Trendlines that update themselves

TDU Auto Trendline finds the ascending and descending trendlines that price is actually respecting, counts every touch, keeps score of how fresh each line is, and marks the exact bar where price breaks out or bounces. No dragging, no re-anchoring, no hindsight. The lines are built from confirmed swing pivots by rule, so the chart you look at today is the chart you would have looked at in real time.

What is the TDU Auto Trendline indicator?

TDU Auto Trendline is a premium chart study for SabrTrader that automates the single most-used and most-abused tool in discretionary trading: the diagonal trendline. It scans the price series for confirmed swing pivots, connects the pivots that form valid sloped support and resistance, validates each candidate line by counting how many times price actually came back and respected it, and then keeps that line alive on your chart — extending it bar by bar — until price breaks it.

Two families of lines are produced. Up trendlines are ascending lines built from rising swing lows; they act as dynamic support underneath an advance. Down trendlines are descending lines built from falling swing highs; they act as dynamic resistance above a decline. Both families can exist simultaneously, which is how the indicator naturally reveals wedges, triangles, channels and coils without you ever labelling a pattern by hand.

Beyond drawing, the indicator publishes eight machine-readable plots: BreakOut, Bounce, Up Trendline Price, Down TrendLine Price, Up Trendline touches, Down TrendLine touches, Up Trendline last touch bars ago and Down TrendLine last touch bars ago. Those series turn a visual tool into data you can reference for alerts, scanning and automation inside SabrTrader — you can ask questions like "is price within four ticks of an ascending line that has already been touched five times and was last touched within the past twenty bars?" and get a numeric answer on every bar.

It is not a separate add-on purchase. TDU Auto Trendline ships inside the premium indicator suite and is included in the Pro, Ultimate and Lifetime plans alongside the rest of the toolkit. Install SabrTrader, sign in, and the study is in your indicator list.

Why hand-drawn trendlines quietly ruin good trading

Everyone draws trendlines. Almost nobody draws them consistently. Ask five traders to mark the ascending support on the same chart and you get five different slopes, because the tool has three degrees of freedom — which low you anchor to, which second low you connect to, and whether you use wicks or bodies. Every one of those choices is made after you already have an opinion about direction, which is precisely when your judgement is least reliable.

The second problem is memory. A hand-drawn line is a snapshot. When price breaks it, most traders delete it, and with it goes the single most valuable piece of information on the chart: where the broken line now sits, because broken support becomes resistance on the retest. The third problem is maintenance. On a 15-second or one-minute chart, structure rotates every few hours. Nobody redraws twenty lines across eight symbols before the open, and if they do, they stop by Wednesday.

TDU Auto Trendline removes all three failure modes. The anchors are chosen by a deterministic pivot rule you configure once. A line is only promoted to the chart once it has been tested the minimum number of times you demand. Broken lines are optionally retained in a neutral grey so retests stay visible. And maintenance is zero: change symbol, change timeframe, change session — the lines rebuild themselves in the time it takes the chart to load.

The result is subtle but important for your P&L: your trendline reads become comparable. When every line on every chart is built by the same rule, a five-touch line on Nasdaq futures means the same thing as a five-touch line on BTC/USDC, and you can build genuine statistics about how you trade breaks and bounces instead of arguing with your own screenshots.

How does the indicator decide where a trendline goes?

The pipeline has four stages, and each stage is exposed to you as a parameter so you can make the engine strict or permissive.

Stage 1 — pivot detection with StrengthLeft and StrengthRight

Everything starts from swing pivots. A swing low qualifies when it is the lowest price across a window of StrengthLeft bars to its left and StrengthRight bars to its right; a swing high is the mirror image. Both default to 10, meaning a pivot must dominate roughly twenty bars of price action before the engine takes it seriously. That is deliberately demanding: it filters out the micro-wiggles that make hand-drawn lines so noisy, and it means the pivots you get are the ones a human eye would also call "the swing".

