div_bull(pS, iS, cp_length_after, cp_length_before, pivot_length, lookback, no_broken, pW, iW, hidW, regW) Test for bullish divergence Parameters: pS: Price series (float) iS: Indicator series (float) cp_length_after: Bars after current (divergent) pivot low to be considered a valid pivot (optional int) cp_length_before: Bars before current (divergent) pivot low to be considered a valid pivot (optional int) pivot_length: Bars before and after prior pivot low to be considered valid pivot (optional int) lookback: Bars back to search for prior pivot low (optional int) no_broken: Flag to only consider divergence valid if the pivot-to-pivot trendline is unbroken (optional bool) pW: Weight of change in price, used in degree of divergence calculation (optional float) iW: Weight of change in indicator, used in degree of divergence calculation (optional float) hidW: Weight of hidden divergence, used in degree of divergence calculation (optional float) regW: Weight of regular divergence, used in degree of divergence calculation (optional float) Returns: [flag,degree,type,lx1,ly1,lx2,ly2] flag = true if divergence exists (bool) degree = degree (strength) of divergence (float) type = 1 = regular, 2 = hidden (int) lx1 = x coordinate 1 (int) ly1 = y coordinate 1 (float) lx2 = x coordinate 2 (int) ly2 = y coordinate 2 (float)
div_bear(pS, iS, cp_length_after, cp_length_before, pivot_length, lookback, no_broken, pW, iW, hidW, regW) Test for bearish divergence Parameters: pS: Price series (float) iS: Indicator series (float) cp_length_after: Bars after current (divergent) pivot high to be considered a valid pivot (optional int) cp_length_before: Bars before current (divergent) pivot highto be considered a valid pivot (optional int) pivot_length: Bars before and after prior pivot high to be considered valid pivot (optional int) lookback: Bars back to search for prior pivot high (optional int) no_broken: Flag to only consider divergence valid if the pivot-to-pivot trendline is unbroken (optional bool) pW: Weight of change in price, used in degree of divergence calculation (optional float) iW: Weight of change in indicator, used in degree of divergence calculation (optional float) hidW: Weight of hidden divergence, used in degree of divergence calculation (optional float) regW: Weight of regular divergence, used in degree of divergence calculation (optional float) Returns: [flag,degree,type,lx1,ly1,lx2,ly2] flag = true if divergence exists (bool) degree = degree (strength) of divergence (float) type = 1 = regular, 2 = hidden (int) lx1 = x coordinate 1 (int) ly1 = y coordinate 1 (float) lx2 = x coordinate 2 (int) ly2 = y coordinate 2 (float)
Versionsinformation
⋅
v2 Return absolute value of degree to handle series with negative values.
Versionsinformation
⋅
v3 - Added support for lookback # of pivots for div_bull() and div_bear() - Fixed divergence degree calculation when indicator series has negative values - Changed default pivot lengths for div_bull() and div_bear()
Updated: div_bull(pS, iS, cp_length_after, cp_length_before, pivot_length, lookback, lookback_pivs, no_broken, pW, iW, hidW, regW) Test for bullish divergence Parameters: pS: Price series (float) iS: Indicator series (float) cp_length_after: Bars after current (divergent) pivot low to be considered a valid pivot (optional int) cp_length_before: Bars before current (divergent) pivot low to be considered a valid pivot (optional int) pivot_length: Bars before and after prior pivot low to be considered valid pivot (optional int) lookback: Bars back to search for prior pivot low (optional int) lookback_pivs: Pivots back to search for prior pivot low (optional int) no_broken: Flag to only consider divergence valid if the pivot-to-pivot trendline is unbroken (optional bool) pW: Weight of change in price, used in degree of divergence calculation (optional float) iW: Weight of change in indicator, used in degree of divergence calculation (optional float) hidW: Weight of hidden divergence, used in degree of divergence calculation (optional float) regW: Weight of regular divergence, used in degree of divergence calculation (optional float) Returns: [flag,degree,type,lx1,ly1,lx2,ly2] flag = true if divergence exists (bool) degree = degree (strength) of divergence (float) type = 1 = regular, 2 = hidden (int) lx1 = x coordinate 1 (int) ly1 = y coordinate 1 (float) lx2 = x coordinate 2 (int) ly2 = y coordinate 2 (float)
div_bear(pS, iS, cp_length_after, cp_length_before, pivot_length, lookback, lookback_pivs, no_broken, pW, iW, hidW, regW) Test for bearish divergence Parameters: pS: Price series (float) iS: Indicator series (float) cp_length_after: Bars after current (divergent) pivot high to be considered a valid pivot (optional int) cp_length_before: Bars before current (divergent) pivot highto be considered a valid pivot (optional int) pivot_length: Bars before and after prior pivot high to be considered valid pivot (optional int) lookback: Bars back to search for prior pivot high (optional int) lookback_pivs: Pivots back to search for prior pivot high (optional int) no_broken: Flag to only consider divergence valid if the pivot-to-pivot trendline is unbroken (optional bool) pW: Weight of change in price, used in degree of divergence calculation (optional float) iW: Weight of change in indicator, used in degree of divergence calculation (optional float) hidW: Weight of hidden divergence, used in degree of divergence calculation (optional float) regW: Weight of regular divergence, used in degree of divergence calculation (optional float) Returns: [flag,degree,type,lx1,ly1,lx2,ly2] flag = true if divergence exists (bool) degree = degree (strength) of divergence (float) type = 1 = regular, 2 = hidden (int) lx1 = x coordinate 1 (int) ly1 = y coordinate 1 (float) lx2 = x coordinate 2 (int) ly2 = y coordinate 2 (float)
Versionsinformation
⋅
v4 Changed divergence strength/degree calculation to use the angle of the divergence line divided by 90 degrees (as opposed to the % change in the indicator value). This will keep the unweighted strength value between 0 and 1 for the indicator portion of the calculation, which is desirable because different indicator series can have wildly different ranges.
