≡ Menu

Hitting the Exoplanet Jackpot

If by any chance you missed Lee Billings’ recent work on BoingBoing, let me direct you to Cosmic Commodities: How Much is a New Planet Worth? Lee has been talking to planet hunter Greg Laughlin (UC-Santa Cruz) about the latter’s equation that quantifies the worth of a given planet. It’s an ingenious concept, one that accepts inputs like planetary mass, estimated temperature, type and brightness of the primary star, and generates a value in cash. Why do this? It’s a way to measure the potential of an exoplanet to be interesting. or in Laughlin’s terms, “a way for me to be able to quantify how excited I should be about any particular planet.”

Reasons to Go

You can follow the genial and long-running musings on planetary value on Laughlin’s systemic blog, but read Billings if you’re not already familiar with the equation, because his interview with Laughlin walks you through all the parameters step by step. One of the interesting things about the equation is that the brighter the star appears to you, the more valuable the planet. That makes basic sense because we use photons to study exoplanets, and the equation says it is easier to do this if we have more photons to work with. Here’s Laughlin on the matter:

Think of it this way: If we’re sitting here on Earth, our sun is extraordinarily bright in the sky. The brightness of the Sun thus makes that term enormous if we run this equation for the Earth. If we evaluate this equation for the Earth, we get an answer of about 5 quadrillion dollars. And that’s basically the value of all our infrastructure, accumulated through history.

Being there, in other words, is everything, and we can watch the equation change with proximity:

If, for instance, there is a planet orbiting in the habitable zone of Alpha Centauri B, part of the closest star system in the sky other than our Sun, that planet’s worth about $6 billion using this scale. But then if you voyage there, Alpha Centauri B appears brighter, and brighter, and brighter, until it is your Sun in your sky and you’re on the planet’s surface. So in going there you have this ability to intrinsically increase value. And that’s an exciting thing because it ultimately provides a profit motive for perhaps going out and making a go of it with these planets.

This is saying that something that is several billion dollars on Earth, could be, if you go there, a quadrillion-dollar payoff.

Image: Exoplanet hunter Greg Laughlin. Credit: UC-Santa Cruz.

Costs of an Imaging Mission

You might object that using dollars and cents to put a value on an exoplanet is an arbitrary way to assign value, but there’s a cunning logic here. What drew me to write this was the fact that if you take the $6 billion that Laughlin’s equation pegs for a planet in the habitable zone of Centauri B and look at it in terms of how you might study it, the $6 billion starts to resonate. If we did find out in the next few years that such a planet existed, we would doubtless want to examine it through a space-based direct imaging telescope, and it’s Laughlin’s guess that the public would be willing to spend well more than a million — and certainly far less than a trillion — dollars to mount such a mission.

In other words, $6 billion might not be so far off if you’re thinking of building a cutting-edge space telescope to study an extremely attractive planet in Centauri B’s habitable zone. It was through a recent post on systemic that I learned about the most ‘valuable’ of exoplanet candidates so far, the object called KOI 326.01, to which Laughlin’s equation assigns a value of $223,099.93. KOI means ‘Kepler Object of Interest,’ a reminder that we haven’t yet even confirmed its existence. It’s a candidate, in other words, but an interesting one because it appears to be a bit smaller than Earth and orbiting in the habitable zone of its star.

Generalizing from Kepler

Laughlin’s valuation formula is great fun to play with — Mars, for example, merits a paltry $14,000, for reasons you can explore in the Billings interview. But let’s back out to the bigger picture discussed recently by Kepler head William Borucki at the American Association for the Advancement of Science’s annual meeting in Washington. Based on Kepler results and extending them throughout the galaxy, we come up with close to 50 billion planets as the roughest of estimates, with at least 500 million of them in the habitable zone of their stars.

All this is still evolving, of course, because we know that stars can have more than one planet, and the Kepler telescope needs more time to identify planets that are, like the Earth, at a greater distance from their star. And even when we have a more or less complete list of planetary candidates, we still have to confirm the findings via techniques like transit timing. It’s interesting to see, too, that a Kepler-like instrument studying our system would probably detect only one planet because our system is not co-planar enough to make multiple detections likely.

So let’s see. Kepler is a $600 million mission, and KOI 326.01 leads all candidates with a perceived value of $223,099.93. What new worlds will Kepler find that can get those two figures a little closer to each other? Surely an Earth-mass planet in the habitable zone of a G-class star would raise the ante, but we have more data collection and analysis ahead before we can make any such calls. Meanwhile, Borucki told the AAAS that 10.5 percent of the stars in the Kepler sample should have Earth-sized planets. Add in the 7.3 percent that should have super-Earths and it’s clear we should expect interesting places to explore both near and far in the galaxy.

