Event, State and Position: A Framework for Timing Trade Entries

Video: THIS TELLS ME EXACTLY WHEN TO ENTER
Creator: Oliver Velez Trading
Original Video: Watch the original video
Difficulty: Advanced

The Big Idea

Oliver Velez combines three questions before entering: What event formed? What state is the market in? Where is the event relative to that state? A powerful-looking candle alone is incomplete. Its meaning changes when it appears just above a compressed area versus far above an already extended move.

The source uses 20- and 200-period simple moving averages, primarily on two-minute charts, to organize those decisions. This guide separates the visual framework from its unverified performance claims, then tests one clearly defined version: a completed bottoming-tail bar just above a flat, narrow state.

If You Remember One Thing

Identify the context before the event, and distinguish a forming candle from a completed one. A historical candle’s final appearance cannot tell you exactly what it looked like when a discretionary trader acted inside it.

The Three Pillars

The transcript promotes high consistency and daily profitability, but these are not outcomes established by the supplied examples. A coherent classification system helps define research; it does not itself prove an edge.

Timeframe: A Choice of Horizon, Not a Promise of Income

Velez prefers two-minute and five-minute charts for his intraday approach, typically two minutes near the start and end of the day and five minutes if trading midday. These are his preferences, not universally superior timeframes.

His eight-bar description is an asserted average trade duration, not a measured distribution provided here. Eight two-minute bars represent 16 minutes; eight five-minute bars represent 40 minutes. Bars held, elapsed time and execution opportunities can differ around session boundaries or missing data.

On a two-minute chart, the 20 and 200 averages summarize approximately 40 and 400 minutes of included bars. Changing to five minutes changes those horizons to 100 and 1,000 minutes. A comparison must acknowledge that the economic context changes, not just the screen’s appearance.

Know the Event

Elephant Bars

The source describes a directional candle larger than recent bars, roughly the preceding ten, with most of its range occupied by the body. Both bullish and bearish versions are used.

A possible research definition would compare the completed body’s size with prior bodies and require a minimum body-to-range ratio. The source does not give those exact thresholds. A large bar can reflect many causes; OHLC candles alone do not establish that multiple institutions acted simultaneously. The presenter’s institutional-flow explanation is an interpretation, not participant identification.

Tail Bars

A bottoming tail has a long lower wick and a small body near the upper end. The source prefers approximately 80–85% of the entire range to be tail and says the body’s color matters less than the tail’s shape. Its confirmation entry is above the completed tail bar’s high, with protection below the event.

For a candle with high $50.10, low $49.60, open $50.00 and close $50.05, range is $0.50 and lower wick is $0.40. The lower-wick fraction is 80%. A $0.25 lower wick on the same range is only 50% and fails our 80% baseline regardless of how attractive the subsequent rally looks.

Color Changes and 180s

A bullish 180 in the source consists of a sizable red candle followed by a stronger green candle that exceeds the red candle’s high; a bearish version reverses direction. The broader color-change category also includes body-based reversals. The presenter later allows ignoring wicks for some non-tail entries.

Those variants should not be mixed casually in a test. Changing color, exceeding a body edge and breaking the full preceding high are different triggers. Specify which one applies before viewing the outcome.

Anticipation Versus Confirmation

The presenter describes entering a forming two-minute elephant bar late in its development, approximately 75–90 seconds into the bar, or waiting for completion and a subsequent break above its high. He favors the earlier price in some cases because it may reduce the distance to the protective level.

Earlier entry can produce a smaller planned stop distance while increasing uncertainty about the candle’s final form. A candle can still change substantially in its final seconds. “Less distance to the stop” is not synonymous with lower overall risk.

Testing an intrabar decision requires information about the partial candle at that time. Turning on realtime tick calculations does not recreate missing historical ticks. Our baseline chooses the completed tail-bar confirmation route instead of using final OHLC to manufacture an earlier signal.

Know the State

Closeness and flatness need numeric definitions for research. Screen angles such as 45 degrees change with zoom and axis scaling, so measure price changes over a fixed number of bars instead. Divide distances by ATR or use a predefined percentage for comparison across price levels.