Versionsinformation
⋅
v5 Added chk_prev_bar flag as an argument to the divergence checks, which will check for divergence using the indicator value from the prior bar. This is necessary in some cases where the indicator is using a different price source than the one passed into the divergence function.
Updated: div_bull(pS, iS, chk_prev_bar, cp_length_after, cp_length_before, pivot_length, lookback, lookback_pivs, no_broken, pW, iW, hidW, regW) Test for bullish divergence Parameters: pS: Price series (float) iS: Indicator series (float) chk_prev_bar: Flag to check for divergence using the indicator value from the prior bar. This is necessary in some cases where the indicator is using a different price source than the one passed in pS. cp_length_after: Bars after current (divergent) pivot low to be considered a valid pivot (optional int) cp_length_before: Bars before current (divergent) pivot low to be considered a valid pivot (optional int) pivot_length: Bars before and after prior pivot low to be considered valid pivot (optional int) lookback: Bars back to search for prior pivot low (optional int) lookback_pivs: Pivots back to search for prior pivot low (optional int) no_broken: Flag to only consider divergence valid if the pivot-to-pivot trendline is unbroken (optional bool) pW: Weight of change in price, used in degree of divergence calculation (optional float) iW: Weight of change in indicator, used in degree of divergence calculation (optional float) hidW: Weight of hidden divergence, used in degree of divergence calculation (optional float) regW: Weight of regular divergence, used in degree of divergence calculation (optional float) Returns: [flag,degree,type,lx1,ly1,lx2,ly2] flag = true if divergence exists (bool) degree = degree (strength) of divergence (float) type = 1 = regular, 2 = hidden (int) lx1 = x coordinate 1 (int) ly1 = y coordinate 1 (float) lx2 = x coordinate 2 (int) ly2 = y coordinate 2 (float)
div_bear(pS, iS, chk_prev_bar, cp_length_after, cp_length_before, pivot_length, lookback, lookback_pivs, no_broken, pW, iW, hidW, regW) Test for bearish divergence Parameters: pS: Price series (float) iS: Indicator series (float) chk_prev_bar: Flag to check for divergence using the indicator value from the prior bar. This is necessary in some cases where the indicator is using a different price source than the one passed in pS. cp_length_after: Bars after current (divergent) pivot high to be considered a valid pivot (optional int) cp_length_before: Bars before current (divergent) pivot highto be considered a valid pivot (optional int) pivot_length: Bars before and after prior pivot high to be considered valid pivot (optional int) lookback: Bars back to search for prior pivot high (optional int) lookback_pivs: Pivots back to search for prior pivot high (optional int) no_broken: Flag to only consider divergence valid if the pivot-to-pivot trendline is unbroken (optional bool) pW: Weight of change in price, used in degree of divergence calculation (optional float) iW: Weight of change in indicator, used in degree of divergence calculation (optional float) hidW: Weight of hidden divergence, used in degree of divergence calculation (optional float) regW: Weight of regular divergence, used in degree of divergence calculation (optional float) Returns: [flag,degree,type,lx1,ly1,lx2,ly2] flag = true if divergence exists (bool) degree = degree (strength) of divergence (float) type = 1 = regular, 2 = hidden (int) lx1 = x coordinate 1 (int) ly1 = y coordinate 1 (float) lx2 = x coordinate 2 (int) ly2 = y coordinate 2 (float)
Versionsinformation
⋅
v6 When calculating the divergence strength on an indicator, Scale the X coordinates to a range similar to the Y scale (X and Y scales should be on the same order of magnitude for the angle value to be meaningful. e.g. if ΔY is 10,000,000 and ΔX is 12, the angle will always be very nearly 90 degrees, which makes for a meaningless value)
Versionsinformation
⋅
v7 Adjusted multiplier in iDelta() to use the average Y value rather than the difference of the two Y values.
Versionsinformation
⋅
v8 Now calculating divergence strength in terms of the standard deviation. = Δ price/StDev + Δ indicator/stDev (multiplied by specified weights)
Versionsinformation
⋅
v9
Versionsinformation
⋅
v10 Changes to check more thoroughly for pivots on indicator series.