Because the right-hand window must fill before a pivot can be confirmed, a brand new low is not treated as an anchor until StrengthRight bars have printed after it. This is the honest trade-off of every pivot engine, and it is the reason the tool does not flatter you with lines that magically appear underneath candles after the fact. If you want faster structure, lower the strengths; if you want the big skeleton only, raise them. Asymmetric settings are perfectly valid — a larger StrengthLeft with a smaller StrengthRight confirms pivots sooner while still requiring meaningful context behind them.

Stage 2 — candidate construction

Confirmed pivot lows are paired to generate ascending candidates; confirmed pivot highs are paired to generate descending candidates. Each pair defines a slope and an origin. The segment between the two anchor pivots is the start segment — the leg that created the line — and it is styled separately (dotted, 50% opacity by default) so you can instantly see where a line was born versus where it is being actively defended.

Stage 3 — validation by touches

A slope is only interesting if price keeps returning to it. The engine walks forward from the origin and counts touches: bars whose price comes to the line within a tolerance of MarginInTicks. Only candidates that reach MinimumTouchPoints are promoted to the chart. The default of 3 is the classic technician's standard: two points make a line, the third makes a trend. Raise it to 4 or 5 on liquid index futures and crypto majors when you only want the lines the whole market can see; drop it to 2 on slow instruments or very high timeframes where three tests may take weeks to accumulate.

Stage 4 — lifecycle management

A validated line is extended forward on every new bar and continuously re-evaluated. It stays active while price respects it, is marked broken when the break rule fires, and can be retained on the chart as inactive grey structure via ShowBrokenTrendLines. Meanwhile the line's current value is published as Up Trendline Price / Down TrendLine Price, its test count as Up/Down Trendline touches, and its freshness as Up/Down Trendline last touch bars ago.

How do I read TDU Auto Trendline on the chart?

The visual language is small on purpose. Once you learn five elements you can read any chart the indicator touches in under two seconds.

• Green solid line (34,139,34) — an active ascending trendline. Support. Price is above it and has been respecting it.
• Crimson solid line (220,20,60) — an active descending trendline. Resistance. Price is below it.
• Grey solid line (128,128,128) — an inactive or broken line. Historical structure retained for retest reference.
• Dotted, half-opacity segment — the start leg between the two original anchor pivots, in green, crimson or grey depending on the line's state.
• Labels — the line's live price (ShowPrice) and its touch count (ShowTouchCount), rendered in your chosen font family and size.
• Markers — ▲ and ▼ for breakouts, ★ for bounces, drawn at the bar where the event was confirmed.

Read the chart in this order. First, count the colours: green only means an uptrend with intact support; crimson only means a downtrend with intact resistance; both present means compression and you should expect an expansion event. Second, read the touch counts: the highest number on the chart is the line the market cares about most. Third, read the freshness: a line last touched three bars ago is live structure; a line last touched two hundred bars ago is a museum piece that price may not even remember. Fourth, look for markers: the ▲/▼ tells you an agreement has just failed, the ★ tells you an agreement has just been reaffirmed.

Active, inactive and broken: why three states matter

Most auto-trendline tools have one state: drawn. TDU Auto Trendline separates the life of a line into stages and gives each stage its own colour, width, style and opacity, because each stage means something different for your risk.

An active line is tradeable structure. You can lean on it: entries near it, stops beyond it, targets measured from it. An inactive or broken line, drawn grey, is no longer support or resistance in its original sense — it is a polarity flip candidate. Broken ascending support routinely becomes resistance on the first pullback; broken descending resistance routinely becomes support. Traders who delete broken lines throw away the highest-probability retest level on the chart. Keeping them grey solves the problem elegantly: they are present but visually subordinate, so they never compete with live structure for your attention.