Two flat averages do not establish zero volatility. An average can remain flat while individual bars swing, and low volatility can persist. Likewise, wide separation does not give a deadline for reversion. A countertrend event can fail while an extended trend continues.

Know the Position

The source calls the area between the averages a trap zone and ranks positions beyond it. Immediately above the state favors bullish events; immediately below favors bearish events. A moderately extended position may support either direction depending on the event, while a far extension can shift attention toward an opposing trade.

These are qualitative categories. Our baseline defines only a near-above position: the tail closes above the upper average but no more than one prior ATR above it, and its low remains near the state. It does not classify every position-two or position-three example.

Use the state measured before the signal candle. Otherwise a large event can move the 20 average and alter the context being used to justify itself. The 20 and 200 are related transformations of the same prices, not independent evidence streams.

The Stop Is a Price Boundary, Not a Guaranteed One-Bar Loss

The source repeatedly recommends protection under a bullish event or above a bearish event. Later, he says an event very close to the state may need protection beyond the state rather than a stop inside the preferred location. These are distinct stop choices with different position-size implications.

For our long baseline, the stop is one tick below the lower of the tail’s low and the lower state boundary. This explicit rule protects both references. It is our implementation of the source’s two ideas, not a uniquely specified formula from the transcript.

If a buy trigger is $50.11 and the frozen stop is $49.39, planned price distance is $0.72. A gap fill at $50.50 increases that distance to $1.11. With 100 shares, planned price loss increases from $72 to $111 before costs. A gap through the stop can increase realized loss further.

“One bar” describes a structural distance in the lesson. It does not guarantee a monetary loss, an exit within one bar of time or a fill at the protective level. A large event bar can be expensive even when the stop is only one event away.

Follow-Through Is Not Trade Profitability

The source asserts 70%, 80% and 90% follow-through or success rates for event/context combinations, and reports profitable automated systems with a lower success rate. The transcript supplies neither complete datasets nor consistent definitions, horizons and cost assumptions that verify those numbers.

A small move beyond an entry can count as follow-through while the completed trade later loses. Profitability depends on actual entry and exit rules, loss size, costs and outcome distributions. For illustration, a 70% win rate with average wins of +0.2R and average losses of −1R has gross expectancy 0.70 × 0.2R − 0.30 × 1R = −0.16R. High win rate alone is insufficient.

Universality across all markets is also a claim requiring evidence. A stock candle does not fully describe an option’s changing price, spread, volatility or assignment exposure. Treat instruments and timeframes as separate research settings.

An Executable Baseline: Tail Confirmation Above a Narrow State

//@version=6
strategy("Tail breakout above flat narrow state: research baseline", overlay=true,
     initial_capital=100000, default_qty_type=strategy.percent_of_equity,
     default_qty_value=10, pyramiding=0, margin_long=100,
     commission_type=strategy.commission.percent, commission_value=0.05,
     slippage=2, process_orders_on_close=false,
     calc_on_every_tick=false, calc_on_order_fills=false)
// Standard two-minute US equity candles, regular-hours chart data only.
if not (timeframe.isminutes and timeframe.multiplier == 2)
    runtime.error("Use a standard two-minute chart with regular-hours data.")
int slopeBars = input.int(5, "Flatness lookback", minval=1, maxval=20)
float flat20 = input.float(0.20, "Maximum 20 SMA change / ATR", minval=0.0, step=0.05)
float flat200 = input.float(0.10, "Maximum 200 SMA change / ATR", minval=0.0, step=0.05)
float separationATR = input.float(0.50, "Maximum MA separation / ATR", minval=0.0, step=0.1)
float maxPositionATR = input.float(1.0, "Maximum close above state / ATR", minval=0.1, step=0.1)
float tailFraction = input.float(0.80, "Minimum lower-wick fraction", minval=0.5, maxval=0.95, step=0.05)
bool requireFlat = input.bool(true, "Require both moving averages flat")
int holdBars = input.int(8, "Experimental holding bars after fill", minval=1, maxval=12)

