Every term this site uses, in plain language — prop firm rules,
backtest metrics, Pine Script vocabulary, and 19 candlestick patterns with a
diagram and the conditions that actually qualify each one.
Last updated 2026-08-25 · 84 terms
Definitions describe what something is. None of them promise what
price will do next, and nothing here is financial advice. Prop firm rules in particular
differ by firm, account size, account stage and platform, and they change without
notice — treat the entries below as an explanation of the mechanism and check your own
firm's current rulebook before you trade.
Interactive
Test yourself
Ten questions drawn from the entries below. Nothing is sent
anywhere — your best score is kept in this browser only.
best
Read the diagram and name the pattern. One attempt per question — you get the
reasoning either way, and a link to the full entry.
Candles & price action
19 terms
Chart formations, what each one is, and where it misleads.
An up candle followed by a down candle whose body fully covers it.
Typically qualifies when
The first candle's body is bullish, the second bearish.
The second body covers the first body's full range top and bottom.
There is a prior up move for it to reverse.
Common false positives
The first candle has a negligible body.
Comparing wick extremes instead of body extremes.
Immediately after a news spike, where a single print can produce the shape mechanically.
The mirror of a bullish engulfing. Buyers finished the first session in control; the second session opened at or above that close and gave the whole range back. Reads as more meaningful at the top of an extended move than in the middle of a range, but that judgement is context, not a rule the pattern itself supplies.
A large up candle followed by a small down candle contained inside its body.
Typically qualifies when
Candle 1 has a large bullish body.
Candle 2 is bearish and its body sits inside candle 1's body.
It follows an advance.
Common false positives
Wicks extending outside candle 1 while the body stays in — still a harami, often misread as broken.
Reading it after a flat, directionless stretch.
The mirror of the bullish harami: a wide bullish session followed by a small bearish one that never leaves the first body. Buying paused. As with all harami, compare bodies, not ranges.
A down candle followed by an up candle whose body fully covers it.
Typically qualifies when
The first candle's body is bearish (close below open).
The second candle's body is bullish (close above open).
The second body's high and low both extend past the first body's — bodies, not wicks.
It appears after a visible move down, so there is something to reverse.
Common false positives
The first candle is a doji or near-doji — almost any candle engulfs a body that tiny.
The second candle's wicks engulf but its body does not.
It forms mid-range with no prior downswing, which is most of the time.
On a low-volume bar such as the futures maintenance-break re-open.
Two candles. The first closes lower than it opened; the second opens at or below the first candle's close and closes at or above its open, so the second body swallows the first. What it describes is a session where sellers finished in control and the next session reversed that entire range. It is a description of what already happened, not a forecast — the same shape appears constantly inside ranges where it means nothing.
A large down candle followed by a small up candle contained inside its body.
Typically qualifies when
Candle 1 has a large bearish body.
Candle 2 is bullish and its body sits inside candle 1's body.
It follows a decline.
Common false positives
Comparing full ranges rather than bodies — harami is a body-inside-body pattern.
Candle 2 is so small it is effectively a doji.
An inside bar with a colour requirement. A wide bearish session is followed by a small bullish one that stays entirely inside the first body. It describes selling that stopped rather than buying that started, which is why it is usually read as weaker than an engulfing.
An up candle followed by a down candle that closes below its midpoint but not below its open.
Typically qualifies when
Candle 1 is bullish with a substantial body.
Candle 2 opens at or above candle 1's close.
Candle 2 closes below the midpoint of candle 1's body but above its open.
Common false positives
Candle 2 closing below candle 1's open — that is a bearish engulfing.
On futures, where the required open above the prior close is often just the overnight drift.
The mirror of the piercing line and the partial form of a bearish engulfing. The second session opens at or above the first close and gives back more than half of the first body without covering it completely.
A candle that opens and closes at essentially the same price.
Typically qualifies when
Open and close are within a small fraction of the candle's total range.
There are wicks on both sides — otherwise it is a flat bar, not indecision.
It appears on a bar with normal participation, not a dead one.
Common false positives
Any 1-minute bar during lunch, overnight, or a holiday session.
Treating every doji as a reversal. Most are noise.
