Troubleshooting Differences Between Python and TradingView STC

# TradingView STC vs any python STC


# TradingView STC vs any python STC












I am trying to use in a trading strategy the STC indicator, but I can not find out why its not working properly.

The chart that I am using is BTC/USDT on UTC as a timeframe.

Chart time: 01 Feb 22 - 16:20 UTC

------------------- TradingView: ------------------------

STC value: 97.66

STC settings:

---------------- Python: ----------------

I've tried the following libraries:

Pands ta(link):

```
dataframe.ta.stc(tclength=12, fast=26, slow=50, factor=0.5, append=True)
```

Technical indicators(link)

```
dataframe['stc_2'] = technical.indicators.stc(dataframe, fast=26, slow=50, length=12)
```

Financial Technical Analysis(link)

```
dataframe['stc'] = fta.STC(dataframe, period_fast=26, period_slow=50, k_period=12, d_period=3, adjust=True)
```

And I've also tried to recreate the indicator by converting the pine script from here to python

```
def stoch(source, high, low, lenght):
    return Series(100 * (source - low[-lenght:].min()) / (high[-lenght:].max() - low[-lenght:].min()))

def fixnan(s: Series):
    mask = np.isnan(s)
    s[mask] = np.interp(np.flatnonzero(mask), np.flatnonzero(~mask), s[~mask])
    return s

def nz(s: Series):
    return s.fillna(0)

def stc(ohlc: DataFrame, fast: int, slow: int, length: int, d1: int, d2: int):
    macd = ta.EMA(ohlc['close'], timeperiod=fast) - ta.EMA(ohlc['close'], timeperiod=slow)
    k = nz(fixnan(stoch(macd, macd, macd, length)))
    d = ta.EMA(k, d1)
    kd = nz(fixnan(stoch(d, d, d, length)))
    stc = ta.EMA(kd, d2)
    r1 = np.where(stc >= 100, 100, stc)
    r2 = np.where(r1 <= 0, 0, r1)
    return r2

dataframe['stc_MINE'] = stc(dataframe, 26, 50, 10, 3, 3)
```

Here is the output from all of them:

As can be seen, none of them is 97.66, could anyone explain to me what I did wrong or what am I missing?

My code:

In order to gather all the data and use the indicators I've used freqtrade

To download the data I've used:

```
freqtrade download-data -t 5m --pairs BCH/USDT --erase --timerange 20210101-20220101
```

And here is my strategy that I've used:

```
class Mine(IStrategy):
timeframe = '5m'

def populate_indicators(self, dataframe: DataFrame, metadata: dict) -> DataFrame:
    dataframe['stc'] = fta.STC(dataframe, period_fast=26, period_slow=50, k_period=12, d_period=3, adjust=True)
    dataframe['stc_2'] = technical.indicators.stc(dataframe, fast=26, slow=50, length=12)
    dataframe.ta.stc(tclength=12, fast=26, slow=50, factor=0.5, append=True)
    dataframe['stc_MINE'] = stc(dataframe, 26, 50, 10, 3, 3)

    return dataframe

def populate_buy_trend(self, dataframe: DataFrame, metadata: dict) -> DataFrame:
    return dataframe

def populate_sell_trend(self, dataframe: DataFrame, metadata: dict) -> DataFrame:
    return dataframe
```

Both exchanges from there the data are downloaded in TradingView and my strategy is Binance

## Answer by Lamari Alaa (score 0)

https://quant.stackexchange.com/a/70976

Maybe your stoch function is not good

try this:

```
def stoch(d, l):
    highest= d.rolling(l).max()
    lowest = d.rolling(l).min()
    stoch = 100 * (d - lowest) / (highest - lowest)
    return stoch
```