float ma20 = ta.sma(close, 20)
float ma200 = ta.sma(close, 200)
float atr = ta.atr(14)
// State is measured before the event, using preceding completed data.
float stateUpper = math.max(ma20[1], ma200[1])
float stateLower = math.min(ma20[1], ma200[1])
float referenceATR = atr[1]
float slope20 = math.abs(ma20[1] - ma20[slopeBars + 1])
float slope200 = math.abs(ma200[1] - ma200[slopeBars + 1])
bool ready = not na(slope200) and referenceATR > 0
bool narrow = ready and stateUpper - stateLower <= separationATR * referenceATR
bool flat = slope20 <= flat20 * referenceATR and slope200 <= flat200 * referenceATR
float barRange = high - low
float lowerWick = math.min(open, close) - low
bool pureTail = barRange > 0 and lowerWick / barRange >= tailFraction
bool nearAbove = close > stateUpper and close <= stateUpper + maxPositionATR * referenceATR
// Prevent a tail far above the state from qualifying merely on its close.
bool touchesArea = low <= stateUpper + 0.25 * referenceATR and low >= stateLower - 0.25 * referenceATR
bool entrySession = not na(time(timeframe.period, "0930-1500:23456", "America/New_York"))
var bool pending = false
var int setupBar = na
var float fixedStop = na

if barstate.isconfirmed
    // Snapshot prevents replacing a setup on its expiry/fill bar.
    bool handledPending = pending
    if pending and bar_index > setupBar
        strategy.cancel("Long")
        if strategy.position_size == 0
            strategy.cancel("Protection")
        pending := false
    if strategy.position_size > 0
        int entryBar = strategy.opentrades.entry_bar_index(0)
        if bar_index - entryBar + 1 >= holdBars
            strategy.close("Long", comment="Experimental timed exit", immediately=true)
    if not handledPending and not pending and strategy.position_size == 0 and entrySession and narrow and (not requireFlat or flat) and pureTail and nearAbove and touchesArea
        // Confirmation order is valid on the next available chart bar only.
        float trigger = high + syminfo.mintick
        // Protect the event AND the lower state boundary when it lies below it.
        fixedStop := math.floor((math.min(low, stateLower) - syminfo.mintick) / syminfo.mintick) * syminfo.mintick
        strategy.entry("Long", strategy.long, stop=trigger)
        strategy.exit("Protection", from_entry="Long", stop=fixedStop)
        setupBar := bar_index
        pending := true
// There is no profit target, trailing stop or addition. Stop remains frozen.
plot(ma20, "20 SMA", color=color.blue)
plot(ma200, "200 SMA", color=color.orange)
plot(pending or strategy.position_size > 0 ? fixedStop : na, "Frozen protective stop", color=color.red, style=plot.style_linebr)

Use standard two-minute US equity candles with regular-hours chart data. This baseline isolates one event and one position; it deliberately excludes elephant bars, 180s, wide-state reversals and additions.

All numeric state thresholds are research choices. The baseline does not separately require a quiet ten-bar price range, so its “flat narrow” classification approximates the moving-average relationship rather than the full discretionary compression description. The eight-bar exit is a test choice inspired by the source’s duration discussion, not his complete profit-taking rule.

What This Code Actually Tests

It tests whether a confirmed bottoming-tail breakout near a quantitatively defined narrow state produces useful results under fixed protection and a timed exit. Disabling flatness compares narrow setups with and without that additional restriction.

The entry and protective stop are submitted together after the event closes. If the next available bar does not fill the entry, it is canceled at that bar’s close. Later setups do not replace it during that waiting bar. If it fills, the protective level stays fixed; subsequent events cannot overwrite it.

The holding count begins with the actual fill bar, which counts as bar one. The strategy closes on the eighth holding bar’s closing tick by default unless the protective stop exits first. There is no target or claimed fixed R multiple. Ten-percent equity sizing controls exposure, not a fixed risk budget.

