How to Take a Baseline
A reading with nothing to compare it to is not a reading. Everything the Survey knows about a building it knows because somebody spent thirty boring minutes at the start writing down what normal looked like.
Almost everybody skips this. It happens with the lights on, it produces no footage, it takes half an hour that everyone would rather spend in the dark, and nothing in it is exciting.
It is also the entire difference between an investigation and an evening out. A reading of 4.2 means nothing. A reading of 4.2 in a room that read 1.1 forty minutes ago means something. A reading of 4.2 in a room that reads 4.2 all night means you have found a fridge.
What a baseline is for
Three things, and it is worth being clear which is which.
It gives you a comparison. This is the obvious one. You cannot call something an excursion without a floor to have excursed from.
It tells you how much the floor moves. This is the one people miss. A position that reads 1.1 and holds is a different position from one that reads 1.1 on average while swinging between 0.4 and 2.6. In the second room, a reading of 2.5 later in the night is nothing at all, and if you only took one instantaneous figure at the start you have no way of knowing that.
It finds your ordinary explanations before you need them. By the end of a proper baseline you know where the consumer unit is, when the fridge cycles, which floorboard creaks, where the draught from the letterbox reaches, and which room has a streetlight coming through the gap in the curtain at head height. All of those will otherwise be discovered at half past one in the morning by somebody frightened.
The procedure
-
Draw the plan first. Rough, by hand, one page per floor. Number the rooms. You will use these numbers for the rest of the night and in the write-up.
-
Mark fixed positions. Three to six per room, marked on the plan and physically marked in the room with tape if the owner permits it. A baseline taken "in the kitchen" is not repeatable. A baseline taken at position K3, 1.2 m above the floor at the corner of the worktop, is.
-
Measure with the lights on and the building normal. Do not turn things off. You are recording what the building does, not what it does under test conditions.
-
Ninety seconds minimum per position, per instrument. Watch the figure settle and record the range as well as the value.
1.1 (0.9–1.4)is a baseline.1.1is a guess. -
Record the compass bearing you were facing. Magnetic measurements are directional. This takes two seconds and saves arguments.
-
Note the time and the state of the building. Heating on or off, boiler firing, which appliances are running, who else is in it, what the weather is doing.
-
Do it again at the end of the night, same positions, same order, same instruments.
| Position | Room number and a description precise enough for somebody else to stand there. |
|---|---|
| Time | To the minute. Timestamps are what let you line the log up with recordings afterwards. |
| Magnetic | Value and range over the window. Note the axis or the bearing you were facing. |
| Temperature | Air temperature, from a probe left to equilibrate. Not an infrared surface reading. |
| Sound | Level in dBA if you have a meter, and a written note of what you can hear if you do not. |
| Light | Whether there is any, and where it comes from. Streetlights move with traffic and cloud. |
| State | Heating, appliances, occupancy, weather, and anything running with a duty cycle. |
Why the second baseline is the good one
Take the same positions at the end and you get something no single measurement can give you: a drift figure for the whole building over the session.
A house cooling from 19 °C to 14 °C over five hours will produce cold spots, timber movement, and a slow rise in the relative significance of every small sound, and it will do it whether or not anything else is happening. A building whose mains loading has dropped as the neighbours go to bed will read differently on a coil meter at 3am than at 9pm for entirely municipal reasons.
If you have the two baselines you can subtract the evening out. If you have one, everything that happened is confounded with the fact that it got later.
Doing it with a phone
The detector on this station will take a baseline for you. It is not a substitute for a proper instrument — it reads the scatter in your compass heading rather than field strength — but it is genuinely measuring the room, it calibrates against a window you choose, and it files the floor, the bearing rose and the session length into your service record so that the figure is still there in six months.
That last part is the argument for doing it at all. A room whose floor was 1.83° in March and 1.83° in November is a room you now know something about. It is also, as it happens, the most unsettling file in our archive, for reasons that have nothing to do with ghosts and everything to do with what a floor is supposed to do.
Questions
- What is a baseline in a paranormal investigation?
- A set of measurements taken before the investigation begins, recording what the building normally reads: magnetic field, temperature, humidity, sound level and light, at fixed marked positions. Everything measured later is compared against it.
- How long should a baseline take?
- Between twenty and forty minutes for an ordinary house, and it is done with the lights on before anybody starts. Each position needs at least ninety seconds of settled reading, because a single instantaneous figure tells you nothing about how much that position normally varies.
- Do I need to repeat the baseline?
- Yes — at the end of the night, at the same positions, in the same order. A building at 2am is not the same building it was at 9pm, and the difference between the two baselines is often the most informative measurement of the session.