Updated: div_bull(pS, iS, cp_length_after, cp_length_before, pivot_length, lookback, lookback_pivs, no_broken, pW, iW, hidW, regW) Test for bullish divergence Parameters: pS: Price series (float) iS: Indicator series (float) cp_length_after: Bars after current (divergent) pivot low to be considered a valid pivot (optional int) cp_length_before: Bars before current (divergent) pivot low to be considered a valid pivot (optional int) pivot_length: Bars before and after prior pivot low to be considered valid pivot (optional int) lookback: Bars back to search for prior pivot low (optional int) lookback_pivs: Pivots back to search for prior pivot low (optional int) no_broken: Flag to only consider divergence valid if the pivot-to-pivot trendline is unbroken (optional bool) pW: Weight of change in price, used in degree of divergence calculation (optional float) iW: Weight of change in indicator, used in degree of divergence calculation (optional float) hidW: Weight of hidden divergence, used in degree of divergence calculation (optional float) regW: Weight of regular divergence, used in degree of divergence calculation (optional float) Returns: [flag,degree,type,lx1,ly1,lx2,ly2] flag = true if divergence exists (bool) degree = degree (strength) of divergence (float) type = 1 = regular, 2 = hidden (int) lx1 = x coordinate 1 (int) ly1 = y coordinate 1 (float) lx2 = x coordinate 2 (int) ly2 = y coordinate 2 (float)
div_bear(pS, iS, cp_length_after, cp_length_before, pivot_length, lookback, lookback_pivs, no_broken, pW, iW, hidW, regW) Test for bearish divergence Parameters: pS: Price series (float) iS: Indicator series (float) cp_length_after: Bars after current (divergent) pivot high to be considered a valid pivot (optional int) cp_length_before: Bars before current (divergent) pivot highto be considered a valid pivot (optional int) pivot_length: Bars before and after prior pivot high to be considered valid pivot (optional int) lookback: Bars back to search for prior pivot high (optional int) lookback_pivs: Pivots back to search for prior pivot high (optional int) no_broken: Flag to only consider divergence valid if the pivot-to-pivot trendline is unbroken (optional bool) pW: Weight of change in price, used in degree of divergence calculation (optional float) iW: Weight of change in indicator, used in degree of divergence calculation (optional float) hidW: Weight of hidden divergence, used in degree of divergence calculation (optional float) regW: Weight of regular divergence, used in degree of divergence calculation (optional float) Returns: [flag,degree,type,lx1,ly1,lx2,ly2] flag = true if divergence exists (bool) degree = degree (strength) of divergence (float) type = 1 = regular, 2 = hidden (int) lx1 = x coordinate 1 (int) ly1 = y coordinate 1 (float) lx2 = x coordinate 2 (int) ly2 = y coordinate 2 (float)
Versionsinformation
⋅
v11 div_bull() and div_bear() - adjusted indicator pivot checking to be more inclusive.
Versionsinformation
⋅
v12 Added function for validating harmonic XABCD patterns: harmonic_validate_xabcd()
Added: harmonic_validate_xabcd(xX, xY, aX, aY, bX, bY, cX, cY, dX, dY, pErr, pAsym, pivot_length, gart, bat, bfly, crab) Validate harmonic XABCD pattern Parameters: xX: X coordinate of point X (int) xY: Y coordinate of point X (float) aX: X coordinate of point A (int) aY: Y coordinate of point A (float) bX: X coordinate of point B (int) bY: Y coordinate of point B (float) cX: X coordinate of point C (int) cY: Y coordinate of point C (float) dX: X coordinate of point D (int) dY: Y coordinate of point D (float) pErr: Acceptable percent error of leg ratios (does not apply to legs defined within a range) (float) pAsym: Acceptable percent asymmetry of leg ΔX (each leg tested against average ΔX of prior legs) (float) pivot_length: Bars before and after prior pivot high to be considered valid pivot (optional int) gart: Flag to validate Gartley pattern (bool) bat: Flag to validate Bat pattern (bool) bfly: Flag to validate Butterfly pattern (bool) crab: Flag to validate Crab pattern (bool) Returns: [flag,t1,t2,t3,t4] flag = true if valid harmonic t1 = true if valid gartley t2 = true if valid bat t3 = true if valid butterfly t4 = true if valid crab
Versionsinformation
⋅
v13
Added pattern_xabcd() - determine if an XABCD pattern has just completed.