The body is a line rather than a block. It says the session finished where it started, whatever happened in between. On its own that is a statement about one bar's net change and nothing else — dojis are extremely common on low-volatility bars and on any short timeframe, where they carry no information at all.
A three-candle sequence: a long up candle, a small-bodied pause, then a long down candle.
Typically qualifies when
Candle 1 has a large bullish body.
Candle 2 has a small body, ideally gapping above candle 1.
Candle 3 has a large bearish body closing well into candle 1's body.
Common false positives
Candle 2 is not actually small relative to the other two.
Reading it on a 5-minute chart, where three-bar shapes recur constantly.
The mirror of the morning star. A strong advance stalls into a small-bodied session and is then given back by a strong decline. As with the morning star, the gaps that the classical definition assumes are largely a stock-market artifact and mostly absent on continuously traded futures.
A candle with a small body at the top and a long lower wick, after a decline.
Typically qualifies when
Lower wick is roughly twice the body or longer.
Upper wick is small or absent.
The body sits in the upper third of the candle's total range.
It follows a decline — that is what separates a hammer from a hanging man.
Common false positives
Reading it in isolation with no prior downswing.
On an illiquid contract or a thin session, where one bad fill draws the wick.
On a daily chart built from a session that included the overnight gap.
Price traded well below the open and closed back near the top of its range. The long lower wick is the whole point: it records rejection of the lower prices within that session. The same shape at the top of a rally is called a hanging man — identical geometry, different location. The location is doing the work, not the candle.
A candle whose entire range sits inside the previous candle's range.
Typically qualifies when
The inside bar's high is at or below the prior high.
Its low is at or above the prior low.
The prior bar is large enough that the containment means something.
Common false positives
A chain of inside bars inside an already-tiny range.
Overnight bars contained inside the prior full RTH session — mechanically inside, informationally empty.
High lower than the prior high, low higher than the prior low. It describes contraction: the second session never left the boundaries the first one set. Traders use the prior bar's high and low as the reference levels for a break either way, which is why inside bars show up so often in breakout logic — they hand you an unambiguous range with no parameters to tune.
A candle with a full body and effectively no wicks.
Typically qualifies when
Wicks are negligible relative to the body on both ends.
The body is large relative to recent bars — compare it against ATR, not against zero.
Common false positives
Tiny bars with no wicks on fast timeframes. Technically marubozu, informationally nothing.
A bar whose range is one or two ticks on an illiquid contract.
The open is the low and the close is the high, or the reverse. One side never took control of the session at any point. On futures this is far more common on short timeframes than on daily bars, and on a 1-minute chart a marubozu is often just a bar with two ticks of range.
A three-candle sequence: a long down candle, a small-bodied pause, then a long up candle.
Typically qualifies when
Candle 1 has a large bearish body.
Candle 2 has a small body of either colour, ideally gapping away from candle 1.
Candle 3 has a large bullish body closing well into candle 1's body.
Common false positives
Candle 3 closes only marginally into candle 1 — the shape is there, the recovery is not.
Applying it on intraday futures charts, where the gaps the pattern assumes rarely exist.
The middle candle is the pattern. A large down session is followed by one where almost nothing happens — the selling stops without buying starting — and then a large up session. The third candle is usually expected to close into the upper half of the first candle's body; how far in is a judgement call, and different sources draw the line differently.
A candle whose high is above and low is below the previous candle's range.
Typically qualifies when
High exceeds the prior high and low undercuts the prior low.
The prior bar was a normal-sized bar, not a doji.
Common false positives
Immediately after a gap, where the first bar mechanically covers the previous range.
Assuming direction from the shape alone.
The opposite of an inside bar: this session traded through both ends of the previous one. It is a range statement, not a direction statement — an outside bar can close anywhere. Where the close lands inside that expanded range is the part worth reading.
A down candle followed by an up candle that closes above its midpoint but not above its open.
Typically qualifies when
Candle 1 is bearish with a substantial body.
Candle 2 opens at or below candle 1's close.
Candle 2 closes above the midpoint of candle 1's body but below its open.
Common false positives
Closing only just past the midpoint.
Candle 2 clearing candle 1's open entirely — that is an engulfing, not a piercing line.
A partial version of the bullish engulfing. The second session opens at or below the first close and recovers more than half of the first candle's body, but stops short of covering it. The midpoint is the conventional threshold, and it is a convention rather than anything the market observes.
