Monday, November 5, 2018 Game Recap and Discussion (SPOILERS)

This is where all of the games are discussed.

Moderators: alietr, dhkendall, trainman, econgator

clprez
Watches Jeopardy! Way Too Much
Posts: 752
Joined: Wed Feb 26, 2014 9:01 pm

Re: Monday, November 5, 2018 Game Recap and Discussion (SPOILERS)

Post by clprez » Mon Nov 05, 2018 10:46 pm

Paddington is a train/tube station in London and thought that perhaps the bear was named after it, so I was pretty sure of my FJ guess.

User avatar
Saturnalia
Loyal Jeopardista
Posts: 200
Joined: Tue Feb 06, 2018 10:16 pm

Re: Monday, November 5, 2018 Game Recap and Discussion (SPOILERS)

Post by Saturnalia » Mon Nov 05, 2018 10:49 pm

Volante wrote:
Mon Nov 05, 2018 10:42 pm

I didn't realize Mortal Kombat changed studios (and, related, that Midway had gone under). Otherwise the 'K' TOM should've been a massive gimme.
Right, I didn't recognize the name of the studio and I think that made me hesitant, because, while I couldn't name offhand the studio that I associated with the game, I knew it was an iconic one.

User avatar
opusthepenguin
The Best Darn Penguin on the Whole JBoard
Posts: 6070
Joined: Thu Aug 11, 2011 2:33 pm
Location: Shawnee, KS
Contact:

Re: Monday, November 5, 2018 Game Recap and Discussion (SPOILERS)

Post by opusthepenguin » Mon Nov 05, 2018 11:03 pm

jeff6286 wrote:
Mon Nov 05, 2018 10:11 pm
opusthepenguin wrote:
Mon Nov 05, 2018 9:24 pm
Congrats to Mary Ann. We liked her a lot. But that wager was unfortunate and I'm glad it didn't bite her. She would have lost on a double-stumper when there was no need to do so--especially since Jonathan overwagered by $200.
Shouldn't she assume Jonathan knows how to bet from second and therefore she's going to lose on a double stumper if she bets to cover?
History does not generally vindicate that hypothesis and definitely did not do so in this instance. In this unusual circumstance, I think she is safest making the minimum bet that still guarantees her a TIE if Jonathan goes all in. (In most cases I'd recommend the minimum shutout bet to WIN, but the numbers were right on the edge tonight.) Had she followed this advice, she would have won less money in tonight's scenario. But my recommendation wins in more scenarios (including tonight's) than her choice did. Had she missed tonight's FJ she would have had occasion to deeply regret her wager.

Yes it's true that she might feel confident in the category. But we all know FJ can throw some awful curves in categories where we feel confident. Is it worth throwing away her best chance to win on a miss in order to maximize her win on a get? I would say no. But that's the way she wagered and it worked out and I'm happy for her. I'll be cheering her on tomorrow. Unless she's up against a boardie or something.

Golf
Watches Jeopardy! Way Too Much
Posts: 1506
Joined: Thu Jan 10, 2013 12:27 pm

Re: Monday, November 5, 2018 Game Recap and Discussion (SPOILERS)

Post by Golf » Mon Nov 05, 2018 11:19 pm

jeff6286 wrote:
Mon Nov 05, 2018 10:11 pm
Shouldn't she assume Jonathan knows how to bet from second and therefore she's going to lose on a double stumper if she bets to cover? In that case she might as well go big or go home. Children's lit also seems like a category that someone could easily feel very strong in, so if she felt she was highly likely to get it right I don't hate her bet at all, take that money while you have the chance.

Actually one more thought, if Jonathan doesn't bet small enough to win on a double stumper, he's quite likely to bet it all, so by leaving $200 behind she does retain a pretty decent chance to win given how often we see second place leave nothing behind. Just because he picked the big round number figure instead of betting either strategically or with typical recklessness doesn't make her bet awful.
You've been here long enough, contestants just don't wager rationally that often. And you're saying that a contestant should assume another contestant should wager rationally and adjust?

Her wager reduced her winning chances significantly, so did Jonathan's. Both wagers were far far far from optimal.

User avatar
jeff6286
Watches Jeopardy! Way Too Much
Posts: 4306
Joined: Fri Jul 01, 2011 7:34 pm
Location: Indianapolis, IN

Re: Monday, November 5, 2018 Game Recap and Discussion (SPOILERS)

Post by jeff6286 » Tue Nov 06, 2018 12:12 am

Golf wrote:
Mon Nov 05, 2018 11:19 pm
jeff6286 wrote:
Mon Nov 05, 2018 10:11 pm
Shouldn't she assume Jonathan knows how to bet from second and therefore she's going to lose on a double stumper if she bets to cover? In that case she might as well go big or go home. Children's lit also seems like a category that someone could easily feel very strong in, so if she felt she was highly likely to get it right I don't hate her bet at all, take that money while you have the chance.

