Mediocre System Bias [Space]

[ericthered]

Originally Posted by Anders

If I were to design a system in Space, I would first determine a world I wanted to place there and then generate the rest of the system randomly. Presumably you have some idea of what you want before you start. If not, that's a good place to start.

I usually get a computer to use the system to generate a large number of systems for me, then go look at the interesting places. Most systems aren't interesting, but then most systems shouldn't be.

If I didn't have a computer to do the heavy lifting, I'd probably only generate systems as necessary.

[Joseph Paul]

Isn't the major detection system we use looking for 'wobble' as a planet transits the primary which means we can only check on systems that are
edge on to us which is unlikely to be a large enough sample to establish norms?

[Polydamas]

Originally Posted by AlexanderHowl

And I would not have any complaints if the average system was interesting, but the average system is not interesting (see the first post). If the average system is not interesting, at least they could have made it possible to create the Sol System. In its current form though, it kind of fails at both.

The average system is not the one with 10 or 11 on 3d, 7 on 2d, and 4 on 1d. A system generated that way is not the median system either. There is nothing special about those numbers, a 7 on 2d is just a way to assign a probability of 6/36 to an outcome.

Why don't you generate 5-10 systems at random, present a short description, and argue whether any of them are interesting?

[ericthered]

Originally Posted by Joseph Paul

Isn't the major detection system we use looking for 'wobble' as a planet transits the primary which means we can only check on systems that are edge on to us which is unlikely to be a large enough sample to establish norms?

Well, we're talking about a few different things here.

The first method that allowed us to detect exoplanets is the wobble method, where we watch for position changes in the stars location. This method is biased towards big planets and close planets, which gives us a lot of "Hot Jupiters". It does not require the planet to transit the primary, though its not exclusive with that.

The most common method we use to find exoplanets is the transit method. It watches the star for repeated, regular dimming of its light. Its also easier to detect planets closer to the star using this method.

We've found a lot of planets out there using these two methods. Over 4,000, which is enough to make predictions with a lot more confidence than the previous 8 planets. The trick is that most of the 4,000 are the only planet identified around their star so far, and the data is biased in some tricky ways.

I expect in the next decade we'll see some firmer theories about planet formation show up. The scientists just need some time to digest all that data. Also, I feel the need to grumble about delays in the James Webb Telescope.

[AlexanderHowl]

The problem with the wobble method is that it works best with very small stars orbited by very close and very large objects (the more massive, the better for detection), which is why we see lots of M-class stars orbited by Hot Jupiters.

The problem with the transit method is that the planets in question have to pass between their star and us, in order to produce a transit, which should only statistically possible in less than 1% of orbits (due to the fact that the orbit of the planet has to transit across their star in a way that we can see). If we were trying to see the Earth through a transit method from another star, we would only have a ((800,000÷93,000,000)×100)%, or 0.86%, chance of being in the correct orientation to see the Earth, we would need a device capable of detecting a 0.01% drop in the relative brightness of Sol, and, since we would could only detect it for 12 hours every year, we would need to look at the star a minimum of one minute out of every 12 hours.

When you look at the odds of detecting any transit, it is a miracle that we ever detect anything, which is why the transit method is also biased towards very close and very large objects. A Hot Jupiter at 0.1 AU around a G2 star would have a correct orientation probability of 8.6%, would cause a 4% drop in relative brightness (it would puff out to a diameter of 160,000 miles due to the heat), and would transit for four hours every 12 days. It is around 40,000x more likely that we would detect the Hot Jupiter than the Earth around a G2 star.

[ericthered]

Originally Posted by Polydamas

Why don't you generate 5-10 systems at random, present a short description, and argue whether any of them are interesting?

