CSV Export | Smart Swing Tools Docs

MTTP's CSV export provides hierarchical swing data across all timeframes. Unlike SSI PRO's single-timeframe export, MTTP exports the complete multi-timeframe structure with cycle numbering.

Enabling Export

Open MTTP settings
Find CSV Export section
Enable Enable CSV Export Mode
Optionally enable CSV Data Only if you don't need visual labels

Once enabled, additional columns appear in TradingView's data window and are included when you export chart data.

Exported Columns

For each timeframe, MTTP exports five data points:

Column	Description	Example
{TF}_Cycle	Current cycle number within parent timeframe	3
{TF}_Phase	Direction: 0=Bottom, 1=Top	1
{TF}_Price	Price at the swing extreme	4525.75
{TF}_Formation_Time	Unix timestamp (milliseconds) when extreme occurred	1731801600000
{TF}_Bar_Index	Bar index where extreme occurred	2847

Timeframe prefixes:

Y = Yearly
Q = Quarterly
M = Monthly
W = Weekly
D = Daily
H4 = 4-Hour
H1 = 1-Hour
M15 = 15-Minute

Complete Column List

Y_Cycle, Y_Phase, Y_Price, Y_Formation_Time, Y_Bar_Index
Q_Cycle, Q_Phase, Q_Price, Q_Formation_Time, Q_Bar_Index
M_Cycle, M_Phase, M_Price, M_Formation_Time, M_Bar_Index
W_Cycle, W_Phase, W_Price, W_Formation_Time, W_Bar_Index
D_Cycle, D_Phase, D_Price, D_Formation_Time, D_Bar_Index
H4_Cycle, H4_Phase, H4_Price, H4_Formation_Time, H4_Bar_Index
H1_Cycle, H1_Phase, H1_Price, H1_Formation_Time, H1_Bar_Index
M15_Cycle, M15_Phase, M15_Price, M15_Formation_Time, M15_Bar_Index
Full_Hierarchy

Understanding Cycle Numbers

Cycle numbers count swings within their parent timeframe:

Example progression:

Year 1 begins
  Quarter 1: Q_Cycle = 1
  Quarter 2: Q_Cycle = 2
  Quarter 3: Q_Cycle = 3
  Quarter 4: Q_Cycle = 4
Year 2 begins → Q_Cycle resets to 1
  Quarter 1: Q_Cycle = 1
  ...

The same logic applies down the hierarchy:

Monthly cycles reset each quarter
Weekly cycles reset each month
Daily cycles reset each week
And so on for intraday timeframes

Hierarchical Tags

Internally, MTTP builds hierarchical tags like:

Y2.Q4.M3.W8.D5.H4-2.H1-3.M15-7

This reads as:

Year cycle 2
Quarter 4 within that year
Month 3 within that quarter
Week 8 within that month
Day 5 within that week
4-Hour swing 2 within that day
1-Hour swing 3 within that 4-hour period
15-Minute swing 7 within that hour

The exported columns break this apart into individual components for easier filtering and analysis.

The Full_Hierarchy Column

A composite number encoding all cycle positions:

Full_Hierarchy = Y * 100000000 + Q * 10000000 + M * 1000000 + 
                 W * 100000 + D * 10000 + H4 * 1000 + H1 * 100 + M15

Example:

Y=2, Q=4, M=3, W=8, D=5, H4=2, H1=3, M15=7
Full_Hierarchy = 243850237

This single number uniquely identifies the hierarchical position, useful for sorting and comparing structural positions across different bars.

Phase Interpretation

The Phase column indicates whether the last confirmed swing was a top or bottom:

Phase	Meaning	Current Direction
0	Last swing was a Bottom (low)	Moving UP from that low
1	Last swing was a Top (high)	Moving DOWN from that high

Combined with the cycle number, you know both where you are in the structure and which direction is currently active.