Added: pattern_xabcd(x_is_low, pivot_length, source) Determine if an XABCD pattern has just completed (i.e. point D is on the previous bar) Parameters: x_is_low: Flag to determine if point X is a low pivot, i.e. bullish pattern (bool, dft = true) pivot_length: Number of bars before and after a valid pivot (int, dft = 5) source: Source series (float, dft = na, will use high and low series) Returns: [flag,xx,xy,ax,ay,bx,by,cx,cy,dx,dy] flag = true if valid XABCD pattern completed on previous bar xx = X coordinate of point X (int) xy = Y coordinate of point X (float) ax = X coordinate of point A (int) ay = Y coordinate of point A (float) bx = X coordinate of point B (int) by = Y coordinate of point B (float) cx = X coordinate of point C (int) cy = Y coordinate of point C (float) dx = X coordinate of point D (int) dy = Y coordinate of point D (float)
Versionsinformation
⋅
v14
Added: pat_xabcd(x_is_low, pivot_length, source) Determine if an XABCD pattern has just completed (i.e. point D is on the previous bar) Parameters: x_is_low: Flag to determine if point X is a low pivot, i.e. bullish pattern (bool, dft = true) pivot_length: Number of bars before and after a valid pivot (int, dft = 5) source: Source series (float, dft = na, will use high and low series) Returns: [flag,xx,xy,ax,ay,bx,by,cx,cy,dx,dy] flag = true if valid XABCD pattern completed on previous bar xx = X coordinate of point X (int) xy = Y coordinate of point X (float) ax = X coordinate of point A (int) ay = Y coordinate of point A (float) bx = X coordinate of point B (int) by = Y coordinate of point B (float) cx = X coordinate of point C (int) cy = Y coordinate of point C (float) dx = X coordinate of point D (int) dy = Y coordinate of point D (float)
Removed: pattern_xabcd(x_is_low, pivot_length, source) Determine if an XABCD pattern has just completed (i.e. point D is on the previous bar)
Versionsinformation
⋅
v15 Fixed point D validation for harmonic XABCD pattern
Versionsinformation
⋅
v16 Fixed harmonic XABCD ΔX symmetry validation
Versionsinformation
⋅
v17
Added: harmonic_xabcd_validate(xX, xY, aX, aY, bX, bY, cX, cY, dX, dY, pErr, pAsym, pivot_length, gart, bat, bfly, crab) Validate harmonic XABCD pattern Parameters: xX: X coordinate of point X (int) xY: Y coordinate of point X (float) aX: X coordinate of point A (int) aY: Y coordinate of point A (float) bX: X coordinate of point B (int) bY: Y coordinate of point B (float) cX: X coordinate of point C (int) cY: Y coordinate of point C (float) dX: X coordinate of point D (int) dY: Y coordinate of point D (float) pErr: Acceptable percent error of leg ratios (does not apply to legs defined within a range) (float) pAsym: Acceptable percent asymmetry of leg ΔX (each leg tested against average ΔX of prior legs) (float) pivot_length: Bars before and after prior pivot high to be considered valid pivot (optional int) gart: Flag to validate Gartley pattern (bool) bat: Flag to validate Bat pattern (bool) bfly: Flag to validate Butterfly pattern (bool) crab: Flag to validate Crab pattern (bool) Returns: [flag,t1,t2,t3,t4] flag = true if valid harmonic t1 = true if valid gartley t2 = true if valid bat t3 = true if valid butterfly t4 = true if valid crab
harmonic_xabcd_prz(type, xY, aY, bY, cY) Get the potential reversal zone (PRZ) levels of a harmonic XABCD pattern Parameters: type: Harmonic pattern type (int - 1 = Gartley, 2 = Bat, 3 = Butterfly, 4 = Crab) xY: Y coordinate of point X (float) aY: Y coordinate of point A (float) bY: Y coordinate of point B (float) cY: Y coordinate of point C (float) Returns: [bc_u, bc_l, xa_u, xa_l] bc_u = nearest BC retracement/extension level (nearest to point C) bc_l = farthest BC retracement/extension level (nearest to point C) xa_u = nearest XA retracement/extension level (or the only XA level, if applicable) xa_l = farthest XA retracement/extension level (or na if not applicable)
Added: harmonic_xabcd_validateIncomplete(xX, xY, aX, aY, bX, bY, cX, cY, pErr, pAsym, pivot_length, gart, bat, bfly, crab) Validate the first 3 legs of a harmonic XABCD pattern Parameters: xX: X coordinate of point X (int) xY: Y coordinate of point X (float) aX: X coordinate of point A (int) aY: Y coordinate of point A (float) bX: X coordinate of point B (int) bY: Y coordinate of point B (float) cX: X coordinate of point C (int) cY: Y coordinate of point C (float) pErr: Acceptable percent error of leg ratios (does not apply to legs defined within a range) (float) pAsym: Acceptable percent asymmetry of leg ΔX (each leg tested against average ΔX of prior legs) (float) pivot_length: Bars before and after prior pivot high to be considered valid pivot (optional int) gart: Flag to validate Gartley pattern (bool) bat: Flag to validate Bat pattern (bool) bfly: Flag to validate Butterfly pattern (bool) crab: Flag to validate Crab pattern (bool) Returns: [flag,t1,t2,t3,t4] flag = true if valid harmonic t1 = true if valid gartley t2 = true if valid bat t3 = true if valid butterfly t4 = true if valid crab
Versionsinformation
⋅
v20 Fixed bug with pat_axbcd() algorithm, checking wrong pivot looking for point C in bearish patterns.