I did it (using a prexisting generator). I generated 10 systems. Tabs indicate binary systems. I listed worlds with a positive affinity or resource value above 2. I'd say that half of the systems had something work digging into deeper.

white dwarf with mass .85 (formerly G8) 
K8 with mass .55 
	M1 with mass .45
M3 with mass .35 -- hot Jupiter
	M7 with mass .1 
M4 with mass .1 -- cold standard garden with very thin atmosphere with breathable pollutants, 70% water, resource value 0, affinity 4
M7 with mass .1 -- Very cold  Large garden with very dense breathable pollutants atmosphere, 90% water, resource value 0, affinity 2 AND resource 2 asteroid belt
M6 with mass .15 -- Warm asteroid (resource value 0)
M4 with mass .3 
	M4 with mass .3 -- Standard tropical gardenwith breathable pollutants, 60% water, resource value 0, affinity 7
K0 with mass .8 -- standard chilly ocean and resource value 1
M3 with mass .35 -- standard infernal world with low oxyen at standard pressure, 80% water, affinity 5
	M7 with mass .1 -- frozen astroids with resource 2,
M0 with mass .5 -- tiny Infernal rock planet with resource 2, Standard cold thin ocean,

I had a planet slot in the goldilocks zone most times. Most of the garden worlds had fairly low affinity scores, but that can be dealt with. I also had a lot of red dwarfs. My experience with this generator tells me that this is a comparatively rich run, as far as garden worlds go. Usually only about 20% of worlds give me gardens (which meant about 40% of systems have ocean or garden worlds). On the other hand, I feel like it was a weak run for resources. And the white dwarf was something of a surprise.

I do a 50/50 split between garden and ocean -- your setting should set that "toggle" to what you want. If you've set it to "all ocean worlds" I'm not sure what you're hoping to find in each system.

[AlexanderHowl]

Correction to the previous post, the probability for proper orientation for transit for Earth detection is ((400,000)^2/(93,000,000)^2)×100%, or 0.00185%, not 0.86%, around 460x less than previously suggested. The probability for proper orientation for transit for Hot Jupiter transit is 0.185%, meaning that it is 400,000x more likely to detect a Hot Jupiter of Jupiter-mass at 0.1 AU from a G2 star than it is to detect a Earth equivalent around a G2 star. As I said, it is a miracle we find anything sometimes.

[David Johnston2]

Originally Posted by ericthered

I did it (using a prexisting generator). I generated 10 systems. Tabs indicate binary systems. I listed worlds with a positive affinity or resource value above 2. I'd say that half of the systems had something work digging into deeper.

white dwarf with mass .85 (formerly G8) 
K8 with mass .55 
	M1 with mass .45
M3 with mass .35 -- hot Jupiter
	M7 with mass .1 
M4 with mass .1 -- cold standard garden with very thin atmosphere with breathable pollutants, 70% water, resource value 0, affinity 4
M7 with mass .1 -- Very cold  Large garden with very dense breathable pollutants atmosphere, 90% water, resource value 0, affinity 2 AND resource 2 asteroid belt
M6 with mass .15 -- Warm asteroid (resource value 0)
M4 with mass .3 
	M4 with mass .3 -- Standard tropical gardenwith breathable pollutants, 60% water, resource value 0, affinity 7
K0 with mass .8 -- standard chilly ocean and resource value 1
M3 with mass .35 -- standard infernal world with low oxyen at standard pressure, 80% water, affinity 5
	M7 with mass .1 -- frozen astroids with resource 2,
M0 with mass .5 -- tiny Infernal rock planet with resource 2, Standard cold thin ocean,

I had a planet slot in the goldilocks zone most times. Most of the garden worlds had fairly low affinity scores, but that can be dealt with. I also had a lot of red dwarfs. My experience with this generator tells me that this is a comparatively rich run, as far as garden worlds go. Usually only about 20% of worlds give me gardens (which meant about 40% of systems have ocean or garden worlds). On the other hand, I feel like it was a weak run for resources. And the white dwarf was something of a surprise.

I do a 50/50 split between garden and ocean -- your setting should set that "toggle" to what you want. If you've set it to "all ocean worlds" I'm not sure what you're hoping to find in each system.

There's something a bit wonky about those values. Manually rolling systems from scratch I find it pretty much impossible to get into the goldilocks zone with all but the largest red dwarfs.

[ericthered]

Originally Posted by David Johnston2

There's something a bit wonky about those values. Manually rolling systems from scratch I find it pretty much impossible to get into the goldilocks zone with all but the largest red dwarfs.

As in its too narrow, or some other problem?

[David Johnston2]

Originally Posted by ericthered

As in its too narrow, or some other problem?

I can't get them close enough without fudging.