The start segment deserves special mention. Because it is dotted and half-transparent by default, you can distinguish the leg that created the line from the extension that is being tested. That matters when judging quality. A line whose start leg is a steep two-bar spike and whose extension has three shallow touches is a different animal from a line whose start leg spans eighty bars of orderly higher lows. If you want the origin to disappear entirely, set the start opacity to zero; if you want to study anchors, raise it to 100 and switch the style to solid.

Touch counts and "last touch bars ago": the quality score you never had

The two numeric families that make this indicator genuinely different are touches and last touch bars ago. Together they form a two-dimensional quality score for every line on your chart.

What does the touch count actually tell you?

Each touch is a moment where price came to the slope and a decision was made in favour of it. More touches mean more participants have transacted using that line as reference, which has two consequences that seem contradictory but are not. First, a heavily touched line is more likely to hold on any single test, because the reference is widely shared. Second, when it finally fails, the move is more violent, because every participant leaning on that line is wrong at the same moment and their stops sit in the same pocket. This is why the touch count is your position-sizing dial as much as your entry filter: three touches is a normal bounce trade, seven touches is either the best bounce of the week or the cleanest break of the month.

What does "last touch bars ago" tell you?

Freshness. A line touched two bars ago is being actively defended right now; the participants who defended it are still in the trade and still have live stops. A line untouched for three hundred bars may be geometrically valid but psychologically forgotten — the market has moved on, the slope may no longer match the prevailing volatility, and the reaction on the next test is far less reliable. Use the plot to build a decay filter: for intraday scalping, treat lines with a last touch beyond a few hundred bars as context only, not as trade triggers. For swing charts, the acceptable staleness window is much wider.

Reading the two numbers together

• High touches, low bars ago — the premium condition. Live, crowded structure. Best bounces and best breaks both come from here.
• High touches, high bars ago — a famous old line. Worth marking, but wait for confirmation before leaning on it.
• Low touches, low bars ago — young structure forming right now. Treat as provisional; a break is not very informative.
• Low touches, high bars ago — noise. Consider raising MinimumTouchPoints so lines like this never reach your chart.

Breakouts: how the indicator confirms that a line has failed

A break is where trendline trading makes and loses most of its money, so the confirmation logic is explicit and fully under your control.

MarginInTicks defines a tolerance band around the line. At the default of 0 the line is honoured literally. Raise it and price must clear the line by that many ticks before the engine counts a violation — the single most effective filter against one-tick wick pokes on thin instruments and against exchange-level noise in crypto. TrendLineBrokenBars defines how much persistence you demand: the default of 1 flags the break on the first qualifying bar, while values of 2 or 3 require the violation to hold across consecutive bars before the line is retired. Increase it when you are tired of being faked out; keep it at 1 when you are trading momentum and cannot afford a two-bar delay.

When the rule fires, ShowBreakOuts paints the marker — ▲ for an upside break through a descending line (BreakOutUpMarker) and ▼ for a downside break through an ascending line (BreakOutDownMarker) — and the BreakOut plot flags that bar so alerts and automation can react. Turn on EnableBreakoutAlert and the platform plays BreakoutAlertSound (default alert-ding.wav) so you can watch eight charts with your eyes and one with your ears.

Why a break is not automatically a trade

Broken trendlines produce two very different outcomes and the difference is context, not geometry. A downside break of a shallow, three-touch ascending line inside a broader uptrend is usually just a change of slope — price rotates, builds a new flatter line, and continues higher. A downside break of a steep, seven-touch line that coincides with a failed swing high and a rejection at the value area high on your Volume Profile is a genuine trend transition. Use the touch count, the freshness plot and the state of your Market Structure read to decide which one you are looking at before the marker prints, not after.

Bounces: trading the line instead of the break

The mirror event to a break is a bounce. When price trades into an active line within the margin tolerance and is rejected rather than absorbed, the engine confirms a bounce, increments the touch count, paints ★ using BounceUpMarker or BounceDownMarker (both default to a star, distinguishable by which line they sit on and by their marker styling), sets the Bounce plot, and optionally fires BounceAlertSound when EnableBounceAlert is on.

