I hate the greeting.
Sometimes, an honest answer feels like a social gaffe.
Depending on your situation, more like always an honest answer is in fact a social gaffe.
GuySrinivasan
Karma: 1,998
In software development, I take joy from (honestly) reporting credible intervals that are far too large for anyone’s comfort.
“Does it help?” you ask.
Well, joy is a good thing.
In our house we heuristic over this problem by, and I quote, “accounting for the planning fallacy twice”. This works very well for us in practice.
Do you have a tldr on why we might think anti-vaxxers were right for the right reasons? Seems like the default positions are “vaccines have obviously worked in the past and we’re pretty sure they’re gonna work in very similar ways today”, and I haven’t seen anything that changes my opinion much about either of those defaults.
Mountaineer’s Fallacy Fallacy—working on the real problem ineffectually when the right move is working on adjacent problems that might be pointless in order to chip away at the edges until the key components of solving the real problem are actually visible and tractable.
But yeah the OG MF is common.
I love this and intend to use it to flex at work until and unless HR tells me to stop because something something gambling something.
The signup was slightly confusing. Entering a user name and password, the but about adding an email said “but first, log in!” and there were two buttons, a “log in” and a “sign up” button. Clicking “log in” said I couldn’t because I needed to sign up. So I clicked “sign up” and then the hiccup was over.
This was a fun one! Post-mortem:
Biggest miss is that I failed to guess that the distributions were identical up to a constant person/resonance multiplier. Actually I did guess that initially, but decided it probably wasn’t true, which is why it’s my biggest miss. I started thinking it when Maria hit more x1s than x0s on Gamma while Janelle hit more x0s than x1s, leading me to think that there were more person-dependent factors than just one. IIRC the nail in the coffin of that theory was looking at Epsilon resonance. It was clear that Epsilon was non-random and inspecting the overlapping areas of the curve showed (e.g.)
A|Janelle|Maria|multiplier
-----
0.28|0.17|0.21|0.86
1.28|0.20|0.26|0.77Later, I lowered my confidence in Will’s Epsilon prediction because I knew our instruments have limited precision. I didn’t connect that with “maybe 0.77 vs 0.86 isn’t that far off given imprecision!”.
Also I think I was modeling the precision incorrectly, probably. I took “for example, since they say Earwax has an amplitude of 3.2 kCept, you can be 100% sure the true value is between 3.15 and 3.25 kCept” to mean that every value could be plus or minus 0.05, but I think now it actually meant that values were rounded to the nearest digit shown, so a listed value of 0.28 kCept was not between 0.23 and 0.33, but rather between 0.275 and 0.285?
Biggest hit, I think, was correctly determining Janelle’s actual chances: I said 25% win, 39% double; actual was 37% win, 30% double. Method was seeing graphs that were clearly 5 linear trends by power, estimating the zero, estimating their slopes, and noticing the multiplier.
I completely forgot that since Earwax’s actions are unprecedented, we’re not entirely confident of its amplitude remaining constant either! Janelle’s Gamma has basically the same characteristics at various amplitudes. Will’s Epsilon works significantly less often if the new amplitude becomes smaller. A bit more reason to stick with Janelle here.
I think the only big remaining two things that could convince me to switch to Will are (a) figuring out a time-based/less-significant-digits-based pattern which tells us that Janelle’s Gamma will have a poor k today/at 3.2 amplitude, or (b) figuring out a simple theory giving the cubic (or whichever) for Maria and Janelle that predicts Will’s Epsilon will always win, removing the model uncertainty and the precision uncertainty.
Using Python in a Jupyter notebook. Seaborn has a fantastic little function to quickly see pairwise graphs, it’s great to begin with. Here’s what Maria looks like after sns.pairplot(df[df.name==‘Maria N.’]): https://drive.google.com/file/d/12Q_11ZTPnyak87EXO89Vbg4am3TXc4px/view?usp=sharing
Summary: Send Janelle, using Gamma Resonance. Great chance of winning, maybe 60-70%, and half the time you win you double also. Honorable mention to Will’s Epsilon Resonance, which if we had more data or a better theory we might be convinced could win 100% of the time, but we just don’t have the data or theory to justify it yet.
Alpha: Maria does not show any real amplitude-dependent EFS, and no potential pilot shows 3.2+ EFS. Reject.
Beta: No amplitude-dependence. Maria and Janelle look pretty similar, but all the trainees do far less well, so this is person-dependent. As such probably Janelle’s chances of winning are probably best estimated using only her data? Either way, I get somewhere between a 2%-3.5% chance of winning for Janelle and nothing for trainees. We can do better.