Actually one more thought, if Jonathan doesn't bet small enough to win on a double stumper, he's quite likely to bet it all, so by leaving $200 behind she does retain a pretty decent chance to win given how often we see second place leave nothing behind. Just because he picked the big round number figure instead of betting either strategically or with typical recklessness doesn't make her bet awful.
You've been here long enough, contestants just don't wager rationally that often. And you're saying that a contestant should assume another contestant should wager rationally and adjust?

Her wager reduced her winning chances significantly, so did Jonathan's. Both wagers were far far far from optimal.
I wasn't making any defense of Jonathan's wager, it was certainly very bad, and only by $200, his big round number just happened to be very close to what would have been a logical choice.

I was kidding when I said she should assume he wagers strategically from second. I do think past history shows that the odds of the random second place contestant betting either less than 4800 or over 9200 from that position are pretty high. He just happened to pick 5000 which would have made Mary's MSB bet drastically better than than her big bet, but for many, many other numbers that he could have gone with, her bet (of anywhere betwen 9601 and 19199) ends up irrelevant.

User avatar
JayK33
Just a Fan
Posts: 499
Joined: Tue Jul 15, 2014 11:10 pm
Location: Canada

Re: Monday, November 5, 2018 Game Recap and Discussion (SPOILERS)

Post by JayK33 » Tue Nov 06, 2018 12:21 am

clprez wrote:
Mon Nov 05, 2018 10:46 pm
Paddington is a train/tube station in London and thought that perhaps the bear was named after it, so I was pretty sure of my FJ guess.
I didn't make the train station connection but luckily I was still able to come up with the correct response.

3/5 on the Lady Gaga category, which has been posted to YouTube.

Anachronism
Watches Jeopardy! Way Too Much
Posts: 282
Joined: Thu Jan 10, 2013 4:45 am

Re: Monday, November 5, 2018 Game Recap and Discussion (SPOILERS)

Post by Anachronism » Tue Nov 06, 2018 4:38 am

Yet how do you feel if you sit there having gotten FJ right from first place and yet you're going home? An optimal bet is only optimal if you know how others are going to bet. In this case, I don't see anything wrong with either bet.

I'm really hating the in-game commercials. But the Grande clue was a good laugh - like they don't know that a Hollywood engagement has a half-life far shorter than a taping delay?

User avatar
alietr
Site Admin
Posts: 5951
Joined: Thu Jun 30, 2011 1:20 pm
Location: Bethesda, MD

Re: Monday, November 5, 2018 Game Recap and Discussion (SPOILERS)

Post by alietr » Tue Nov 06, 2018 8:17 am

Elijah Baley wrote:
Mon Nov 05, 2018 3:19 pm
Kudos (?) to Gerald for going all in but wow, that was a supremely bad guess - like, how do you not even guess someone from the correct country?? And, after just getting what was a much harder clue. I'll take "Things That Make You Scratch Your Head," Alex!
I know a lot of people here like the all-in wager, but I'm not sure a bottom-of-the-board clue in DJ is the best place to go all in. It happened to have been a pretty easy clue, but he did miss it.

Paddington came to mind first, but I did take a detour to consider Thomas the Tank Engine before going back to Paddington as being more logical.

Golf
Watches Jeopardy! Way Too Much
Posts: 1506
Joined: Thu Jan 10, 2013 12:27 pm

Re: Monday, November 5, 2018 Game Recap and Discussion (SPOILERS)

Post by Golf » Tue Nov 06, 2018 8:32 am

Anachronism wrote:
Tue Nov 06, 2018 4:38 am
Yet how do you feel if you sit there having gotten FJ right from first place and yet you're going home? An optimal bet is only optimal if you know how others are going to bet. In this case, I don't see anything wrong with either bet.
No. Just no. Please be sarcasm I'm not seeing the sign for.

Nobody is suggesting the leader ever not wager to cover a double up. And yes, there are optimal wagers in Jeopardy. That's what game theory is all about.
alietr wrote:
Tue Nov 06, 2018 8:17 am
I know a lot of people here like the all-in wager, but I'm not sure a bottom-of-the-board clue in DJ is the best place to go all in. It happened to have been a pretty easy clue, but he did miss it.
I think it could go either way. Yes, it's a bottom row clue, but if you don't go big, you've got to hope for misses from both contestants. I'm OK with the TDD, I would have also been OK with $5 and hope for a FJ TS. Just off the top of my head I'm thinking if you have a better chance of correctly answering the clue than there is of both leaders missing FJ, you go for it.

