Verification of ghost storm tracks of September-October 2018

I falsely detected (sub)tropical cyclogenesis in Atlantic 39 times in September and October 2018. Tracks were made of 14 systems.

At the lead time of 1 days (24 hours), there were 2 false detections.
At the lead time of 2 days (48 hours), there were 4 false detections.
At the lead time of 3 days (72 hours), there were 2 false detections.
At the lead time of 4 days (96 hours), there were 2 false detections.
At the lead time of 5 days (120 hours), there were 2 false detections.
At the lead time of 6 days (144 hours), there were 1 false detections.
At the lead time of 7 days (168 hours), there were 4 false detections.
At the lead time of 8 days (192 hours), there were 3 false detections.
At the lead time of 9 days (216 hours), there were 7 false detections.
At the lead time of 10 days (240 hours), there were 10 false detections.
At the lead time of 11 days (264 hours), there were 13 false detections.
At the lead time of 12 days (288 hours), there were 14 false detections.
At the lead time of 13 days (312 hours), there were 16 false detections.
At the lead time of 14 days (336 hours), there were 19 false detections.
At the lead time of 15 days (360 hours), there were 19 false detections.
At the lead time of 16 days (384 hours), there were 20 false detections.

Circle coloration scheme

Ghost storm #51

 These tracks began in August already.

September 1, 12Z run:

September 2, 12Z run:

September 4, 12Z run:

September 5, 06Z run:

 September 7, 12Z run:

September 10, 12Z run:

System #52 was not a ghost storm. Rather, it was Gordon.

 Ghost storm #53

A track of this system was made in August already.

 September 2, 12Z run:

System #54 was not a ghost storm. Rather, it was TD 11.

System #55 was not a ghost storm. Rather, it was Leslie

System #56 was not a ghost storm. Rather, it was Joyce.

System #57 was not a ghost storm. Rather, it was Michael.

Ghost storm #58

  This isolated incident is not identifiable due to large lead time.

September 14, 12Z run:

System #59 was not a ghost storm. Rather, it was an assumed regeneration of Michael, otherwise represented by #57.

Ghost storm #60

  

This is best identifiable with a wave, here near 10N 30W.

September 15, 18Z run:

September 16, 06Z run:

September 16, 18Z run:

September 17, 12Z run:

September 18, 06Z run:

September 19, 06Z run:

System #61 was not a ghost storm. Rather, it was Kirk.

System #62 was not a ghost storm. Rather, it was the regeneration of Leslie.

Ghost storm #63

This system gradually became identifiable with this complex low pressure area near Azores.

 September 23, 12Z run:

September 25, 06Z run:

September 26, 06Z run:

October 4, 06Z run:

Ghost storm #64

 I have no idea what this is.

September 25, 06Z run:

September 26, 06Z run:

October 1, 06Z run:

System #65 was not a ghost storm. Rather, it was Nadine.

Ghost storm #66

  This is identifiable with this low, marked just south of Azores.

October 1, 06Z run:

 

October 4, 06Z run:

 October 15, 12Z run:

 October 18, 06Z run:

October 23, 06Z run:

Ghost storm #67

This isolated incident is identifiable with a wave, marked here south of Cape Verde Islands.

  

 October 2, 12Z run:

Ghost storm #68

 This is best identifiable with a wave, partially visible here in western Caribbean.

October 7, 06Z run:

October 8, 06Z run:

October 9, 06Z run:

October 11, 06Z run:

System #69 was not a ghost storm. Rather, it was Oscar.

Ghost storm #70

 This is not identifiable.

October 9, 06Z run:

  October 14, 18Z run:

 Ghost storm #71

 This is not identifiable, due to long lead time, and proximity to Oscar.

October 16, 18Z run:

October 17, 06Z run:

October 18, 06Z run:

Ghost storm #72

 A tropical cyclone forms just off Ivorian coast, and becomes a Cat 5 equatorward of 4 N ? Really ? Ridiculous.

October 21, 06Z run:

 Ghost storm #73

 I have no idea about identity of this.

October 24, 06Z run:

Ghost storm #74

Wut

  

October 25, 06Z run: