diff --git a/freqtrade/data/btanalysis.py b/freqtrade/data/btanalysis.py
index 4db88332d..90c228833 100644
--- a/freqtrade/data/btanalysis.py
+++ b/freqtrade/data/btanalysis.py
@@ -412,30 +412,36 @@ def calculate_mdd(data: dict, trades: pd.DataFrame, *, date_col: str = 'close_da
     
     for pair, df in data.items():
         # Init colonne boolean à 0
-        df_bis = df.copy()
-        df_bis["is_into_trade"] = np.zeros(len(df_bis), dtype=bool)
+        # df_bis = df.copy()
+        # df_bis["is_into_trade"] = np.zeros(len(df_bis), dtype=bool)
         
         # Récupération des index
-        open_close_date = trades.loc[trades['pair']==pair][["open_date","close_date"]]
-        mdd_df = df[df["date"].isin([open_close_date["open_date"],open_close_date["close_date"]])]
-        l_open=df[df["date"].isin(open_close_date["open_date"])].index.values.tolist()
-        l_close=df[df["date"].isin(open_close_date["close_date"])].index.values.tolist()
+        # open_close_date = trades.loc[trades['pair']==pair][["open_date","close_date"]]
+        # mdd_df = df[df["date"].isin([open_close_date["open_date"],open_close_date["close_date"]])]
+        # l_open=df[df["date"].isin(open_close_date["open_date"])].index.values.tolist()
+        # l_close=df[df["date"].isin(open_close_date["close_date"])].index.values.tolist()
         
         # Set 1 aux bornes open et close dans le df
-        df_bis["is_into_trade"].iloc[l_open] = 1
-        df_bis["is_into_trade"].iloc[l_close] = 1
+        # df_bis["is_into_trade"].iloc[l_open] = 1
+        # df_bis["is_into_trade"].iloc[l_close] = 1
         
         # cumulative sum modulo 2
-        df_bis["is_into_trade"] = df_bis["is_into_trade"].cumsum()&1
+        # df_bis["is_into_trade"] = df_bis["is_into_trade"].cumsum()&1
         
         # close boundaries set to 1 (again)
-        df_bis["is_into_trade"].iloc[l_close] = 1
+        # df_bis["is_into_trade"].iloc[l_close] = 1
 
-        # trades_pair = trades.loc[trades['pair']==pair][["open_date","close_date"]]
-        # trades_pair = trades_pair['is_in_trade'] = np.ones(len(trades_pair))
-        # trades_list = pd.concat([trades_pair.rename(columns={'open_date':'date'})[['date','is_in_trade']],trades_pair.rename(columns={'close_date':'date'})[['date','is_in_trade']]]).sort_values(by='date')
-        # trades_join = df.set_index('date').join(trades_list.set_index('date'))
-        # trades_join = trades_join.fillna(0)
+        trades_aux = trades.loc[trades['pair']==pair][["open_date","close_date"]]
+        trades_aux['open_close_mark'] = np.ones(len(trades_aux))
+        trades_aux = pd.concat([trades_aux.rename(columns={'open_date':'date'})[['date','open_close_mark']],trades_aux.rename(columns={'close_date':'date'})[['date','open_close_mark']]]).sort_values(by='date')
+        data_join = df.set_index('date').join(trades_aux.set_index('date'))
+        data_join = data_join.fillna(0)
+        data_join['is_in_trade'] = data_join.open_close_mark.cumsum()%2
+        data_join['is_in_trade'].loc[data_join['open_close_mark'] == 1] = 1
+        data_join['close_cummax'] = 0
+
+        data_join['close_cummax'] = data_join.groupby(data_join.open_close_mark.cumsum()).close.cummax()
+        data_join['close_cummax'].loc[data_join['is_in_trade'] == 0] = 0
         
         # print(mdd_df)
     return mdd_df