A candle whose wick is far longer than its body, marking a rejected price area.
Typically qualifies when
The wick on one side is at least two thirds of the candle's total range.
The body is small and sits at the opposite end.
The wick pokes past a level that was already on the chart.
Common false positives
Long wicks at the session open or the maintenance-break re-open.
Any bar during a scheduled economic release.
A long wick in the middle of a range with no reference level nearby.
An umbrella term covering the hammer, the shooting star and the inverted hammer. The shared idea is that price visited a level and did not stay. Pin bars are most often read where they intersect something already on the chart — a prior high, a session level, a moving average — rather than in open space.
A candle with a small body at the bottom and a long upper wick, after a rally.
Typically qualifies when
Upper wick is roughly twice the body or longer.
Lower wick is small or absent.
The body sits in the lower third of the range.
It follows an advance.
Common false positives
At the open of the RTH session, where the first minutes routinely draw long wicks.
Around a scheduled release, where the wick is one headline rather than rejection.
Without a prior rally, where it is just a bar with a long wick.
Price pushed well above the open and gave all of it back before the close. The long upper wick records rejection of the higher prices. The identical shape after a decline is an inverted hammer and is read the opposite way — again, position in the swing carries the meaning.
Three consecutive down candles, each opening inside the last body and closing lower.
Typically qualifies when
Three consecutive bearish bodies.
Each opens within the previous body and closes below the previous close.
Lower wicks stay short.
Common false positives
One of the three is a doji or near-doji.
The sequence spans a weekend or a contract roll.
The mirror of three white soldiers: three sessions of orderly selling, each closing near its low. Read the lower wicks the same way — long ones say the selling was being absorbed each session, which is a different story from the shape alone.
Three consecutive up candles, each opening inside the last body and closing higher.
Typically qualifies when
Three consecutive bullish bodies.
Each opens within the previous candle's body.
Each closes above the previous close, near its own high.
Upper wicks stay short.
Common false positives
The third candle is much smaller than the first two — momentum already fading.
After an extended run, where it describes the end of a move as readily as a start.
A description of steady, orderly buying across three sessions with little given back. The detail that matters is the wicks: small upper wicks mean each session closed near its high. Long upper wicks on the same three-candle sequence describe something quite different — buying that kept getting sold into.
Two adjacent candles that share almost exactly the same low.
Typically qualifies when
The two lows are within a tick or two of each other.
The candles are adjacent.
The pair follows a decline.
Common false positives
Two lows that merely look level at the chart's zoom level — check the actual prices.
A shared low at a round number that everything touches anyway.
Two sessions probed the same price and neither went through it. The shared low is the observation; it is a level that has now been tested twice in quick succession. Whether the second test means anything depends entirely on whether that price was already significant.
Two adjacent candles that share almost exactly the same high.
Typically qualifies when
The two highs are within a tick or two of each other.
The candles are adjacent.
The pair follows an advance.
Common false positives
Highs that are 'close enough' visually but several ticks apart.
Two bars capped by the same session high purely because the session ended.
The mirror of a tweezer bottom. Two sessions reached the same price and stopped. As with the bottom, the value is in whether that price already mattered — an untested high in open space is a weaker reference than a prior swing high or the top of a well-defined range.
Heavy volume that produces almost no price movement.
Large size is arriving and something is taking the other side of all of it, so price barely moves. That much is observable. What is not observable is who is absorbing or why — a passive buyer defending a level and a seller quietly distributing into strength both look like this on a chart. Absorption tells you a large participant was willing to transact here, never who they were or whether they will keep doing it.
Two readings of the same thing
A level being defended
Someone is willing to keep taking the other side at this price, so the level holds.
Distribution into strength
Someone is quietly unloading into the demand, so the level fails once they are done.
A cumulative volume line weighted by where the close sat inside each bar's range.
A refinement of on-balance volume: instead of counting a bar as wholly up or wholly down, it weights that bar's volume by where the close finished within the high-low range. A close near the high adds most of the volume; a close in the middle adds almost nothing. It handles indecisive bars far better and still cannot see who was buying, or why.
The period when open interest and volume migrate from the expiring contract to the next one.