Versionsinformation
⋅
v21 Fixed bug in pat_xabcd() where intermediate low wasn't being detected between points C and D
Versionsinformation
⋅
v22
Added: harmonic_xabcd_przClosest(l1, l2, l3, l4) Get the confluent PRZ levels (i.e. the two closest PRZ levels) Order of arguments does not matter Parameters: l1: level 1 (float) l2: level 2 (float) l3: level 3 (float) l4: level 4 (optional, float) Returns: [lL,lH] lL = lower confluent PRZ level lH = higher confluent PRZ level
Versionsinformation
⋅
v23
Added: success(entry, stop, t1, t2) Determine if trade is successful Parameters: entry: Entry bar index (int) stop: Stop level (float) t1: Target 1 level (float) t2: Target 2 level (float) Returns: [t1Hit,t2Hit,t1x,t1y,t2x,t2y]
Versionsinformation
⋅
v24
Added: harmonic_xabcd_rAndE(type, l, l1, l2) Get the ratio of two pattern legs, and the percent error from the theoretical harmonic ratio Order of arguments does not matter Parameters: type: Harmonic pattern type (int - 1 = Gartley, 2 = Bat, 3 = Butterfly, 4 = Crab) l: Leg ID ("xab", "abc", "bcd", or "xad") (string) l1: Line 1 height (float) l2: Line 2 height (float) Returns: [r,e]
Versionsinformation
⋅
v25
Added: harmonic_xabcd_przRange() Get upper and lower PRZ levels
harmonic_xabcd_eD() Measure closeness of D to either of the two closest PRZ levels, relative to height of the XA leg
przScore() Measure the closeness of the two closest PRZ levels, relative to the height of the XA leg
harmonic_xabcd_eAvg() Get the avg retracement ratio % error
harmonic_xabcd_targets() Get target level
pat_xabcd_asym() Get the avg asymmetry %
Versionsinformation
⋅
v26
Added: harmonic_xabcd_przScore() Measure the closeness of the two closest PRZ levels, relative to the height of the XA leg
Removed: przScore() Measure the closeness of the two closest PRZ levels, relative to the height of the XA leg
Versionsinformation
⋅
v27
Updated: pat_xabcd(x_is_low, pivot_length, source, conf_length) Determine if an XABCD pattern has just completed (i.e. point D is on the previous bar) Parameters: x_is_low: Flag to determine if point X is a low pivot, i.e. bullish pattern (bool, dft = true) pivot_length: Number of bars before and after a valid pivot (int, dft = 5) source: Source series (float, dft = na, will use high and low series) conf_length: Number of trailing bars after pivot point D to confirm a valid pattern (int, dft = 1) Returns: [flag,xx,xy,ax,ay,bx,by,cx,cy,dx,dy] flag = true if valid XABCD pattern completed on previous bar xx = X coordinate of point X (int) xy = Y coordinate of point X (float) ax = X coordinate of point A (int) ay = Y coordinate of point A (float) bx = X coordinate of point B (int) by = Y coordinate of point B (float) cx = X coordinate of point C (int) cy = Y coordinate of point C (float) dx = X coordinate of point D (int) dy = Y coordinate of point D (float)
Versionsinformation
⋅
v28
Versionsinformation
⋅
v29
Versionsinformation
⋅
v30
Added: harmonic_xabcd_score() Get score values for a pattern
Versionsinformation
⋅
v31 Added Shark and Cypher XABCD patterns
Updated: harmonic_xabcd_validate(xX, xY, aX, aY, bX, bY, cX, cY, dX, dY, pErr, pAsym, pivot_length, gart, bat, bfly, crab, shark, cyph) Validate harmonic XABCD pattern Parameters: xX: X coordinate of point X (int) xY: Y coordinate of point X (float) aX: X coordinate of point A (int) aY: Y coordinate of point A (float) bX: X coordinate of point B (int) bY: Y coordinate of point B (float) cX: X coordinate of point C (int) cY: Y coordinate of point C (float) dX: X coordinate of point D (int) dY: Y coordinate of point D (float) pErr: Acceptable percent error of leg ratios (does not apply to legs defined within a range) (float) pAsym: Acceptable percent asymmetry of leg ΔX (each leg tested against average ΔX of prior legs) (float) pivot_length: Bars before and after prior pivot high to be considered valid pivot (optional int) gart: Flag to validate Gartley pattern (bool) bat: Flag to validate Bat pattern (bool) bfly: Flag to validate Butterfly pattern (bool) crab: Flag to validate Crab pattern (bool) shark: Flag to validate Shark pattern (bool) cyph: Flag to validate Cypher pattern (bool) Returns: [flag,t1,t2,t3,t4,t5,t6] flag = true if valid harmonic t1 = true if valid gartley t2 = true if valid bat t3 = true if valid butterfly t4 = true if valid crab t5 = true if valid shark t6 = true if valid cypher
harmonic_xabcd_validateIncomplete(xX, xY, aX, aY, bX, bY, cX, cY, pErr, pAsym, pivot_length, gart, bat, bfly, crab, shark, cyph) Validate the first 3 legs of a harmonic XABCD pattern Parameters: xX: X coordinate of point X (int) xY: Y coordinate of point X (float) aX: X coordinate of point A (int) aY: Y coordinate of point A (float) bX: X coordinate of point B (int) bY: Y coordinate of point B (float) cX: X coordinate of point C (int) cY: Y coordinate of point C (float) pErr: Acceptable percent error of leg ratios (does not apply to legs defined within a range) (float) pAsym: Acceptable percent asymmetry of leg ΔX (each leg tested against average ΔX of prior legs) (float) pivot_length: Bars before and after prior pivot high to be considered valid pivot (optional int) gart: Flag to validate Gartley pattern (bool) bat: Flag to validate Bat pattern (bool) bfly: Flag to validate Butterfly pattern (bool) crab: Flag to validate Crab pattern (bool) shark: Flag to validate Shark pattern (bool) cyph: Flag to validate Cypher pattern (bool) Returns: [flag,t1,t2,t3,t4] flag = true if valid harmonic t1 = true if valid gartley t2 = true if valid bat t3 = true if valid butterfly t4 = true if valid crab t5 = true if valid shark t6 = true if valid cypher
