diff --git a/ctaplot/gammaboard/gammaboard.py b/ctaplot/gammaboard/gammaboard.py
index 555e344..e897198 100644
--- a/ctaplot/gammaboard/gammaboard.py
+++ b/ctaplot/gammaboard/gammaboard.py
@@ -11,7 +11,7 @@
 from ipywidgets import HBox, Tab, Output, VBox, FloatSlider, Layout, Button, Dropdown, Text, Label
 from sklearn.metrics import roc_auc_score, precision_recall_curve
 from sklearn.multiclass import LabelBinarizer
-from astropy.table import Table
+from astropy.table import Table, vstack
 import astropy.units as u
 from tqdm.auto import tqdm
 from .. import plots
@@ -83,20 +83,19 @@ def load_trig_events(experiment, experiments_directory):
     result_files = find_data_files(experiment, experiments_directory)
     trig_energies = []
     for file in result_files:
-        try:
-            result_file = tables.open_file(file)
-        except Exception as e:
-            print('Could not open data file {}'.format(file))
-            print(e)
-        else:
+        if guess_particle_type_from_file(file) == GAMMA_ID:
             try:
-                if guess_particle_type_from_file(file) == GAMMA_ID:
-                    result_file.close()
-                    trig_energies.append(pd.read_hdf(file, key='triggered_events'))
-            except:
-                print("Cannot load the number of triggered events for experiment {} file".format(experiment))
-                return None
-    energies = Table.from_pandas(pd.concat(trig_energies))
+                dl1_params = Table.read(file, path='dl1/event/telescope/parameters/LST_LSTCam')
+                _, indices = np.unique(dl1_params[['obs_id', 'event_id']], axis=0, return_index=True)
+                trig_energies = dl1_params['mc_energy'][indices]
+            except OSError:
+                try:
+                    trig_energies.append(Table.from_pandas(pd.read_hdf(file, key='triggered_events')))
+                except KeyError as e:
+                    print("Cannot load the number of triggered events for experiment {} file".format(experiment))
+                    return None
+    energies = vstack(trig_energies)
+    energies.rename_column('mc_energy', 'mc_trig_energies')
     energies['mc_trig_energies'] *= u.TeV
     return energies