Mostly fair, but tiers did have a slight other impact in that they were used to bias the final room: Clay Golem and Hag were equally more-likely to be in the final room, both less so than Dragon and Steel Golem but more so than Orcs and Boulder Trap.

Yes, that's a sneaky part of the scenario.  In general, I think this is a realistic thing to occur: 'other intelligent people optimizing around this data' is one of the things that causes the most complicated things to happen in real-world data as well.

Christian Z R had a very good comment on this, where they mentioned looking at the subset of dungeons where Rooms 2 and 4 had the same encounter, or where Rooms 6 and 8 had the same encounter, to factor out the impact of intelligence and guarantee 'they will encounter this specific thing'.

(Edited to add: actually, there are ~100 rows in the dataset where Room2=4, Room6=8, and Room3=5=7.  This isn't enough to get firm analysis on, but it could have served as a very strong sanity-check opportunity where you can look at a few dungeons where you know exactly what the route is.)

I think puzzling out the premise could have been a lot more fun if we hadn't known the entry and exit squares going in

I think this would have messed up the difficulty curve a bit: telling players 'here is the entrance and exit' is part of what lets 'stick a tough encounter at the entrance/exit' be a simple strategy.

The writing was as fun and funny as usual - if not more so! - but seemed less . . . pointed?/ambitious?/thematically-coherent? than I've come to expect.

This is absolutely true though I'm surprised it's obvious: my originally-planned scenario didn't quite work out as intended (I'm still trying to assemble mechanics for it that actually work the way I want them to) and this was my backup scenario.

imo a 4x4 or 5x5 dungeon would probably have been easier than the 3x3, especially for reliably distinguishing between hypotheses A and B

Interesting.  I trimmed it down to 3x3 as part of Plan 'Try Not To Make Everything Too Overcomplicated', trying to use the smallest dungeon that would still make pathing relevant in order to avoid dropping 16 separate encounters on players.

There was one aspect about which I have unreservedly positive feelings: the chrono effects, the hag poem and the varying numbers of adventurers were all excellent red herrings, seeming like they might hint towards subtle opportunities for performance improvement (and/or a secret Bonus Bonus Objective) but being quickly dismissable as fingertraps. 

This...is not really quite how those were intended.  The intent was something more along the lines of 'Easter Eggs'.

Here’s a third paper, showing that sports betting increases domestic violence. When the home team suffers an upset loss while sports betting is legal, domestic violence that day goes up by 9% for the day, with lingering effects. It is estimated 10 million Americans are victims of domestic violence each year. 

I was suspicious of the methodology here (e.g. the difference between 'when the home team loses violence goes up by 9% if and only if gambling is legalized' and 'when the home team loses violence goes up by 10% if gambling is not legalized but by 10.9% if it is legalized' is something that I don't trust sociology to track honestly).

I went to take a look at the paper, and do not think it really supports the argument at all.

The relevant charts I believe are on p26 here.  The first one shows how intimate partner violence (IPV) varies with 'expected outcome of game' and 'actual outcome of game':

Note that 'expected outcome of game' is the thing that actually seems predictive, not 'actual outcome of game'.  When the home team is expected to lose, domestic violence is high even if they win.  When the home team is expected to win, domestic violence is low even if they lose (though even lower if they win).

This looks to me like a study that's been massively confounded by other effects.  Perhaps good sports teams tend to be favored to win, and also to be in wealthy regions with little domestic violence?  Regardless of the reason, though, this makes me very suspicious of anything this study claims to show.

The second chart shows how IPV varies with the outcome of the game based on whether sports betting is legal:

This does, indeed, show that areas with legalized sports betting had higher rates of domestic violence when the home team lost (~0.45 vs ~0.43).  However, it also showed that they had lower rates of domestic violence when the home team won, by more. (~0.38 vs ~0.42).  If we assume that half of games are wins and half are losses (seems...pretty reasonable?), I believe this chart depicts legalized sports betting lowering domestic violence (though again I don't know if I believe that either due to how obviously confounded this data is).

Somehow we seem to have gone from "a clearly confounded paper that (if you believe it) shows sports betting on average lowering domestic violence" to "there is strong evidence of sports betting increasing domestic violence".

I find this somewhat depressing.

The dungeon is laid out as depicted; Room 3 does not border Room 4, and does border Room 6.  You don't, however, know what exactly the adventurers are going to do in your dungeon, or which encounters they are going to do in which order.  Perhaps you could figure that out from the dataset.

(I've edited the doc to make this clearer).

I think you may have mixed up the ordering halfway through the example: in the first and third tables 'Emma and you' is $90 while 'Emma and Liam'is $30, but in the second it's the other way around, and some of the charts seem odd as a result?

I don't think you should feel bad about that!  This scenario was pretty complicated and difficult, and even if you didn't solve it I think "tried to solve it but didn't quite manage it" is more impressive than "didn't try at all"!

1. There is a problem I want solved.

2. No-one, anywhere in the world, has solved it for me.

3. Therefore, Silicon Valley specifically is bad.

Were whichever markets you're looking at open at this time? Most stuff doesn't trade that much out of hours.

I think this is just an unavoidable consequence of the bonus objective being outside-the-box in some sense: any remotely-real world is much more complicated than the dataset can ever be.

If you were making this decision at a D&D table, you might want to ask the GM:

• How easy is it to identify magic items?  Can you tell what items your opponent uses while fighting him?  Can you tell what items the contestants use while spectating a fight?
• Can we disguise magic items?  If we paint the totally powerful Boots of Speed lime green, will they still be recognizable?
• How exactly did we get these +4 Boots?  Did we (or can we convincingly claim to have) take them from people who stole them, rather than stealing them ourselves?
• How honorable is House Cadagal's reputation?  If we give the Boots back, will they be grateful enough that it's worth it rather than keeping the Boots?

I can't realistically explain all of these up front in the scenario!  And this is just the questions I can think of - in my last scenario (linked comment contains spoilers for that if you haven't played it yet) the players came up with a zany scheme I hadn't considered myself.

Overall, I think if you realized that the +4 Boots in your inventory came from the Elf Ninja you can count yourself as having accomplished the Bonus Objective regardless of what you decided to do with them.  (You can imagine that you discussed the matter with the GM and your companions, asked all the questions above, and made a sensible decision based on the answers).