What This Code Does Not Test

It does not recreate a partial elephant candle at 75 seconds, prove institutional participation, estimate the source’s 70–90% claims or test every state and position. It has no adds, trailing stop, profit-taking tiers, independent volume filter or short trades.

The clock admits setup submissions between 09:30 and 15:00 New York time; it is not an exchange-calendar model. Extended-hours bars would change the averages. Early-close sessions, missing bars and insufficient warm-up history need review. Holding bars can span a session boundary if data are missing or the market closes early; there is no automatic session-end liquidation.

Execution and Validation

Stop-entry fills can occur beyond the trigger after a gap. The frozen absolute stop retains its original level, so actual fill-to-stop risk changes; the script does not assume the trigger was the fill. Submitted protection can operate on the fill bar. If entry and protection are both touched, historical path assumptions can affect the result.

The timed exit’s immediate closing-tick fill is a simulation shortcut and not a guarantee that a trader can transact at that exact close. Defaults are 0.05% commission per fill and two ticks of slippage. Costs, quantity rounding and product specifications must be adjusted to the intended market.

See TradingView’s strategy documentation (paid link) for order emulation and repainting documentation (paid link) for differences between developing and completed bars. The baseline uses no higher-timeframe request and makes setup decisions on completed bars. Intrabar fill ordering still requires scrutiny; Bar Magnifier can help where suitable data exist.

Validation status: The script has been statically reviewed and its full code interface checked. It has not been compiled or backtested in TradingView. No returns, win rates or measured holding-time averages are claimed.

Backtest Lab

1. Does flatness help? Compare the flatness requirement enabled versus disabled with identical event, position and exit rules. Report trade count, expectancy after costs, exposure and drawdown.
2. Tail purity: Compare minimum wick fractions of 0.75, 0.80 and 0.85. Preserve a later untouched evaluation period rather than selecting the best past threshold.
3. Exit dependence: Compare holding limits of four, eight and twelve bars. Record actual exit causes and fill-to-stop distances. This investigates exit sensitivity, not the asserted universal eight-bar average.
4. Execution stress: Increase commission to 0.10% and slippage to five ticks. Inspect gaps, entry-and-stop bars, expired orders and session-boundary exposure across several liquid stocks.

A separate anticipation study would require lower-timeframe or timestamped information sufficient to reconstruct the candle at the decision instant. Final two-minute OHLC cannot supply that history. Adding tick calculations to a script alone does not solve the missing-data problem.

Check Your Understanding

1. Does a final elephant candle prove an entry was valid 90 seconds earlier? No. Its partial state may have differed.
2. Are flat averages evidence of zero volatility? No. Individual price movements can remain substantial.
3. Does a 90% follow-through claim establish net profitability? No. Definitions, exits, costs and loss size are required.
4. Does an event stop guarantee one-bar monetary risk? No. Bar size, actual fill, quantity and gaps determine monetary outcomes.
5. Why are event, state and position separate? They distinguish the trigger from the context and the degree of extension.

Practice Exercise

Label 25 consecutive tail candidates without revealing later bars. Record wick fraction, pre-event average separation, pre-event slopes, position distance, trigger and protection. Keep rejected examples and valid setups that never fill.

For every fill, calculate actual entry minus the frozen stop and note the number of bars held. A system with attractive examples can still fail if the typical event is too expensive, confirmation gaps worsen fills or timed exits give back favorable movement.

Final Takeaway

The source offers a useful entry vocabulary: event, state and position. Make that vocabulary measurable before relying on it. Separate anticipated bars from confirmed events, structural stops from guaranteed loss limits, and follow-through claims from completed-trade returns. The baseline is a starting experiment, not proof of the advertised odds.

Risk Disclosure

This guide is educational and is not personalized financial advice. Trading can produce substantial losses, and leverage may amplify them. Stops do not guarantee execution prices or planned loss limits. Hypothetical examples, reported results and historical simulations do not guarantee future performance; liquidity, costs and market conditions can alter outcomes.

CandleChasers is an independent educational publisher and is not affiliated with Oliver Velez Trading. The calculations, experimental definitions and research exercises here are CandleChasers’ independent analysis of the supplied material.