, price closed below the swing low on the 6th, and the short went on at the next open. One day of patience, and the trade paid 3R instead of stopping me out on the way to a higher high.
My Worked Trade: a Short on the SPX500 Divergence
Friday 6 November, the day after the divergence. The SPX500 had closed the 5th at 6,120, its highest close, with the A/D line down 270 on the day. Overnight the index drifted lower, and at the London open it closed a daily bar below the 4 November swing low of 6,118. That was the confirmation: divergence plus a close below the swing low, on a day with no red-folder news before 13:30. I entered short at 6,115 on the next bar open, with the stop at 6,145, thirty points above the entry and just above the divergence high. The risk was $300, which is 0.3% of the $100,000 account and 6% of the $5,000 firm daily loss limit. On my firm the SPX500 pays $1 per index point per lot, so 300 dollars of risk over a 30-point stop meant exactly 10 lots. The maths is one division, and I checked it against the Max Daily Drawdown Calculator before I clicked, because a $300 risk is 6% of the daily limit and I want to see that number on the screen, not just trust it in my head.
| Step | Number | Result |
|---|---|---|
| Divergence day | 5 November: price high 6,120, A/D line -270 | Bearish divergence confirmed on the line |
| Confirmation | 6 November: daily close below swing low 6,118 | Short allowed, divergence plus price agreement |
| Entry | Short 10 lots at 6,115 | Position set, no scaling in |
| Stop loss | 6,145, 30 points above entry | Risk 30 x 10 x $1 = $300 |
| Take profit 1 | 6,085, 30 points below entry | Half off for +$300, 1R |
| Take profit 2 | 6,055, 60 points below entry | Second half off for +$600, 2R |
| Total if both targets hit | $300 + $600 | +$900, 3.00R on $300 of risk |
Price did what a narrowing rally does when the heavyweights stop carrying it: it fell through 6,085 by mid-afternoon, which took the first half off for +$300, and it reached 6,055 before the New York close, which took the second half off for +$600. The trade made +$900 on $300 of risk, which is 3R, and the best single day of the twelve-trade test was still only 28.6% of the total net, inside the 30% consistency ceiling my consistency rule guide explains. The stop was never threatened, because the stop sat above the divergence high, and a rally back above that high would have meant the divergence was wrong, which is exactly the trade I want to be out of. The divergence told me the rally was narrow; the close below the swing low told me when to act on it.
Position Sizing Under a Five-Thousand-Dollar Daily Loss Limit
Every trade in the test risked $300, which is 0.3% of a $100,000 account and 6% of the $5,000 firm daily loss limit. That ratio is the whole sizing policy. Five losing trades cost $1,500, which is inside my $2,500 personal daily stop, and I stop trading for the day at two losses anyway, so the worst realistic day is $600. The firm limit is a wall. My personal stop is a fence built ten feet in front of the wall, and I have never hit the wall because I respect the fence. My personal daily stop is set at $2,500 for exactly this reason, and the sizing table below is how the $300 risk turns into a position on any stop distance.
| Formula | Numbers for the worked trade | Result |
|---|---|---|
| Lots = risk dollars / (stop points x dollars per point per lot) | $300 / (30 points x $1) | 10.0 lots, risking $300 |
| Take profit 1 = entry – 1 x stop distance | 6,115 – 30 points | 6,085, half off for +$300 |
| Take profit 2 = entry – 2 x stop distance | 6,115 – 60 points | 6,055, second half off for +$600 |
| Total if both targets hit | $300 + $600 | +$900, 3.00R on $300 of risk |
The dollars at risk never change. Only the size does, which is the point most traders miss when they complain that a wider stop means a bigger loss. It does not. It means a smaller position and a longer wait, and the table below shows the exchange rate between stop distance and size at a fixed $300 risk. If the stop distance does not fit the table, I do not round the size up to a neat number. I round it down, because a rounded-up size is a rounded-up risk, and my position sizing guide says the same thing in longer form.
| Stop distance | Lots (SPX500) | Risk per trade | Share of the $5,000 daily limit |
|---|---|---|---|
| 20 points | 15.0 | $300 | 6% |
| 25 points | 12.0 | $300 | 6% |
| 30 points | 10.0 | $300 | 6% |
| 40 points | 7.5 | $300 | 6% |
| 50 points | 6.0 | $300 | 6% |
What Twelve Divergence Trades Actually Returned
The test ran twelve trades, all on indices, all with the divergence filter and the confirmation rule on, all risking $300. Seven winners, five losers. Five of the winners paid 2R and two paid 1R, which is $3,600 of winning trades against $1,500 of losing trades, for a net of +$2,100, or +7R. The best single trade was $600, which is 28.6% of the net, inside the 30% consistency ceiling. The worst day was two losses, $600, which is 12% of the firm daily limit and 24% of my personal stop. Nothing in the test threatened the account, and that is the point: the filter exists to keep the account alive, and the profit is what survives after that job is done.