For a few sessions each quarter, volume is split across two contracts and neither chart shows the whole market. Any volume comparison spanning a roll is comparing different things, and a continuous back-adjusted chart stitches the price series without stitching the volume — which is a standing source of fake volume divergences in backtests that nobody checks the dates on.
A running total of volume that traded at the ask minus volume that traded at the bid.
An attempt to split volume into aggressive buying and aggressive selling by classifying each trade by which side of the spread it hit. It is the closest a chart gets to directional volume, and it is still an approximation: it captures who crossed the spread, not who was passively sitting there absorbing it. Delta rising while price stalls is the classic absorption picture — aggression meeting something willing to take it.
Two readings of the same thing
Aggression
Buyers are lifting offers, so the pressure is genuine.
Absorption
Buyers are lifting offers and price is not moving, so someone larger is selling into every one of them.
Comparing how much volume was spent against how far price actually moved.
Volume is the effort, range is the result. Big effort with a small result says the move is meeting real opposition; small effort with a big result says almost nobody is standing in the way. It is one of the few volume concepts that is a genuine comparison rather than a threshold, which is why it survives changing conditions better than "volume above X" ever does.
Two readings of the same thing
Opposition
High volume, small range: the move is being fought. The level matters to someone.
Emptiness
Low volume, large range: nobody is participating. Common overnight and around holidays, where it means thin liquidity rather than conviction.
A running total that adds a bar's whole volume on an up close and subtracts it on a down close.
A cumulative line, so only its direction means anything — the absolute value depends entirely on where the calculation started. Its weakness is the all-or-nothing rule: a bar that closes a single tick higher contributes exactly as much as one that closes on its high, which throws away most of what happened inside the bar. The accumulation/distribution line exists to fix precisely that.
The single price level with the most traded volume in a profile.
The price at which the most business was agreed over the period — the profile's fattest row. It is used as a reference level in both directions, which is the usual trap: the same level is read as a magnet by mean-reversion traders and as a barrier by breakout traders, and the level itself does not tell you which it is going to be today.
Two readings of the same thing
Magnet
Price is drawn back toward the most-traded price. The common reading inside a balanced, ranging market.
Barrier
Price has to work through a lot of previously agreed business to pass. The common reading when price is trending away from it.
Current volume compared with the usual volume for that time of day.
Raw volume is close to meaningless without a baseline: the first minute of the cash session dwarfs anything at 2am, every day, in every market. Relative volume divides the current figure by the average for that same slot, so 2.0x means twice the normal amount of business for this time of day. It is the form of the number worth putting in a strategy — an absolute volume threshold silently becomes a time-of-day filter.
A count of price changes in a bar, used where real volume is unavailable.
Not volume. It counts how many times the price updated, which correlates with activity but is not a measure of size — one 500-lot trade and one 1-lot trade each produce a single tick. It is standard on spot forex and on some CFD feeds. On futures you have real contract volume and there is no reason to use this; if a strategy was developed on tick volume and is then run on a futures feed, the input has quietly changed meaning.
The price band containing the bulk of a session's traded volume, conventionally 70%.
The range within which most of the period's business was actually done, bounded by the value area high and low. Its usefulness is as a definition of "normal" for that session: trading inside it is a market that agrees on price, and an excursion outside it is a market testing whether a new price is acceptable. The 70% figure is a convention borrowed from a standard deviation, not a property of markets.
The number of contracts that changed hands during a bar.
Every contract traded has a buyer and a seller, so volume is not "buying pressure" — it is activity. It counts how much business was done, never which side won. On futures it is a real, exchange-reported figure; on spot forex there is no central exchange and the "volume" on your chart is usually tick count instead. The single most common error in volume analysis is treating a large number as directional.
Two readings of the same thing
Participation
A high figure means more participants agreed that this price was worth transacting at. That is information about interest.
Conviction
A high figure means one side was aggressive. That is an inference, not an observation — the tape shows size, not intent.
An extreme-volume bar at the end of a move, usually with a long rejecting wick.
The idea is that a move ends when the last willing participants have finally acted — capitulation produces the largest volume of the whole swing, and there is then nobody left to continue it. The tell is the combination: enormous volume and a failure to hold the extreme. Volume alone does not make a climax, which is why the label is applied far more often in hindsight than in advance.