How to Export

Enable CSV Export Mode in settings
Click the camera/snapshot icon in TradingView's toolbar
Select Export chart data
Choose CSV format
Open in Excel, Google Sheets, or your analysis tool

Working with the Data

Filtering to Swing Rows

Most rows will have NaN values for swing columns (swings only occur at specific bars). Filter to rows where at least one {TF}_Price column has data:

Excel:

=NOT(ISNA(Y_Price)) OR NOT(ISNA(Q_Price)) OR NOT(ISNA(M_Price)) ...

Python (pandas):

python

swing_rows = df[df[['Y_Price', 'Q_Price', 'M_Price', 'W_Price', 'D_Price']].notna().any(axis=1)]

Converting Timestamps

Formation timestamps are Unix milliseconds. To convert:

Python:

python

import pandas as pd
df['Y_Date'] = pd.to_datetime(df['Y_Formation_Time'], unit='ms')

JavaScript:

javascript

const date = new Date(row.Y_Formation_Time);

Excel:

=(A2/86400000)+DATE(1970,1,1)

Building Cycle Sequences

To analyze how cycles progress:

python

# Get all weekly swings
weekly_swings = df[df['W_Price'].notna()][['time', 'W_Cycle', 'W_Phase', 'W_Price']]

# Calculate swing-to-swing relationships
weekly_swings['prev_price'] = weekly_swings['W_Price'].shift(1)
weekly_swings['range'] = abs(weekly_swings['W_Price'] - weekly_swings['prev_price'])

CSV Data Only Mode

When CSV Data Only (No Labels) is enabled:

Swing detection runs normally
Cycle data is collected and exported
No visual labels are created

Benefits:

Bypasses the 500 label limit entirely
No dynamic window scanning overhead
Faster chart performance
Clean chart with data-only collection

Use case: You want MTTP's hierarchical data for external analysis but don't need labels cluttering your chart.

Date Filter

MTTP applies a 3-year recency filter to exported data. Swings older than 3 years from the current time are excluded from the CSV state variables.

This prevents ancient historical swings from persisting indefinitely in the export columns while ensuring recent and relevant data is captured.

Combining with SSI PRO Export

For comprehensive swing analysis, you can run both indicators and export both:

SSI PRO CSV: Detailed single-timeframe swing data with all 16 metrics
MTTP CSV: Multi-timeframe hierarchical position data

Merge by timestamp or bar_index to create a unified dataset showing both detailed swing metrics and structural context.

Database Integration

The exported data works well for database storage:

Suggested Schema

sql

CREATE TABLE mttp_swings (
    id SERIAL PRIMARY KEY,
    symbol VARCHAR(20),
    bar_time TIMESTAMP,
    bar_index INTEGER,
    
    -- Yearly
    y_cycle INTEGER,
    y_phase INTEGER,
    y_price DECIMAL(18,8),
    y_formation_time BIGINT,
    
    -- Quarterly
    q_cycle INTEGER,
    q_phase INTEGER,
    q_price DECIMAL(18,8),
    q_formation_time BIGINT,
    
    -- Monthly
    m_cycle INTEGER,
    m_phase INTEGER,
    m_price DECIMAL(18,8),
    m_formation_time BIGINT,
    
    -- Weekly
    w_cycle INTEGER,
    w_phase INTEGER,
    w_price DECIMAL(18,8),
    w_formation_time BIGINT,
    
    -- Daily
    d_cycle INTEGER,
    d_phase INTEGER,
    d_price DECIMAL(18,8),
    d_formation_time BIGINT,
    
    -- Full hierarchy
    full_hierarchy BIGINT,
    
    UNIQUE(symbol, bar_index)
);

Query Examples

Find all Weekly bottoms:

sql

SELECT * FROM mttp_swings 
WHERE w_phase = 0 AND w_price IS NOT NULL;

Find convergence (Weekly + Monthly bottom at same bar):

sql

SELECT * FROM mttp_swings 
WHERE w_phase = 0 AND m_phase = 0 
  AND w_price IS NOT NULL AND m_price IS NOT NULL;

Limitations

2-Day and 3-Day swings are not currently included in CSV export (they use a different detection system)
Historical data beyond 3 years is filtered out
Chart must load sufficient history for higher timeframe swings to appear

Next Steps

Settings — Configure export options
Overview — Return to core concepts