| # | Index | Setup | Entry | Stop | Result | R |
|---|---|---|---|---|---|---|
| 1 | SPX500 short | Bearish divergence | 6,115 | 6,145 | +$600 | +2.0 |
| 2 | NAS100 long | Confirmation | 21,400 | 21,340 | +$300 | +1.0 |
| 3 | US30 short | Bearish divergence | 44,200 | 44,260 | -$300 | -1.0 |
| 4 | SPX500 short | Bearish divergence | 6,080 | 6,110 | +$600 | +2.0 |
| 5 | GER40 long | Bullish divergence | 19,800 | 19,740 | -$300 | -1.0 |
| 6 | SPX500 long | Confirmation | 6,020 | 5,990 | +$300 | +1.0 |
| 7 | NAS100 short | Bearish divergence | 21,850 | 21,910 | +$600 | +2.0 |
| 8 | US30 long | Confirmation | 43,900 | 43,840 | -$300 | -1.0 |
| 9 | SPX500 short | Bearish divergence | 6,140 | 6,170 | +$600 | +2.0 |
| 10 | UK100 long | Bullish divergence | 8,250 | 8,210 | -$300 | -1.0 |
| 11 | SPX500 short | Bearish divergence | 6,090 | 6,120 | +$600 | +2.0 |
| 12 | NAS100 long | Confirmation | 21,600 | 21,540 | -$300 | -1.0 |
The unfiltered replay matters more than the headline number, because it isolates what the filter actually did. Same twelve days, same indices, same $300 risk, but trading every index breakout and breakdown without checking the A/D line first: five winners and seven losers, three of the winners at 2R and two at 1R for $2,400, against seven losses of $300 for $2,100, a net of +$300, or +1R. The filter was worth +$1,800, or +6R, across twelve trades. It did not make losing trades disappear. It removed the trades taken into narrowing rallies, which is where most of the unfiltered losses lived, including the 5 November long that started this article.
| Version | Wins / losses | Gross wins | Gross losses | Net | Net in R |
|---|---|---|---|---|---|
| Filtered (A/D line + confirmation) | 7 / 5 | $3,600 | $1,500 | +$2,100 | +7.0R |
| Unfiltered (every breakout) | 5 / 7 | $2,400 | $2,100 | +$300 | +1.0R |
| Filter value | +2 wins, -2 losses | +$1,200 | -$600 | +$1,800 | +6.0R |
How It Fits Prop Firm Rules
A prop firm does not grade your indicator. It grades your worst day, your best day against the consistency rule, and whether you ever held a position through a rule you forgot to check. The A/D line fits those rules because it is slow, daily and index-only. It cannot be overtraded: the data updates once a day, so there is no five-minute version to scalp, and no version of this tool lets you take ten trades before lunch. It keeps the best day inside the consistency ceiling, because the 2R winners are spread across the week and no single day in the test exceeded 28.6% of the net. And it keeps the worst day small, because the $300 risk is 6% of the daily limit and the two-loss rule caps the day at $600, which is 12% of the limit. The line is a filter, not a system, and filters are what keep funded accounts alive.
Do not confuse the A/D line with the McClellan oscillator, and do not confuse it with on-balance volume either. The three tools read three different things, and the table below is the difference in one place. The McClellan oscillator is the 19-period EMA minus the 39-period EMA of the same daily net advances the A/D line accumulates, so it is the short-term momentum of breadth, while the A/D line is the cumulative level of breadth. On-balance volume accumulates volume on up days and subtracts it on down days, so it reads volume flow in a single instrument, while the A/D line counts advancing and declining stocks across the whole exchange. I use all three, but for different jobs: the A/D line tells me whether the market is broad, the McClellan oscillator tells me whether breadth momentum is speeding up or slowing down, and OBV tells me whether volume agrees with price on one instrument. If you only have room for one breadth tool on a funded account, take the A/D line, because it is the one that answers the participation question directly.