Near, of course, is better, which is why we need to move from a statistical sampling mission like Kepler to an active examination of stars close to our Sun. Both the PLATO mission (PLAnetary Transits and Oscillations of Stars) and TESS (Transiting Exoplanet Survey Satellite) point us in that direction, and the hope here is that the exciting Kepler results will give both missions further impetus. Meanwhile, we continue to await with interest the results of ongoing investigations of the Alpha Centauri stars, in the hopes that a planet there will hit Greg Laughlin’s jackpot.


Comments on this entry are closed.

  • Erik Anderson February 22, 2011, 12:13

    “You might object that using dollars and cents to put a value on an exoplanet…”

    Ha! Not me! Did I not allude to something akin to this last week? I had dismissed as folly the nitwit suggestion to use NASA funds to build a dedicated ground-based telescope to harvest the highest possible quantities of high-mass planets hugging low-mass stars.

    Future exoplanet research programs should be about quality — not quantity. This will require putting more observatories in space. If cost is an issue, then our immediate priority is to develop more efficient launch vehicles/technology.

  • Erik Anderson February 22, 2011, 12:18

    “Based on Kepler results and extending them throughout the galaxy, we come up with close to 50 billion planets as the roughest of estimates, with at least 500 million of them in the habitable zone of their stars.”

    P.S. This estimate is loaded with non sequiturs. Sorry. :-(

  • Adam February 22, 2011, 17:30

    Hi Paul
    Does make one wonder just how much a planet has to be worth to justify a probe like Icarus? A hab-zone planet, with confirmed atmospheric signs of an oxic biosphere, around Alpha Centauri B – would that be enough?

  • Paul Gilster February 22, 2011, 18:53

    I guess Greg would say yes, assuming you could build Icarus for $6 billion… ;-) Although that interesting biosphere you’re talking about would set a higher value on the planet than the $6 billion the formula yields for a planet with no such traces. Seriously, though, it’s interesting to speculate on how a flyby probe would fit into the formula — you’re gathering more photons all the way, but you can’t stop at the target unless you and the rest of the Icarus team come up with a solution.

  • Michael Simmons February 22, 2011, 23:58

    “at least 500 million of them in the habitable zone of their stars.”

    My thoughts on reading this are:
    Say 1 in a million have the conditions for life and that life exists (think of all those venus, mars, gas giants, bad spin axis, no spin/ tidally locked, no magnetic field , no water and so on like planets) then 1 in 10 evolves complex life and then 1 in 10 of those have evolved sentient intelligent life since the big bang then that leaves us with 5 civilisations to find in 39 million million cubic light years of our Milky way…

  • Philip Horzempa February 23, 2011, 2:43

    I agree with Erik Anderson that NASA’s Exoplanet research should be about quality, not quantity. That is why I strongly oppose NASA’s cancellation of the Space Interferometry Mission, or SIM. The agency has already spent $600 million on design and engineering risk reduction over the past 10 years. The mission is ready to go.
    And what a mission it will be! SIM is an astrometry mission and is capable of finding Earth-mass planets around our NEAREST neighboring FGK stars.
    If you want to add your voice to those objecting to the cancellation of SIM, then visit the online Save SIM petition at –


    I just want to add that if SIM is put on hold, its funds will still be spent, only not on SIM. Putting the brakes on SIM will NOT lower NASA’s budget. It will merely delay the day when we find our nearest Earth clones.

  • Ronald February 23, 2011, 9:57

    Philip, great initiative, I signed it, however, the link was slightly faulty and I had to search for the right one.

    Michael, I think the first part of your statistics are too gloomy (1 in million chance of life on a habitable planet, that should be (much) higher), whereas the last part of your statistics may be too optimistic: I doubt whether 1 in 10 of all planets with complex life will develop intelligence advanced enough to call a civilization.

    All in all, as I recently argued in another post, I agree that the number of civilizations, and in particular those overlapping in time, in our MW galaxy must be very very small.

