I was supposed to get the stuff put on my teeth today, but it got postponed; since I'm leaving town I probably won't get it applied for another month or two.
It takes about a year for it to become effective, so this market is really "Will I get any new cavities from 1 year after application (latest December 2023, so final cavity check Dec 2024), until the end of 2026?"
I'll do a dental visit around 1 year from application to check for baseline cavities, and won't count any that exist.
I am historically very cavity prone, I've had I think 14 cavities? I have good but not great oral hygeine; I brush 1-2 times a day and go through phases of regular flossing and not flossing. I last visited the dentist 6 months ago and they said "everything looks pretty good."
Related questions
@Aella I'm also happy with you restarting the clock. Lumina's effectiveness is a really interesting open question, and it'd be a bummer to have one of the big prediction markets about it fail.
@Aella Wasn't the baseline dental check going to be one year after application?
I'll do a dental visit around 1 year from application to check for baseline cavities
Plus
It takes about a year for it to become effective, so this market is really "Will I get any new cavities from 1 year after application (latest December 2023, so final cavity check Dec 2024), until the end of 2026?"
Traders may be interested in a podcast I recorded with the founder: https://www.youtube.com/watch?v=tPyZt3EJCFM
Assuming the rate of cavities hasn't changed much and there aren't other major interventions, the prior should be ~60% (if you model as a Poisson process with on avg slightly fewer than 1 cavities per 2 yrs)
So at 15%, the market seems to imply that ~3 out of 4 new biotech startups claiming to solve a problem on the level of eradicating cavities actually succeeds or makes very serious progress towards their claim.
I don't know much about biotech startups, but this seems far too high to me. If that were the real base rate for new biotech startups, I have to imagine we'd have much fewer health problems as a society by now.
@georgeyw I mean, one might claim to know more about Lantern Bioworks beyond just that they are a biotech startup claiming to solve a problem on the level of eradicating cavities
I don't actually myself know anything that would especially suggest they're unusually reliable, though—or at least, not other than the fact that Manifold seems to maybe think so
@April yeah that's fair, though I think by default I'd still be suspicious of whatever evidence they make available, since startups are incentivized to optimize that to look impressive to investors. For example, they claim that this cure was created ~40 yrs ago and that it's a solution worth ~$45B/yr. That's a lot of money, and if it really is the harmless silver bullet that Lantern Bioworks claims it is, then I'm extremely confused about what caused it to be forgotten about (and if there's a coherent story to that, then are there just tons of golden gooses laying around that no one's picked up??). I also didn't notice a citation on their site, and if there was bulletproof evidence of their solution, I'd expect it to be easier to find.
Now that I'm thinking about it, I have other questions too, like why does it take a year to become effective if part of the mechanism is outcompeting all the other bacteria in your mouth? That seems like a very long time relative to normal bacterial timescales.
Without looking up actual numbers, my guess would be that 3 out of 4 is at least a couple orders of magnitude higher than the actual base rate. I don't think that's insurmountable from an inside view, but I think I'd be surprised if it were on average a losing strategy to bet based on outside view when there's that much cushion. I'm new to this though, so I appreciate/welcome any corrections.
@georgeyw My bet is not premised on the outside view but on the specifics of their claims - I think there's reasonably good evidence to believe they used fairly well-established technology to solve the mechanics of a fairly well-understood problem. It's certainly possible it doesn't work for unforeseeable reasons, but I'm giving them good odds.