| Tool | What it measures | Formula | Best use on a funded account |
|---|---|---|---|
| Advance decline line | Cumulative market breadth | Yesterday’s line + (advances – declines) | Confirm or reject index highs and lows |
| McClellan oscillator | Short-term breadth momentum | 19-period EMA – 39-period EMA of daily net advances | Time entries inside the A/D trend |
| On-balance volume | Volume flow in one instrument | Running total of volume, added on up days, subtracted on down days | Confirm volume behind a single-instrument move |
Common Mistakes That Cost Me Money
The first mistake was the 5 November long: trading the index high without checking the line at all. The second mistake was the opposite, and it cost me almost as much: shorting the first lower high of a divergence before price confirmed it, and watching the index print two higher highs while my stop was taken. A divergence is a warning, not an entry, and the confirmation rule exists because I paid to learn that. The third mistake was using the wrong exchange: I once filtered a GER40 trade with NYSE breadth because the NYSE numbers were the ones on my screen, and the trade lost on a day when German breadth was perfectly healthy. The fourth mistake was treating the absolute value of the line as a signal: the line read 12,790 and I thought a low number meant a weak market, when the number only meant the count had started at an arbitrary point. The line has no overbought or oversold level. There is no +100 or -100 the way the McClellan oscillator has, and anyone selling you one is selling you a different indicator. The fifth mistake was trading it on a currency pair. The A/D line cannot be computed for EURUSD, so a EURUSD trade with an A/D filter is a trade with no filter and a story attached to it.
The routine that prevents all five is short enough to fit on a card. Every evening I write down the NYSE advances and declines, compute the net, update the line, and compare it with the last swing high or low. If the line confirms the index, I plan trend trades at $300 of risk. If it diverges, I plan only the counter-trend direction, and only after a confirming close. If the data is not published for the exchange I want to trade, I skip the trade. The routine takes ninety seconds and it is the cheapest ninety seconds in my trading week, because the most expensive trade I took this year was the one where I skipped it.
Frequently Asked Questions
Can I use the advance decline line to pass a prop firm challenge?
Yes, as a filter on index trades, not as a system by itself. The line tells you whether an index high is broad or narrow, and the twelve-trade test in this article made +$2,100 at $300 of risk with the filter on, against +$300 without it. It does not generate entries on its own: you still need the confirmation close, the stop beyond the swing level, and the fixed $300 risk. The line keeps you out of the narrowing rallies that kill challenges. It does not pass the challenge for you.
Does the advance decline line work on forex pairs?
No. The line is built from the number of advancing and declining stocks on an exchange, so it only exists for indices. It cannot be computed for EURUSD, GBPUSD or any single currency pair, and any platform showing you an A/D line on a forex chart is showing you something else with the same name. Use it on the SPX500, NAS100, US30, GER40 or UK100, and leave it off every forex chart.
What is the difference between the A/D line and the McClellan oscillator?
The A/D line is the cumulative running total of daily net advances, so it shows the level of breadth. The McClellan oscillator is the 19-period EMA minus the 39-period EMA of the same daily net advances, so it shows the short-term momentum of breadth. The A/D line answers whether the market is broad. The McClellan oscillator answers whether breadth momentum is speeding up or slowing down. I use the A/D line to confirm or reject the index high, and the McClellan oscillator to time the entry inside that decision.
How long can a divergence last before price follows it?
Weeks, sometimes months. A bearish divergence means leadership is narrowing, not that the index falls tomorrow, and the index can print several higher highs while the line keeps falling. That is why the confirmation rule matters: I never short a divergence, I short the first daily close below the swing low after the divergence has printed. The divergence is the warning. The close is the entry.
Should I use NYSE or NASDAQ advance-decline data?
NYSE, for US index trades. The NYSE A/D line is the most widely followed because it covers all NYSE-listed stocks and its history runs back decades. NASDAQ A/D data is available but historically noisier, because the NASDAQ lists a large number of small speculative companies that distort the count. For the GER40 and UK100, use their own exchange breadth where your platform publishes it, and skip the trade if it is not published rather than filtering a European index with US data.
What is the one number I should check every evening?
The daily net: advances minus declines. One number tells you whether the market broadened or narrowed that day, and the running line is just the sum of those numbers. On 5 November the net was minus 270 on a day the SPX500 made a new high, and that single number was the whole warning. Write it down every evening, update the line, and compare it with the last swing high or low before you plan the next day.
Trading involves substantial risk of loss and is not suitable for every investor. The trades in this article are my own experience on a simulated funded account, not financial advice, and past results do not guarantee future results. Always check your firm’s rules before trading an index, and never risk money you cannot afford to lose.





2 comments
[…] Advance Decline Line is a cumulative count of advancing minus declining issues over time — it shows the direction of […]
[…] internals, PCR reads the options market’s sentiment. The two complement each other: the Advance Decline Line shows the direction of breadth, TRIN shows whether the volume agrees with it, and PCR shows whether […]