  • coolstar February 23, 2011, 11:54

    @Michael and Ronald I did a similar back-of-the-envelope calc after the latest Kepler results were released. I found them pretty gloomy for the prospects of ETI. One needs also, very importantly, to multiply by the ratio of your guess for the lifetime of a technological civilization by the period in which such a civilization could have existed in the MW to get the number of civilizations NOW. It’s all too easy to get numbers like 10^-5 or or 10^-6 for that part of the exercise. Even though it’s still early in terms of data releases from Kepler, the total number of earth-sized planets in habitable zones seems unlikely to increase by more than a factor of 2.

  • Erik Anderson February 23, 2011, 12:41

    The loss of SIM was a veritable tragedy — not merely its planet-hunting potential but also for Galactic astronomy in general. But let’s not kid ourselves. SIM ~is~ on hold. It was de-funded years ago. The mission is nowhere near “ready to go.” All that has been produced lately are blueprint revisions with repeatedly diminished specifications. Meanwhile, ESA has been forging ahead with assembling Gaia. We are now just one year away from Gaia’s launch. Look to Europe show the USA how its done.

  • Ronald February 23, 2011, 13:58

    Coolstar: yes, the percentage of stars with an earth-sized planet in the HZ now stands at about 0.5% but is eventually expected to roughly doubly to about 1% of all stars.

    Although the total number of earthlike habitable planets (near solar type stars) has been estimated at various occasions as anywhere between 50 and 200 million (see for instance https://centauri-dreams.org/?p=11625&cpage=1#comments, and more recent posts and comments here on this topic), with 250 – 500 million being either overly optimistic or taking all stars (i.e. not just the solar type), the number of (co-existing) civilizations must be very small indeed.

    I did my own back-of-the-envelope estimate and came to the conclusion that, if the average civilization lasts for 10,000 years, the chance that there are two of them existing in our MW galaxy at the same time in something on the order of 1%. My guesstimate may be overly optimistic or slightly pessimistic, but either way it is (much) more likely that at any given time (such as now) a civilization (such as us) is the only one present in the MW galaxy. Or maybe just a few others (5 – 10 orso) at the most, most probably scattered vey very widely through the galactic disc, and not all even being at the same advanced technological level.

  • Porterhouse February 23, 2011, 19:25

    Even if there is only, on average, one technological civilization per galaxy at any given time, it follows that there are billions of such civilizations in existence right now accross the visible universe. Wonders we shall never know, but wonders nonetheless.

  • Rob Henry February 23, 2011, 21:46

    Surely the prospects for the emergence of intelligent life given a world with high metabolism multicellular animals are very much greater than many portray here.

    It is often thought that only a minority of planets that support advanced life will develop intelligent life, yet the one case that we know of does not support this gloom. I would go so far as to state that if we had to deduce the main effect of evolution in advancing complex animals, we would surely have to state increasing brain complexity and encephalisation index was its major thrust – often seemingly its only thrust and changes in the immune system are the only rival that I can think of. Such a trend is seen over such a wide range of animal phyla that I wonder why so many feel that many other groups would not have got there in just a few hundred million more years if two (possibly more) species of hominids had not got there first.

  • Michael Simmons February 23, 2011, 23:01

    Ronald said “if the average civilization lasts for 10,000 years,”
    This seems very pessimistic.

    I think once a civilisation gets past a certain point its going to be around for a very long time.

    Our own risk of self annihilation, asteroid extinction and so on has been going down rapidly as our awareness has grown. Peak Oil/fossil fuels running out, a massive agricultural disaster or a pandemic might kill billions of us and/or put us back a few hundred years, but I don’t see us going extinct.

  • Ronald February 24, 2011, 7:10

    Rob: interesting viewpoint and definitely a valid one, however, although indeed increasing brain size can be seen in many different animal groups, I would still counter that virtually none of them led to the high level intelligence enabling what we could call self-awareness, let alone civilization.
    The only lineages that developed brain size and intelligence anywhere near ours are (of course) the primates and cetaceans, the latter not leading to any technology probably because of their watery environment (another obstacle for the development of technological civilization on water planets).

    For instance, the dinosaurs and birds, both highly successful lineages, have had ample time, diversification and opportunity to develop something like civilization and yet they have not.

    Moreover, when we consider not just a few large and conspicuous animal groups but the whole animal kingdom, it is striking that by far the greatest groups of them, particularly the arthropods and above all the insects within those, have thrived, adapted, survived and diversified, yes, even developed certain kinds and levels of ‘organized society’, without having to develop high intelligence or large brains.

    Finally, what life and evolution are really about is survival and adaptaton, which can apparently be achieved very successfully without high intelligence, as microorganisms have proven for billions of years.