Two readings of the same thing
Exhaustion
The move is finishing. Supported when the bar rejects its own extreme and volume falls away sharply on the bars that follow.
Continuation
A large, committed participant is entering with size and the move continues. Supported when price holds near the extreme and the following bars stay there on lower volume.
Price making new extremes while volume steadily falls.
Each new high is being made on less business than the one before it. The standard reading is that the move is running out of participants. It is worth holding loosely: volume declines for calendar reasons too — into a holiday, ahead of a major release, or across a contract roll — and none of those say anything about the trend.
Two readings of the same thing
Weakening
Fewer participants are willing to transact at each new level.
Calendar
Everyone is simply away, or positioning ahead of an event. Check the date before reading anything into it.
Volume falling away sharply during a pullback or a range.
A pullback that nobody is participating in. The classical reading is bullish for an uptrend — if the pullback had real selling behind it you would see it in the volume — but the identical picture appears when a market has simply stopped being interesting, and a market nobody is trading can drift a long way on very little.
Two readings of the same thing
Healthy pause
The prior move is intact and profit-taking is light. More likely when the pullback holds above a level that mattered.
Loss of interest
Participation has left entirely. More likely when the dry-up persists for many bars and the range keeps narrowing.
Volume plotted by price level rather than by time.
A standard volume histogram answers "how much traded during this bar?" A profile answers "how much has traded at this price?" — which is the more useful question for finding levels, because it shows where business was actually done rather than where a bar happened to end. Prices with a lot of traded volume behind them tend to attract price back; thin areas tend to be crossed quickly.
A bar with far more volume than the bars around it.
The observation is unambiguous and the conclusion is not. A spike marks a price area where an unusual amount of business was transacted — which is exactly what you would expect both at the start of a move and at the end of one. What separates the two is what price does afterwards, which means a spike is confirmable only in hindsight and is never a signal on its own.
Two readings of the same thing
Initiation
New participants are entering and the move is beginning. More likely when the bar closes near its extreme and the next bars hold the new ground.
Climax
The last participants are entering and the move is ending. More likely when the bar has a long wick, closes back in its own range, or arrives after an extended run.
A one-time or monthly charge to turn a passed evaluation into a funded account.
Charged after you pass, sometimes with a choice between a larger one-off payment and a smaller recurring one. It is frequently omitted from the advertised cost of an evaluation and changes the real break-even meaningfully.
Breaking a hard account rule, which ends the account immediately.
Distinct from a soft violation such as a consistency shortfall, which only delays a payout. Breaches are usually enforced by the platform in real time: the position is liquidated and the account is closed the moment the threshold is crossed, with no opportunity to recover within the same session.
A cap on how much of your total profit may come from a single day.
Typically expressed as a percentage: if your best day accounts for more than that share of total profit, a payout is delayed or refused until the rest of the account catches up. It is a payout gate, not a trading rule — it does not breach the account, and it is separate from the profit target. The practical effect is that one enormous day can be worse than several ordinary ones, which is the opposite of how most strategies are optimised.
The most you may lose in one trading day before the account is locked or breached.
A per-day cap, reset at the daily boundary. It is a separate rule from the maximum drawdown and both are live at the same time — you can be comfortably inside your overall drawdown and still fail the day. Firms differ on whether the limit is measured on closed P&L only or on equity including open positions, and on whether hitting it locks you out for the day or ends the account outright.
Stopping for the day once a set profit is reached, to protect it from later trades.
Usually self-imposed rather than firm-mandated, and implemented in a strategy as a flag that blocks new entries for the remainder of the session. Its value on a trailing-drawdown account is specific: on an equity-trailed account, giving back an intraday gain can leave the floor raised while the balance falls, so protecting the day's profit also protects the distance to the floor.
A drawdown floor recalculated only once per day, from the closing balance.
The floor moves at the daily settlement rather than tick by tick. In practice this is far more forgiving than an intraday trail: a position that goes deeply against you and recovers before the close never touches the floor, because the floor never saw the excursion. It is the single most important thing to establish about an account before sizing a strategy, and the two are routinely confused because both get called 'trailing drawdown'.
The simulated trading test you must pass to be given a funded account.