Bounce trading is statistically the friendlier side of trendline work because your invalidation is tight and unambiguous: you are wrong if the line gives way. That converts a diagonal into an actual risk unit. The Up Trendline Price and Down TrendLine Price plots give you the exact numeric value of the line on the current bar, so you can place stops a fixed distance beyond it instead of eyeballing pixels — and because the line's value changes every bar, a stop referenced to it is a trailing stop that follows the slope of the trend for free.

ShowBounces controls the markers. Leave them on while you are learning the engine's tolerance: after a few sessions you will develop a feel for how close price has to come, at your chosen margin, before a touch is registered. Then decide whether to keep them or declutter.

The eight plots and what to do with them

Every visual element has a numeric counterpart. This is what separates a premium study from a drawing script.

Two derived measurements are worth building yourself. Distance to line = current price minus Up Trendline Price (or Down TrendLine Price minus price) tells you objectively whether you are entering at the edge of structure or chasing into the middle of a range. Channel width = Down TrendLine Price minus Up Trendline Price gives you a squeeze meter: when the width contracts toward the size of an average bar, an expansion is imminent and you should be positioned for a break rather than a fade.

How to trade it: five concrete playbooks

These are complete, rule-based setups built only from what the indicator actually publishes. Trade them on a simulator first, log the touch count and freshness of every line you use, and you will quickly discover which variant suits your instrument.

Playbook 1 — The Third-Touch Bounce

Context: an active green ascending line with touches ≥ 3 and last touch bars ago inside your freshness window. Higher swing highs on the chart, no crimson line immediately overhead.

Trigger: price trades into the line within the margin and the ★ bounce marker confirms, or a bar closes back above Up Trendline Price after dipping to it.

Entry: at or just above the close of the confirming bar. Stop: below Up Trendline Price minus your margin, plus a small buffer for spread — because the plot updates each bar, the stop trails up with the slope. Targets: the prior swing high first, then the value of the opposing Down TrendLine Price if a descending line exists.

Invalidation: a confirmed ▼ breakout on the same line, or two consecutive closes below the line at your TrendLineBrokenBars setting. Do not average down into a breaking trendline; the whole point of the tool is that it tells you the agreement has ended.

Playbook 2 — The Crowded-Line Break

Context: a line with an unusually high touch count for your instrument — typically 5 or more — and a recent last touch. Everyone is leaning on it and stops are stacked just beyond.

Trigger: the ▲ or ▼ marker prints with your margin and persistence filters satisfied, ideally on expanding range and volume.

Entry: on the confirmation bar's close, or on the first shallow pullback that does not re-enter the old line's territory. Stop: back on the far side of the broken line, now drawn grey — if price reclaims it, the break was a trap. Targets: measured move equal to the height of the coil at the point of the break; second target at the next horizontal reference from your Volume Profile.

Invalidation: a close back through the grey broken line within the next few bars. This is the cleanest invalidation in trendline trading because it is a line, not a guess.

Playbook 3 — The Failed Break Reclaim (trap reversal)

Context: a break marker fired, the line went grey, and price failed to follow through — it is now pushing straight back through the broken line.

Trigger: price closes back on the original side of the grey line with conviction, after the break marker printed within the last handful of bars.

Entry: on the reclaim close. Stop: beyond the extreme of the failed break. Targets: the opposite boundary of the structure — often the opposing active line's price. This setup is where the ShowBrokenTrendLines option earns its keep: without retained grey lines you cannot even see the reclaim.

Invalidation: failure to hold the reclaim, i.e. an immediate second violation. Cut, and stand down on that line for the session — twice-broken structure rarely gets a third chance.

Playbook 4 — The Squeeze Expansion

Context: an active green ascending line and an active crimson descending line converging. Channel width (Down TrendLine Price minus Up Trendline Price) contracting toward one or two average bar ranges. Both lines reasonably fresh.

Trigger: the first confirmed break marker of either line.