    This is all not just telling, but from a viewpoint of evolution and the development of intelligenc it is outright worrisome. It looks as if large brains and high intelligence are very exceptional and expensive flukes of evolution, rather than an inevitable trend.

  • M.Hajnal February 24, 2011, 9:18

    Hey folks!
    Thanks for the post, it’s a very interesting equation. I just made a simple online calculator for it. Check it out at the following URL:


    Just enter the parameters of the planet, and hit ‘Calculate value’ to see the result. And send me a mail to hajnalm {at} gmail {dot} com if you have any suggestions, questions or nice things to say.

  • coolstar February 24, 2011, 12:16

    Ronald: I got essentially the same results as you for my calc of N (actually, for L=10,000 I think I may have gotten N=1). I think I’m more optimistic in some respects as I included all K and most M stars. The controlling factor really does seem to be L, as guesses for it seem to cover a much larger range than for the other parameters. L really has to be on order of a million or 10 million to give very many technological civilizations in the galaxy, given what we know (or think we know) now. I suspect that most of the earth-sized planets in habitable zones yet to be found by Kepler will actually be around early K stars as they’re much more numerous than G stars and the parameter space for their periods hasn’t been searched well yet (or, at least, yet released).
    I think your remarks about the evolution of intelligence on earth are spot on. One need only compare the time period where intelligence COULD have evolved to the time during which it HAS to see their reasonableness
    (which doesn’t mean, of course, that we’re right as I for one would very much wish to be wrong on this).
    I suspect the “big-shots” in the planet hunting field and SETI are purposely ignoring these arguments for the time being. Which is reasonable, for now, but in only a decade’s time, if the numbers don’t get better…..

  • Rob Henry February 24, 2011, 18:03

    Ronald, I understand your stated reservations over my assumption of a high fi in the Drake equation, but your main one is hard to quantify.

    I have also noticed that arthropods are so extremely successful, and that if I was right, a few should be well along the road to true intelligence by now, even if this process has been slowed by vertebrates already occupying most of the niches available for ‘clever animals’. As you noted, the most successful among them are the insects, but these particular arthropods have adopted a tracheal breathing system, that is extremely effective but limits their size. The result is that their most advanced forms are eusocial, and the individuals within some of these have strange mental attributes for their brain sizes, for example the ability of bees to communicate the whereabouts of food with each other, and possible even the ability to count. More importantly, these colonies have an intelligence above that of the individuals. It has been shown that much of this collective intelligent is just formed by the repeated following of simple rules, and thus should have little flexibility, but it certainly has not been shown that all of it is of this nature. In order to rate the success of the case of the arthropods against the high fi postulate we must therefore first have a theory that shows how and if true intelligence can be achieved by a collective of small brained animals. If we can prove such intelligence possible then we must measure how far along the path these insect have gone to date, and only then can we rate them a success or failure.

    I feel that the most telling factor is that the general trend for increase in brain size is against evolutionary pressures. Brains are expensive to maintain, and complex brains place additional physiological strains on their holders. The only payoff to this trend could be increased intelligence, thus it must be an exceedingly useful commodity in a wide range of circumstances.

  • Rob Henry February 24, 2011, 19:30

    Coolstar: I am sorry I must side with Ronald over a value for L of just 10,000 years being likely. You share a gut feeling with me that this is far to low, the reasoning that I would give is as follows:

    I could deign to agree that most technological civilisations will rapidly selfdestruct, but the nature of each would be so different from each other that it is difficult to conceive of less than 10% of them being immune from this problem. Given that most civilisations should have arisen a few billion years before us we can assign a lifetime of circa 1 billion year to these immune civilisations even if they die with their star. This would make L at least 100 million years.

    The problem is that all such arguments for high L are destroyed by the doomsday argument that runs as follows:

    We have first had the ability to attempt to messages the stars for about 100 years or so (I will pretend that this is an exact figure). We are not examining this position because of a privileged event, such as the recept of a reply. We thus have complete statistical validity in assuming that there is a 95% chance that we are thinking of this ability within the first 5% of its existence, and so can have 95% confidence that this technological situation will not last more than another 1900 years.

    Thus Coolstar, you can see that in order to remain reasonable and push L well beyond 10,000 years we must KNOW (not just assume) that humans are unusually selfdestructive for intelligent creatures.

  • spaceman February 24, 2011, 21:07


    I have several questions about the figures quoted in this thread.