A paid, simulated account with a profit target and a set of risk rules. Passing converts it to a funded account, which may be simulated as well, with payouts drawn from the firm rather than from a live exchange position. The distinction between simulated-funded and live-funded is rarely advertised and is worth checking before choosing a firm.
The daily deadline by which all positions must be closed.
Positions still open at flat time are liquidated by the firm, usually with a fee and sometimes as a rule violation. Any automated strategy needs an explicit flatten before that time — relying on the firm's forced liquidation means accepting its fill and its penalty. This is why every strategy sold here ships with an end-of-day flatten built in rather than as an option.
The highest value an account has reached, used as the anchor for a trailing floor.
Whether it is recorded on equity or on closed balance is the single detail that determines how punishing a trailing drawdown is. On an equity high-water mark, an open trade that goes 20 handles your way and then comes back has permanently raised your floor without ever adding to your balance.
The number of separate days you must trade before an evaluation can pass.
Prevents passing on a single lucky session. Firms differ on what counts as a trading day — some require any fill, some require a minimum volume or a minimum P&L movement. An automated strategy that only trades a specific setup can sit idle for days and fail this requirement while being profitable.
A window around scheduled economic releases in which trading is restricted.
More common at forex-oriented firms than at futures firms, and the restrictions vary from a full ban to a prohibition on opening new positions. Where it applies, an automated strategy has to know the calendar, which most do not — this is a common way for an otherwise compliant strategy to violate rules.
Withdrawing profit from a funded account, subject to the firm's schedule and gates.
Gated by several rules at once: a minimum profit balance, a minimum number of trading days since the last payout, the consistency rule, and often a requirement to leave a buffer above the drawdown floor. The headline profit split is the least interesting number here — the gates decide when you actually see money.
The share of funded-account profit you keep, with the rest going to the firm.
Commonly 80–90% to the trader, sometimes 100% on a first tranche. Compare it against the total cost of getting and keeping the account — monthly fees, activation fees and resets — rather than in isolation.
The profit needed to pass an evaluation and move to a funded account.
A fixed dollar figure tied to the account size. It is only half the pass condition — the other half is not having breached any drawdown, daily loss or minimum-trading-day rule on the way there. Reaching the target quickly with one outsized day frequently trips a consistency rule instead of passing.
Paying to restart a failed or breached evaluation from the beginning.
Cheaper than buying a new evaluation at most firms, and the main recurring cost of a strategy that breaches often. It is the number worth comparing a monthly tool subscription against — a strategy that prevents one reset a month has paid for a good deal of tooling.
A cap on contract size that increases as the account's profit grows.
Early in an account you may be limited to a fraction of the account's headline maximum, with the full size unlocked at profit milestones. This matters for automation more than for discretionary trading: a strategy backtested at full size will not reproduce those results while the plan restricts it, and exceeding the cap is usually an immediate violation rather than a warning.
A loss floor that rises with your account's high-water mark and never falls back.
You start with a floor a fixed distance below your starting balance. As the account makes new highs the floor follows at that same distance; when the account falls, the floor stays where it is. Two details decide whether a strategy survives it. First, whether the high-water mark is measured on equity — including open, unrealised profit — or only on closed balance: an equity-based trail can ratchet upward on a trade you have not exited yet, and stay there after you give the profit back. Second, whether it stops trailing once you are a set amount above your start. Both vary by firm and by account.
Running a strategy over historical data to estimate how it would have performed.
An estimate whose quality depends entirely on its assumptions: fill model, commission, slippage, whether the data includes the overnight session, and whether the logic could have known at the time what it appears to know. Most backtests that look extraordinary are wrong for one of those reasons rather than the market having been extraordinary.
The broker and exchange cost of a trade, usually quoted per round turn.
A round turn is one entry plus one exit on one contract. Prop-firm accounts often carry higher per-contract costs than a retail account, and a high-frequency strategy can be profitable gross and unprofitable net. Model it in the backtest rather than subtracting it afterwards — it changes which trades are winners, not just the total.
The decline from an equity peak to the following trough.
In a backtest this is usually reported on closed trades, which understates what you would have lived through. Firm drawdown rules measure something different again — a floor relative to a high-water mark, not a peak-to-trough distance — so a strategy's reported maximum drawdown is not directly comparable to an account's drawdown allowance.