Entry: in the direction of the break. Stop: the mid-point of the coil, or the opposite line's value — whichever gives you a risk you can actually take. Targets: the widest point of the coil projected from the break, taken in two pieces.

Invalidation: a break that reverses through the coil and triggers the opposite line's break marker. When a squeeze breaks both ways, the structure is finished; switch to horizontal references and wait for new pivots to form.

Playbook 5 — The Polarity Retest

Context: a previously broken line, now grey, with a high historical touch count. Price is drifting back toward it from the other side.

Trigger: first tag of the grey line from the new side with a rejection bar, ideally aligning with a footprint imbalance stack or absorption on your Footprint chart.

Entry: on the rejection close. Stop: a fixed distance beyond the grey line — if price passes cleanly back through, polarity did not flip. Targets: the swing that produced the original break, then trend continuation.

Invalidation: a close through the grey line. Note that when polarity holds, a fresh line of the opposite family will often form off the very pivot your retest created — which is exactly how the engine hands you the next trade.

How do I set up TDU Auto Trendline in SabrTrader?

1. Install the platform. Download SabrTrader for Windows and run the installer. It sits alongside your existing tools and does not require you to move brokers to evaluate it.
2. Sign in and activate a plan. TDU Auto Trendline is unlocked with Pro, Ultimate or Lifetime — see pricing. There is no separate indicator purchase and no per-indicator licence to manage.
3. Open a chart of the instrument you actually trade — index futures, a single stock, or a crypto pair such as BTC/USDC — on the timeframe you actually trade.
4. Add the indicator from the study list and accept the defaults for the first session: strengths 10/10, MinimumTouchPoints 3, MarginInTicks 0, TrendLineBrokenBars 1, breakouts and bounces visible.
5. Calibrate the strengths. If the chart looks crowded, raise StrengthLeft and StrengthRight together until only the lines you would have drawn yourself remain. If structure feels missing, lower them in steps of two.
6. Calibrate the margin. Scroll back through a few hundred bars and look for break markers that were obviously single-wick pokes. Raise MarginInTicks until those disappear without erasing the genuine breaks.
7. Decide your persistence. Momentum traders keep TrendLineBrokenBars at 1. Swing and mean-reversion traders usually prefer 2 or 3.
8. Style for your theme. Set widths, opacities and the start-segment style so active lines dominate and grey lines recede. Set MarkerFontSize and TextFontSize so labels are readable at your normal zoom without covering candles.
9. Turn on alerts last. Enable EnableBreakoutAlert and EnableBounceAlert only after the visual settings feel right, otherwise you will train yourself to react to noise.
10. Save as a template and apply it across your watchlist. From here, maintenance is zero.

Complete settings reference

Structure detection

Events, markers and alerts

Labels

Ascending (up) line styling

Descending (down) line styling

Works on every market you trade

Trendlines are pure price geometry, so the engine is instrument-agnostic. What changes between markets is tick size, session structure and how much wick noise you have to tolerate. Use this as a starting grid, then calibrate on your own data.

Two market-specific notes. In crypto, 24/7 trading means no session gaps, so lines can survive for very long stretches — lean on the freshness plots harder than you would on futures, and be generous with MarginInTicks because wick excursions are larger relative to structure. In stocks, overnight gaps regularly jump straight over a line without ever touching it; keep ShowBrokenTrendLines on so you can trade the retest of a gapped-through line, which is one of the most reliable applications of the whole tool.

How it compares

Auto-trendline scripts are common. Auto-trendline engines with state, counting and data output are not.

Combining Auto Trendline with the rest of the suite

Diagonals answer "is the trend's slope intact?" They do not answer "where is value?" or "who is actually transacting?" The strongest reads come from stacking a diagonal against a horizontal and an order-flow confirmation.