    1). I thought the planet frequencies for earths, super-earths, Neptunes, and Jovians were 6%, 7%, 17%, and 4%, respectively. This gives an overall planet frequency (albeit one based only a portion of the expected Kepler data) of ~34%. In contrast, the planet frequencies for earths, super-earths, Neptunes, and Jovians mentioned by Burocki at the AAAS meeting were 10.5%, 7.3%, 20.8%, and 5.2%, respectively. This gives an overall planet frequency of ~44% which is approximately 10 percentage points off from the figure mentioned in the actual scientific paper released earlier this month. What accounts for this obvious difference? Again, this gets back to: how incomplete is the fp estimate based on the initial Kepler data? How much of a lower limit is the current estimate of planetary frequency?

    2). Burocki seems to be rounding 44% up to 50% of stars with Kepler-detectable planets in a galaxy of ~100 billion stars. Some parts of the galaxy such as the outer rim will not have as many planets as does our region inward to the galactic center. In any case, it was said that 1 in 200 stars have at least one planet in the habitable zone based on the existing data set, but 50 billion/200 = 250 million and not 500 million. Where is the 500 million figure coming from?

    Many of the approximately 250 million will not have conditions conducive to the origin of life as we know it. The reason being is that within this estimate are planets that are too large–think all of the Neptunes and sub-neptunes found. Many will not have enough water. Many will not have protective magnetic fields. I am not sure if I agree that a planet has to have all of the properties Rare Earth hypothesizers think it must in order to be a promising site for the origin of complex life, but enough water and a protective magnetic field to protect against cosmic rays seem to be musts.


  • Eniac February 25, 2011, 0:12

    All the “doomsday” scenarios above neglect the very important possibility that a civilization may advance sufficiently to colonize nearby stars. Once that happens, the value of L does not matter. What matters is that more colonies are formed than self-destruct, and the galaxy will be filled in a cosmic blink of the eye. Because of the the many billions of years available for this to happen, you have to put the probability of expansion VERY close to zero to explain the absence of ETI. This conflicts with the rapid and seemingly unstoppable increase in our technological abilities and is the essence of the Fermi paradox, as understood by Fermi.

  • Ronald February 25, 2011, 15:14

    Spaceman: I think the reason for the seeming discrepancy is that the quoted ‘1 in 200’ is not 1 in 200 planets (i.e. 1 in 400 stars) but 1 in 200 of *all* stars, so you would have to divide 100 billion by 200, giving 500 million indeed.
    Furthermore, my impression is that Burocki means that 1 in 200 stars has a more or less earthlike (or at least more or less earthsized) planet in its HZ.
    From the Kepler data it can be concluded that the total number of planets in the HZ of stars is considerably higher (about 10 times), but the combination of near-earthsized and being in the HZ reduces the fraction to about 0.5% so far, a figure which is expected to approx. double to about 1% (of all stars), when Kepler data are complete.

  • coolstar February 27, 2011, 19:47

    @Rob Henry & Eniac I don’t actually have much of an opinion about what a reasonable value of L should be. Until (if) we find some evidence of ETI I doubt we’ll even be able to make reasonable guesses for that number. That’s why I think it’s instructive to try lots of different values, differing by several orders of magntiude, and to explore the various implications. I do agree that that interstellar colonization is certainly likely to increase L for a given civilization. My suspicion is that perhaps only quite a small fraction of technological civilizations will do that, for a wide variety of possible reasons.
    As has been pointed out though, the colonization time for the entire galaxy is short compared to even the solar lifetime so we’re back to the Fermi Paradox once again. There are reasonable explanations as to where all the ETI artifacts are, even if we assume the galaxy has been colonized (perhaps more than once).
    If L can be very large, and a species does have interstellar travel, it may be quite simplistic to think of the resulting civilization (or species) as being monolithic, as speciation times on the earth (without directed evolution) seem to be short compared to the time to colonize the MW.
    Still, I find the Kepler results troubling, as I don’t think many had suspected, before we had decent data, that terrestrial sized planets in habitable zones would be so rare. A reasonable approximation is that they’re 50-100 times rarer than most people in the field thought even 5-10 years ago. Data has that troubling quality of changing things!

  • Eniac February 28, 2011, 1:02

    If L can be very large, and a species does have interstellar travel, it may be quite simplistic to think of the resulting civilization (or species) as being monolithic, as speciation times on the earth (without directed evolution) seem to be short compared to the time to colonize the MW.