The average profit or loss per trade, in dollars or in R.
(Win rate x average win) − (loss rate x average loss). It is the number that actually matters, because it is the one you multiply by trade frequency to get an expected return. Expectancy stated in R rather than dollars survives changes in position size, which makes it the right unit for comparing two strategies.
The furthest a trade went against you before it closed.
The most useful single statistic for prop-firm work, and the one most backtest summaries omit. A strategy's reported drawdown is measured on closed trades; MAE measures the unrealised low points along the way — which is exactly what an intraday trailing drawdown reacts to. Two strategies with identical equity curves can have completely different MAE profiles, and only one of them will survive a $2,000 trailing floor.
The furthest a trade went in your favour before it closed.
The counterpart to MAE. Comparing MFE against actual realised profit tells you how much of each move the exit gave back, which is the evidence for or against changing a target. Large MFE with small realised profit is the signature of exits that are too tight, not of entries that are wrong.
Tuning a strategy until it describes the past instead of the market.
Every added parameter and every re-run of an optimiser makes the historical result better and the future result worse. The tell is fragility: change the timeframe slightly, shift the session by fifteen minutes, or move the sample forward six months, and a genuinely fitted strategy falls apart while a real edge merely gets worse.
Deciding how many contracts to trade, given the stop distance and the risk budget.
Risk per trade divided by (stop distance in ticks x tick value) gives the contract count. On a prop account the risk budget is not a preference — it is set by the distance to the drawdown floor and the daily loss limit, whichever binds first, and it shrinks as you approach either.
Above 1.0 means the strategy made money over the sample. It says nothing about how the money arrived: a profit factor of 2.0 built from one enormous winner is a different object from the same figure built from four hundred small ones, and only one of them is likely to repeat. Always read it next to the trade count and the largest single win.
A trade's result expressed as a multiple of the amount risked on it.
If you risked $200 and made $600, that is +3R. Working in R rather than dollars makes results comparable across account sizes and contract counts, and it is the only sane unit for a prop-firm account where the permitted size changes as you progress through a scaling plan.
The profit or loss locked in by positions you have already closed.
Balance-based rules measure this and ignore whatever an open position is doing. Backtest summaries are almost always reported on realised P&L, which is exactly why they hide the intraday excursions that an equity-based trailing drawdown reacts to.
A strategy with 40 trades has told you almost nothing, however good the numbers look. Small samples are where curve-fitting hides, and where a single outlier carries the whole result. Look for a few hundred trades across more than one market regime before treating a figure as an estimate of anything.
The difference between the price you expected and the price you were filled at.
Backtests default to optimistic fills. On futures the realistic assumption for a market order in a liquid contract is a tick or so, more at the open, around releases and in the illiquid overnight session. A strategy with a small average win is far more sensitive to this than one with a large average win — at a $12.50 average edge, one tick of slippage on MES is most of it.
The profit or loss on positions you still hold and have not closed.
It moves with every tick and it is not yours until the position is closed. Whether a firm's drawdown and daily-loss rules watch equity (which includes this) or balance (which does not) is the difference between a rule that reacts to every excursion and one that only sees your exits.
Optimising on one stretch of data and testing on the next, repeatedly.
Parameters are chosen using only data available at the time, then evaluated on the period that follows, and the window rolls forward. It is more work than a single backtest and it is the only common method that produces a number you can reasonably expect to see again, because every result it reports was produced out-of-sample.
On its own it is close to meaningless — a strategy can win 90% of the time and lose money, or win 30% and be excellent. It only becomes informative paired with the average win-to-loss ratio. High win rates in particular deserve suspicion, because the usual way to manufacture one is a distant stop, which hides the risk rather than removing it.
A URL TradingView posts your alert message to when a strategy signal fires.
The bridge between a chart and a broker. TradingView sends a JSON payload to an execution service, which places the order with the broker where the prop-firm account lives. Every hop adds latency and a way to fail silently, so the message format and the receiving service's error handling matter as much as the strategy.
The average size of recent bars, including any gap from the prior close.
A volatility measure, not a direction one. Its main use is making stops and targets adapt: a fixed 10-point stop is loose in a quiet session and tight in a fast one, whereas 1.5x ATR is roughly the same amount of room in both. On a drawdown-constrained account that consistency is worth more than the extra parameter costs.