With Volume Profile

The highest-quality bounce is one where an active green line intersects a well-defined value area low or a high-volume node from your Volume Profile. Diagonal support plus horizontal acceptance means two independent groups of participants have reason to defend the same price. Conversely, a break of a crowded line that lands in a low-volume pocket tends to travel fast, because there is nothing beneath it to slow the move — ideal conditions for Playbook 2.

With the Footprint chart

Use the Footprint to grade the touch. Absorption at the line — heavy selling into it with no price progress — is what a real bounce looks like from the inside. Thin, unopposed prints through the line are what a real break looks like. The trendline tells you where to look; the footprint tells you whether it worked.

With Market Structure

The Market Structure indicator classifies swings and shifts. Align the two: a break of ascending support that also coincides with a structural break of a prior swing low is a trend change; a break of ascending support with structure still making higher lows is usually just a slope reset. This single filter eliminates a large share of losing trendline breakout trades.

With TDU Price Action

The TDU Price Action study supplies bar-level confirmation — the rejection candles, engulfs and exhaustion signatures that make a touch actionable. Combine it with the ★ bounce marker and you have both the level and the trigger in one glance.

Tuning guide and the mistakes to avoid

Why is my chart cluttered with lines?

Almost always because the strengths are too low for the timeframe, or MinimumTouchPoints is set to 2. Raise StrengthLeft and StrengthRight together in steps of two until the remaining lines match the ones you would have drawn by hand, then raise the touch minimum to 4 if the chart is still busy. Reducing UpInactiveTrendOpacity and DownInactiveTrendOpacity also helps: it keeps broken structure available for retests while removing it from your foreground attention.

Why do I get breakouts that immediately reverse?

Two causes. MarginInTicks at 0 on an instrument with wide wicks means a single tick of overshoot retires a valid line — raise the margin until obvious wick pokes stop registering. And TrendLineBrokenBars at 1 flags on the first bar — move to 2 or 3 if you would rather miss the first leg than take three fakeouts a day. Then remember the context rule: breaks of shallow, low-touch lines are rotations, not reversals.

Why did a line disappear?

Because it broke. If you want a record of it, set ShowBrokenTrendLines to True so the line survives in the inactive style. If you never want to see history at all, set it to False and accept that you lose polarity retests.

Why does a new line take a while to appear?

Anchors require StrengthRight bars of confirmation, and validated lines require MinimumTouchPoints tests. That delay is the price of not being lied to. If you need earlier structure, lower StrengthRight specifically while keeping StrengthLeft high — you will confirm pivots sooner while still requiring real context behind them.

Common mistakes

• Trading every marker. The markers are events, not signals. Filter by touch count, freshness and trend context.
• Ignoring the touch count. A 2-touch line and a 7-touch line deserve different position sizes and different expectations.
• Using one setting everywhere. A daily stock chart and a one-minute futures chart need different strengths and margins. Save separate templates.
• Deleting broken lines. The retest of grey structure is frequently the better trade than the break itself.
• Fighting the slope. If a green line keeps holding, stop looking for shorts underneath it. The tool is telling you who is in control.
• Alerts before calibration. Enable sounds only once the visual settings produce lines you would actually trade.

Workflow: how a session with self-updating trendlines actually looks

Before the open you load your template across the watchlist. In ten seconds per chart you note three things: which family of lines is active, the highest touch count on the chart, and whether any green and crimson pair is converging. That gives you a bias and a list of the two or three levels that matter, without drawing a single line.

During the session you keep breakout alerts on for the compression charts and bounce alerts on for the trending charts. When a ★ fires on a fresh, high-touch line, you check the footprint for absorption and take the bounce with a stop referenced to the line's plotted price. When a ▲ or ▼ fires on a crowded line, you check whether structure agrees before you commit. Everything else you ignore.

After the close you review. Because the lines are deterministic, your screenshots are honest: the line you traded is still exactly where it was, with the same touch count it had at the moment of entry. That is the difference between a journal that improves you and a journal that flatters you — and it is the quietest, most underrated benefit of automating your trendlines.

Frequently asked questions