Versionsinformation
⋅
v32
Updated: harmonic_xabcd_przScore() Measure the closeness of the two closest PRZ levels, relative to the height of the XA leg
Versionsinformation
⋅
v33
Added: harmonic_xabcd_fibDispTxt() Get fib ratio display text
Versionsinformation
⋅
v34
Added: harmonic_xabcd_symbol() Get pattern symbol
Versionsinformation
⋅
v35
Added: harmonic_xabcd_scoreTot() Get total weighted score value for a pattern
Versionsinformation
⋅
v36
Removed: harmonic_xabcd_scoreTot() Get total weighted score value for a pattern
harmonic_xabcd_fibDispTxt() Get fib ratio display text
harmonic_xabcd_symbol() Get pattern symbol
Versionsinformation
⋅
v37
Added: harmonic_xabcd_scoreTot() Get pattern's total weighted score
Versionsinformation
⋅
v38
Added: harmonic_xabcd_fibDispTxt() Get fib ratio display text
harmonic_xabcd_symbol() Get pattern symbol
Versionsinformation
⋅
v39
Versionsinformation
⋅
v40
Versionsinformation
⋅
v41
Versionsinformation
⋅
v42
Added: harmonic_xabcd_stop() Get stop level
Versionsinformation
⋅
v43 Fixed bug with cypher validation of point D (retracement of XC, not XA)
Versionsinformation
⋅
v44
Versionsinformation
⋅
v45 Removed redundant "==true" comparisons
Versionsinformation
⋅
v46 Fixed bug with cypher PRZ level
Versionsinformation
⋅
v47
Added: pat_xabcd_avgDev() Get the average % deviation of an XABCD pattern
Versionsinformation
⋅
v48
Versionsinformation
⋅
v49 Testing
Versionsinformation
⋅
v50 Testing
Versionsinformation
⋅
v51
Added: pat_xabcd_validate(xX, xY, aX, aY, bX, bY, cX, cY, dX, dY, pErr, pAsym) Validate custom XABCD pattern Parameters: xX: X coordinate of point X (int) xY: Y coordinate of point X (float) aX: X coordinate of point A (int) aY: Y coordinate of point A (float) bX: X coordinate of point B (int) bY: Y coordinate of point B (float) cX: X coordinate of point C (int) cY: Y coordinate of point C (float) dX: X coordinate of point D (int) dY: Y coordinate of point D (float) pErr: Acceptable percent error of leg ratios (does not apply to legs defined within a range) (float) pAsym: Acceptable percent asymmetry of leg ΔX (each leg tested against average ΔX of prior legs) (float) Returns: TRUE if pattern is valid
pat_xabcd_validateIncomplete(xX, xY, aX, aY, bX, bY, cX, cY, pErr, pAsym) Validate the first 3 legs of a custom XABCD pattern Parameters: xX: X coordinate of point X (int) xY: Y coordinate of point X (float) aX: X coordinate of point A (int) aY: Y coordinate of point A (float) bX: X coordinate of point B (int) bY: Y coordinate of point B (float) cX: X coordinate of point C (int) cY: Y coordinate of point C (float) pErr: Acceptable percent error of leg ratios (does not apply to legs defined within a range) (float) pAsym: Acceptable percent asymmetry of leg ΔX (each leg tested against average ΔX of prior legs) (float) Returns: TRUE if first 3 legs are valid
pat_xabcd_prz(xY, aY, bY, cY) Get the potential reversal zone (PRZ) levels of a custom XABCD pattern Parameters: xY: Y coordinate of point X (float) aY: Y coordinate of point A (float) bY: Y coordinate of point B (float) cY: Y coordinate of point C (float) Returns: [xad_lvl,bcd_lvl,xcd_lvl]
Versionsinformation
⋅
v52
Updated: harmonic_xabcd_eAvg() Get the avg retracement ratio % error
Versionsinformation
⋅
v53 Fixed bug with success(), extra checks on entry bar
Versionsinformation
⋅
v54
Added: tradeClosed() Determine if Target or Stop was hit on the current bar
Versionsinformation
⋅
v55 Bug fix for closed trade check
Versionsinformation
⋅
v56 Bug fix for closed trade check
Versionsinformation
⋅
v57
Updated: pat_xabcd(x_is_low, pivot_length, source, conf_length, incomplete) Determine if an XABCD pattern has just completed (i.e. point D is on the previous bar) Parameters: x_is_low: Flag to determine if point X is a low pivot, i.e. bullish pattern (bool, dft = true) pivot_length: Number of bars before and after a valid pivot (int, dft = 5) source: Source series (float, dft = na, will use high and low series) conf_length: Number of trailing bars after pivot point D to confirm a valid pattern (int, dft = 1) incomplete: Flag to return an incomplete XABC pattern (bool, dft = false) Returns: [flag,xx,xy,ax,ay,bx,by,cx,cy,dx,dy] flag = true if valid XABCD pattern completed on previous bar xx = X coordinate of point X (int) xy = Y coordinate of point X (float) ax = X coordinate of point A (int) ay = Y coordinate of point A (float) bx = X coordinate of point B (int) by = Y coordinate of point B (float) cx = X coordinate of point C (int) cy = Y coordinate of point C (float) dx = X coordinate of point D (int) dy = Y coordinate of point D (float)
Versionsinformation
⋅
v58 Short circuit incomplete XABCD pattern search
Versionsinformation
⋅
v59
Updated: harmonic_xabcd_przRange() Get upper and lower PRZ levels
Versionsinformation
⋅
v60
Added: harmonic_xabcd_entry()
Versionsinformation
⋅
v61
Added: xabcd_entryHit() Determine if entry level was reached. Assumes pattern is active/not timed out.