Gamma: Very clear dependence on amplitude A, which is good. Each person has some base value B; Maria’s is 0.66, Janelle’s is 0.89. The EFS generated is of the form B x (1 + k x A), for observed values of k from 0 to 4, and maybe one day we’ll see higher. I don’t see a way to predict k, but this is quite promising for Janelle. k=0 loses, k=1 wins, and k=2, 3, or 4 wins and doubles. That’s a 64% chance of winning overall, with 39% (60% of the time given that we win) of doubling. Going off of observed frequency of Janelle hitting various k; the relative ratios are different enough from Maria’s that it’s not clear we can combine them in any nice way. They both show large k=0,1,2 and small k=3 and tiny k=4. (None of the trainees has a higher B.)
Delta: Maria’s EFS shows a linear upward trend dependent on amplitude. Janelle’s is too low to be interesting, and the trainees’ are all very low and none suggest having a super-positive slope. Reject.
Epsilon: Ooookay this one’s interesting. I didn’t get sin to fit as well as a cubic, and I did get a cubic form that can be described with just one parameter varying per Maria vs Janelle, which I thought was likely given how Gamma worked, which means we can generate the entire cubic for Will and check how it does at A=3.2 and… it predicts 3.31 EFS. The Epsilon Resonance is entirely predictable given an amplitude. However, there are two big problems with simply sending Will to use Epsilon. First, even if our model is precisely right, the precision of our instruments is not perfect, and once we take that into account, Will’s Epsilon EFS predictions vary a fair amount, leading to only about a 60% chance of winning. Second, our model is almost certainly wrong, because we haven’t found a simple model. So Janelle’s Gamma is better than Will’s Epsilon in every way according to our current uncertainty.
Zeta: You might get 0, or your base Z, or rarely 3.5 x Z. The problem is that Janelle’s 3.5 x Z only just barely beats 3.2 (and doesn’t beat 3.25!), and she gets 3.5 x Z maybe like 5% of the time which is << 64%. While Corazon’s 3.5 x Z would handily double, it loses otherwise. Again << 64%.
Eta: Janelle’s too small here, but Flint has a 2.3, and it looks like the possible EFS’s are a base E, or x1.5, or x1.5x1.5, or x1.5x1.5x1.125, or 1.5x1.5x1.125x1.25. If Flint’s E=2.3, this might be promising. Unfortunately it seems much more likely that 2.3 is one of the multiples, and if there is an amplitude dependence, it’s probably one of the high multiples.
I am liking this one a lot. There are enough hints, some obvious and others more subtle, that indicate many resonance strengths are simpler than they appear at first. TBD: whether I’m reading too much into the data and seeing more hints than actually exist.
Oh! Branch-Loop Analysis is much different than I expected. Good to know.
Did we not record which resonance each pilot actually used each time? Usually it’s clear
and holistically it’s even more clear, but
it’d be nice to have confirmation in those cases where we have all the counterfactual EFS’s and I’m pretty sure that data should be available.
Feedback:
“please don’t shitpost and when you engage with me please avoid all attempts at humor because these pattern-match to ways I am abused and if you do those things even if in good faith it will only hurt our communication, perhaps disastrously, never help” would, I think, cover basically everything you want to cover without also signaling that it will be extremely emotionally draining to engage with you.
OTOH if it will be extremely emotionally draining to engage with you then you have successfully signaled that.
Possibly this isn’t fair but I’m pretty sure it’s an accurate reading.
I also live with an immunocompromised individual who cannot be successfully vaccinated. After research including reasoning very similar to yours, we concluded that if she wore a mask, we felt safe enough with vaccinated folks who had not had any obvious infection opportunities within the past 12 hours to be indoors for substantial periods of time with known-vaccinated folks not wearing a mask. This tracks almost exactly with your guess of “MIGHT be roughly 1.5 days[...] Maybe hours?”
I know about 16 vaccinated people I who I expect would have noticed+told me about easily noticeable >3 days side effects if they had occurred. 0⁄16 told me. Most mid-30s, two 60s. 40% W, 60% M. 80% basically healthy, 20% with significant health issues. Several groggy/sore/etc for a day, three very much so. Not sure about type distribution, though predominantly Pfizer.
The single most important thing I got from PJ Eby was the “what’s good about that?” question.
The Jewel Beetle was weird. It was what, like 8% to auto-win everything by winning the Beetle? Except there was just one roll, overall. So in each group of four, one person auto-wins, and then it becomes a cross-group auction where whoever got the Beetle for way less ends up winning. Seems like with very few people participating overall, going for the Beetle caps your odds of winning at 8%, which is not great. With very many people participating, like 100, going for the Beetle caps your odds of winning at the chance there is not a cohort with pure non-beetlers, otherwise whichever of them wins the beetle probably just wins.
Converges in the limit, we’re all good here.