    Indeed. Rather than monolithic, the galactic population would be a very loosely connected mass of independent colonies. So independent, in fact, that L for the ensemble would certainly be infinite, because there could never be an event that would affect all at once.

    Think of Earth. Life took hold, spread across the globe, and never ever remotely came close to disappearing again for billions of years. And this is in a comparatively small space, where events affecting all at once do exist. Once we (or anyone) head for the stars, the same is certain to happen to the Milky Way. It is a direct consequence of self-replication.

    And just as life is evident on Earth, it would be evident in the galaxy, if it existed.

  • Ronald March 6, 2011, 11:14

    coolstar: “Still, I find the Kepler results troubling, as I don’t think many had suspected, before we had decent data, that terrestrial sized planets in habitable zones would be so rare. A reasonable approximation is that they’re 50-100 times rarer than most people in the field thought even 5-10 years ago.”
    With ref. to my own post just before yours, that seems a bit too pessimistic: it now seems that about 0.5 – 1 % of all stars have a terrestrial planet in their HZ. That is indeed not 5% (but 5 – 10 times lower) than some optimists suggested, but not 50 – 100 times lower.

    And this is for all stars surveyed. It is possible that the statistics are a bit more favorable for the solar type stars, because their very characteristics (mass, metallicity, luminosity, etc.) may be more conducive to the desired combination of a terrestrial planet in the HZ.

    Eniac: “And just as life is evident on Earth, it would be evident in the galaxy, if it existed.”

    As I have also stated in the recent discussion thread on SETI, this is mainly true for *intelligent* life. The MW galaxy and the universe may be teeming with all sorts of (primitive) life with complex life being much rarer and intelligent life exceedingly rare.
    As our technologies advance, particularly telescopic and spectro-analytical technologies, life in the galaxy will indeed become increasingly evident to us. This is dependent on available detection technologies, not on life itself.

  • ljk March 10, 2011, 6:25

    The Occurrence Rate of Earth Analog Planets Orbiting Sunlike Stars

    Authors: Joseph Catanzarite, Michael Shao (Jet Propulsion Laboratory, California Institute of Technology)

    (Submitted on 8 Mar 2011)

    Abstract: Kepler is a space telescope that searches Sun-like stars for planets. Its major goal is to determine {\eta}_Earth, the fraction of Sunlike stars that have planets like Earth. When a planet ‘transits’ or moves in front of a star, Kepler can measure the concomitant dimming of the starlight.

    From analysis of the first four months of those measurements for over 150,000 stars, Kepler’s science team has determined sizes, surface temperatures, orbit sizes and periods for over a thousand new planet candidates.

    Here, we show that 1.4% to 2.7% of stars like the Sun are expected to have Earth analog planets, based on the Kepler data release of Feb 2011. The estimate will improve when it is based on the full 3.5 to 6 year Kepler data set.

    Accurate knowledge of {\eta}_Earth is necessary to plan future missions that will image and take spectra of Earthlike planets.

    Our result that Earths are relatively scarce means that a substantial effort will be needed to identify suitable target stars prior to these future missions.

    Comments: 13 pages, 5 figures

    Subjects: Earth and Planetary Astrophysics (astro-ph.EP); Solar and Stellar Astrophysics (astro-ph.SR)

    Cite as: arXiv:1103.1443v1 [astro-ph.EP]

    Submission history

    From: Joseph Catanzarite [view email]

    [v1] Tue, 8 Mar 2011 06:11:58 GMT (636kb)


  • Ronald March 10, 2011, 19:58

    Ref. ljk: very interesting and relevant publication!

    I checked it to find that earthlike planets are defined as having a radius between 0.8 and 2 * earth, and also as being in the star’s Habitabel Zone (as defined by Kasting et al., 1993: between 0.95 and 1.37 Au in our solar system).
    I am not sure how they defined ‘sunlike’ stars, but my impression from the summary is that it is defined as all FGK stars.

    This also explains their % of 1.4 to 2.7, compared with o.5 – 1% derived previously from preliminary Kepler results.
    0.8 earth radius (corresponding to about 0.5 earth mass) may be a bit pessimistic as a lower limit, but on the other hand 2 earth radii (roughly corresponding to 6-8 earth masses) seems rather high, I would call that super-earth.
    And to take *all* FGK stars is definitely overly optimistic as a definition of sunlike stars.
    So as for now I stick with my previous guesstimate of about 1% of all sunlike stars having an earthlike planet in the HZ.