Acting only on a bar's final values, once that bar has closed.
The condition is evaluated on the confirmed bar and the order goes in at the next bar's open. It gives up part of the move in exchange for a signal that cannot change afterwards, and it is the only way to make a live strategy match its backtest. Every strategy sold here is bar-close by default for that reason.
On a prop account this is bounded by three things at once: the account's maximum, the scaling plan's current allowance, and what the distance to the drawdown floor actually permits. The last is usually the binding one and is the one no rulebook tells you.
Logic that stops a strategy trading for the rest of the day once a loss cap is hit.
Tracks the session's realised P&L and blocks new entries past a threshold, usually flattening whatever is open. It is what keeps one bad session from turning into a breach, and it has to live inside the strategy rather than in a mental note — an automated system will keep taking its setup all day otherwise.
Logic that uses information which was not available at the time it acts.
Distinct from repainting, though they often occur together: repainting is a display problem, lookahead is a logic problem. In Pine the classic sources are barmerge.lookahead_on and any comparison against a higher-timeframe value that has not finished forming. The symptom is a backtest that looks close to perfect.
The rate at which price is moving, rather than the level it has reached.
Implemented as a moving-average relationship, a rate-of-change calculation, or an oscillator. All momentum measures lag by construction — they are computed from bars that have already closed — so the practical question is never whether they lag but whether the lag is short enough to be worth the noise reduction it buys.
The high and low of the first stretch of a session, used as breakout levels.
Commonly the first 5, 15 or 30 minutes. It gives a strategy two unambiguous levels that need no parameters and no fitting, which is why opening-range logic tends to hold up across markets better than most. It also naturally limits how many signals a session can produce — useful on an account with a daily loss limit.
The dollar value of one full point of price movement in a contract.
Tick value multiplied by the number of ticks in a point. MNQ is $2 per point, NQ is $20, MES is $5, ES is $50. Pine's strategy tester needs this to be right or every dollar figure it reports is wrong by a constant factor while still looking entirely plausible.
An indicator or signal that changes after the bar it appeared on has closed.
Historical bars show the signal in one place; live, it appears somewhere else or vanishes. The usual causes are calculations on the developing bar, request.security without proper offsetting, and functions that look at a bar's final values before it has one. A repainting backtest is not a pessimistic estimate — it is a description of trades that were never available.
The block of hours a strategy is allowed to trade.
RTH is the regular cash session; ETH covers the overnight. They behave differently enough that a strategy tuned on one usually fails on the other — overnight liquidity is thinner, spreads wider and slippage larger. A backtest that silently includes ETH will show trades that would have been expensive or impossible.
An ATR-based trailing line that flips between above and below price.
The line sits a multiple of ATR away from a midpoint and only ever moves in the direction of the current state until price closes through it, at which point it flips. It produces one unambiguous state at a time, which makes it well suited to automation — there is no interpretation step between the indicator and the order.
Price areas where earlier trading has repeatedly stopped a move.
Prior swing highs and lows, session opens, and the previous day's high, low and close. They are areas rather than lines, and their usefulness comes from being widely watched rather than from anything intrinsic. A candle pattern occurring at one of these is the context that most pattern descriptions quietly assume.
The dollar value of the smallest price increment a contract can move.
MES moves in 0.25-point increments worth $1.25; ES moves in the same increments worth $12.50. Tick value converts a stop measured in points into a stop measured in dollars, which is the only form a drawdown rule cares about. Getting it wrong by a factor of ten between a micro and its full-size parent is the most common sizing error there is.
A service that receives TradingView alerts and places the orders with your broker.
The common middle layer for TradingView-to-futures automation: the strategy fires an alert, TradersPost translates it into a broker order at Tradovate or Rithmic, where the prop-firm account sits. It also handles position reconciliation, which matters more than it sounds — a missed exit leaves the strategy and the broker disagreeing about what is open.
The session's average traded price, weighted by volume at each price.
Anchored to the session start and reset each day. It is widely watched, which is most of why it works as a reference: it is where a large participant measuring execution quality is benchmarked. Distance from VWAP is used both as a mean-reversion signal and as a trend filter, and those two uses want opposite settings.