Versionsinformation
⋅
v62 Updated przClosest check to exclude values of 0 (only used when actual retracement value < 0)
Versionsinformation
⋅
v63 backed out last change
Versionsinformation
⋅
v64 test
Versionsinformation
⋅
v65 dont allow entry value < 0
Versionsinformation
⋅
v66 Updated to use latest version of Fib script (improved efficiency for precise fibs)
Versionsinformation
⋅
v67 test
Updated: pat_xabcd(x_is_low, pivot_length, source, conf_length, incomplete) Determine if an XABCD pattern has just completed (i.e. point D is on the previous bar) Parameters: x_is_low: Flag to determine if point X is a low pivot, i.e. bullish pattern (bool, dft = true) pivot_length: Number of bars before and after a valid pivot (int, dft = 5) source: Source series (float, dft = na, will use high and low series) conf_length: Number of trailing bars after pivot point D to confirm a valid pattern (int, dft = 1) incomplete: Flag to return an incomplete XABC pattern (bool, dft = false) Returns: [flag,xx,xy,ax,ay,bx,by,cx,cy,dx,dy] flag = true if valid XABCD pattern completed on previous bar xx = X coordinate of point X (int) xy = Y coordinate of point X (float) ax = X coordinate of point A (int) ay = Y coordinate of point A (float) bx = X coordinate of point B (int) by = Y coordinate of point B (float) cx = X coordinate of point C (int) cy = Y coordinate of point C (float) dx = X coordinate of point D (int) dy = Y coordinate of point D (float)
Versionsinformation
⋅
v68 Reverted test changes
Updated: pat_xabcd(x_is_low, pivot_length, source, conf_length, incomplete) Determine if an XABCD pattern has just completed (i.e. point D is on the previous bar) Parameters: x_is_low: Flag to determine if point X is a low pivot, i.e. bullish pattern (bool, dft = true) pivot_length: Number of bars before and after a valid pivot (int, dft = 5) source: Source series (float, dft = na, will use high and low series) conf_length: Number of trailing bars after pivot point D to confirm a valid pattern (int, dft = 1) incomplete: Flag to return an incomplete XABC pattern (bool, dft = false) Returns: [flag,xx,xy,ax,ay,bx,by,cx,cy,dx,dy] flag = true if valid XABCD pattern completed on previous bar xx = X coordinate of point X (int) xy = Y coordinate of point X (float) ax = X coordinate of point A (int) ay = Y coordinate of point A (float) bx = X coordinate of point B (int) by = Y coordinate of point B (float) cx = X coordinate of point C (int) cy = Y coordinate of point C (float) dx = X coordinate of point D (int) dy = Y coordinate of point D (float)
Versionsinformation
⋅
v69 Test allowing divergence checks with pivot length of 0
Versionsinformation
⋅
v70
Updated: pat_xabcd(x_is_low, pivot_length, source, conf_length, incomplete, inc_skipC) Determine if an XABCD pattern has just completed (i.e. point D is on the previous bar) Parameters: x_is_low: Flag to determine if point X is a low pivot, i.e. bullish pattern (bool, dft = true) pivot_length: Number of bars before and after a valid pivot (int, dft = 5) source: Source series (float, dft = na, will use high and low series) conf_length: Number of trailing bars after pivot point D to confirm a valid pattern (int, dft = 1) incomplete: Flag to return an incomplete XABC pattern (bool, dft = false) inc_skipC: skip incomplete patterns whose point C is at or before this bar index (to avoid returning the same pattern on consecutive bars) Returns: [flag,xx,xy,ax,ay,bx,by,cx,cy,dx,dy] flag = true if valid XABCD pattern completed on previous bar xx = X coordinate of point X (int) xy = Y coordinate of point X (float) ax = X coordinate of point A (int) ay = Y coordinate of point A (float) bx = X coordinate of point B (int) by = Y coordinate of point B (float) cx = X coordinate of point C (int) cy = Y coordinate of point C (float) dx = X coordinate of point D (int) dy = Y coordinate of point D (float)
Versionsinformation
⋅
v71
Versionsinformation
⋅
v72
Updated: pat_xabcd(x_is_low, pivot_length, source, conf_length, incomplete, inc_skipC) Determine if an XABCD pattern has just completed (i.e. point D is on the previous bar) Parameters: x_is_low: Flag to determine if point X is a low pivot, i.e. bullish pattern (bool, dft = true) pivot_length: Number of bars before and after a valid pivot (int, dft = 5) source: Source series (float, dft = na, will use high and low series) conf_length: Number of trailing bars after pivot point D to confirm a valid pattern (int, dft = 1) incomplete: Flag to return an incomplete XABC pattern (bool, dft = false) inc_skipC: skip incomplete patterns whose point C is at this bar index (to avoid returning the same pattern on consecutive bars) Returns: [flag,xx,xy,ax,ay,bx,by,cx,cy,dx,dy] flag = true if valid XABCD pattern completed on previous bar xx = X coordinate of point X (int) xy = Y coordinate of point X (float) ax = X coordinate of point A (int) ay = Y coordinate of point A (float) bx = X coordinate of point B (int) by = Y coordinate of point B (float) cx = X coordinate of point C (int) cy = Y coordinate of point C (float) dx = X coordinate of point D (int) dy = Y coordinate of point D (float)
Versionsinformation
⋅
v73
Versionsinformation
⋅
v74
Versionsinformation
⋅
v75
Updated: pat_xabcd(x_is_low, pivot_length, source, conf_length, incomplete) Determine if an XABCD pattern has just completed (i.e. point D is on the previous bar) Parameters: x_is_low: Flag to determine if point X is a low pivot, i.e. bullish pattern (bool, dft = true) pivot_length: Number of bars before and after a valid pivot (int, dft = 5) source: Source series (float, dft = na, will use high and low series) conf_length: Number of trailing bars after pivot point D to confirm a valid pattern (int, dft = 1) incomplete: Flag to return an incomplete XABC pattern (bool, dft = false) Returns: [flag,xx,xy,ax,ay,bx,by,cx,cy,dx,dy] flag = true if valid XABCD pattern completed on previous bar xx = X coordinate of point X (int) xy = Y coordinate of point X (float) ax = X coordinate of point A (int) ay = Y coordinate of point A (float) bx = X coordinate of point B (int) by = Y coordinate of point B (float) cx = X coordinate of point C (int) cy = Y coordinate of point C (float) dx = X coordinate of point D (int) dy = Y coordinate of point D (float)
Versionsinformation
⋅
v76
Updated: tradeClosed(eX, eY, stop, t1h, t2h, t1, t2) Determine if Target or Stop was hit on the current bar Parameters: eX eY stop t1h t2h t1 t2
Versionsinformation
⋅
v77
Updated: pat_xabcd(x_is_low, pivot_length, source, conf_length, incomplete) Determine if an XABCD pattern has just completed (i.e. point D is on the previous bar) Parameters: x_is_low: Flag to determine if point X is a low pivot, i.e. bullish pattern (bool, dft = true) pivot_length: Number of bars before and after a valid pivot (int, dft = 5) source: Source series (float, dft = na, will use high and low series) conf_length: Number of trailing bars after pivot point D to confirm a valid pattern (int, dft = 1) incomplete: Flag to return an incomplete XABC pattern (bool, dft = false) Returns: [flag, x, a, b, c, d] flag = true if valid XABCD pattern completed on previous bar x = point X a = point A b = point B c = point C d = point D
Versionsinformation
⋅
v78 test revert d.y to 0.0
Versionsinformation
⋅
v79 Reverted changes using point object for pat_xabcd()
Updated: pat_xabcd(x_is_low, pivot_length, source, conf_length, incomplete) Determine if an XABCD pattern has just completed (i.e. point D is on the previous bar) Parameters: x_is_low: Flag to determine if point X is a low pivot, i.e. bullish pattern (bool, dft = true) pivot_length: Number of bars before and after a valid pivot (int, dft = 5) source: Source series (float, dft = na, will use high and low series) conf_length: Number of trailing bars after pivot point D to confirm a valid pattern (int, dft = 1) incomplete: Flag to return an incomplete XABC pattern (bool, dft = false) Returns: [flag,xx,xy,ax,ay,bx,by,cx,cy,dx,dy] flag = true if valid XABCD pattern completed on previous bar xx = X coordinate of point X (int) xy = Y coordinate of point X (float) ax = X coordinate of point A (int) ay = Y coordinate of point A (float) bx = X coordinate of point B (int) by = Y coordinate of point B (float) cx = X coordinate of point C (int) cy = Y coordinate of point C (float) dx = X coordinate of point D (int) dy = Y coordinate of point D (float)
Versionsinformation
⋅
v80 Add stop flag to tradeClosed()
Versionsinformation
⋅
v81
Added: test_cd(cd, bc, xa, xc, ad, pErr, p_types) Validate CD leg of XABCD Parameters: cd bc xa xc ad pErr p_types
Versionsinformation
⋅
v82
Added: pat_xabcdIncomplete(x_is_low, pivot_length, source, conf_length) Determine if an XABCD pattern is in progress (point C was just confirmed) Parameters: x_is_low: Flag to determine if point X is a low pivot, i.e. bullish M pattern (bool, dft = true) pivot_length: Number of bars before and after a valid pivot (int, dft = 5) source: Source series (float, dft = na, will use high and low series) conf_length: Number of trailing bars after pivot point D to confirm a valid pattern (int, dft = 1) Returns: [flag,xx,xy,ax,ay,bx,by,cx,cy] flag = true if valid XABC pattern completed on bar_index[conf_length] xx = X coordinate of point X (int) xy = Y coordinate of point X (float) ax = X coordinate of point A (int) ay = Y coordinate of point A (float) bx = X coordinate of point B (int) by = Y coordinate of point B (float) cx = X coordinate of point C (int) cy = Y coordinate of point C (float) dx = X coordinate of point D (int) dy = Y coordinate of point D (float)
This script is very nice, but there is one problem with calculating the "Shark Pattern". According to Harmonic Theory, "The Shark Pattern" doesn't require checking the validity of the retracement value of leg AB (or XA) against leg XA (or OX), but in practice there is a problem: In a particular shape called XABCD, if three retracement values are determined, mathematically the remaining one must be within a certain range! We need to define it in the script to capture the pattern correctly. It's probably between 0.382 and 0.618. Thank you for all your hard work.