John Boy
Watches Jeopardy! Way Too Much
Posts: 2191
Joined: Fri Jul 15, 2011 7:11 am

Re: Monday, November 5, 2018 Game Recap and Discussion (SPOILERS)

Post by John Boy » Tue Nov 06, 2018 10:08 am

Euphonium wrote:
Mon Nov 05, 2018 3:45 pm

Holy crap Mary Ann is good. Maybe it was just me but she seemed to be just a bit behind on the buzzer in J, but came on strong in DJ when everyone else was slowing down coming up with responses for the harder clues.
My impression exactly. She looked OK early on, but once she got up a head of steam she was awesome, and seemed to know a lot of really difficult material. Kudos to her. I'm not crazy about her FJ wager, but I'm imagining she thought it was a strong category, and her sole get proved her right. We'll need to see how her wagering skills look in her subsequent game(s).

And all that after not knowing the difference between a flush and a full house? Well, we all have our strong subjects and weak subjects.

One of my weaker ones is children's lit., so this FJ category made me groan. I made the same mistake as Jonathan, going with Harry Potter. At about 20 seconds Paddington popped into mind, and I don't know why since I've never read any Paddington books. Thought about changing but felt zero confidence in doing so. poop.

Got the last DD (troglodytes) primarily from the Star Trek episode of several hundred years ago.

John Boy
Watches Jeopardy! Way Too Much
Posts: 2191
Joined: Fri Jul 15, 2011 7:11 am

Re: Monday, November 5, 2018 Game Recap and Discussion (SPOILERS)

Post by John Boy » Tue Nov 06, 2018 10:12 am

Fun Fact: Cleveland does not have a sister city, but we do have a suicide pact with Detroit.

User avatar
opusthepenguin
The Best Darn Penguin on the Whole JBoard
Posts: 6070
Joined: Thu Aug 11, 2011 2:33 pm
Location: Shawnee, KS
Contact:

Re: Monday, November 5, 2018 Game Recap and Discussion (SPOILERS)

Post by opusthepenguin » Tue Nov 06, 2018 10:53 am

Anachronism wrote:
Tue Nov 06, 2018 4:38 am
Yet how do you feel if you sit there having gotten FJ right from first place and yet you're going home? An optimal bet is only optimal if you know how others are going to bet. In this case, I don't see anything wrong with either bet.
For that to be true, there would have to be situations where my bet loses and Mary Ann's wins and vice versa, depending on what Jonathan bets and who gets FJ. Play with the numbers and see if you're right. Here's the situation going into FJ:

Jonathan Greenan: 14400
Mary Ann Borer: 19200

To make things easy (because this is actually a special situation where one can think strategically about tying, but let's ignore that) I'll recommend that Mary Ann bet $9601.

Mary Ann actually bet $19000.

The challenge: Find me a situation where Mary Ann would have won with her bet but lost with mine.

User avatar
opusthepenguin
The Best Darn Penguin on the Whole JBoard
Posts: 6070
Joined: Thu Aug 11, 2011 2:33 pm
Location: Shawnee, KS
Contact:

Re: Monday, November 5, 2018 Game Recap and Discussion (SPOILERS)

Post by opusthepenguin » Tue Nov 06, 2018 10:55 am

John Boy wrote:
Tue Nov 06, 2018 10:12 am
Fun Fact: Cleveland does not have a sister city, but we do have a suicide pact with Detroit.
:lol:

User avatar
LucarioSnooperVixey
No Life
Posts: 481
Joined: Sun Jan 15, 2017 8:41 pm
Location: New Jersey

Re: Monday, November 5, 2018 Game Recap and Discussion (SPOILERS)

Post by LucarioSnooperVixey » Tue Nov 06, 2018 12:45 pm

53 R
DD: 3/3
FJ: :mrgreen:
LT: Column, (Alfred the Great), Mammoth Cave, Lasceaux, Baseball

Only misses were A Star is Born, Sarah Huckabee Sanders, Cleveland Browns, Plug, Claire Foy, and Larder.
Known in the real world as Douglas Squasoni.

User avatar
AFRET CMS
Jeopardy! Champion
Posts: 723
Joined: Thu Jan 26, 2017 2:48 pm
Location: Colorado

Re: Monday, November 5, 2018 Game Recap and Discussion (SPOILERS)

Post by AFRET CMS » Tue Nov 06, 2018 1:50 pm

Golf wrote:
Tue Nov 06, 2018 8:32 am
And yes, there are optimal wagers in Jeopardy. That's what game theory is all about.
Unfortunately, "optimal actions" in classical game theory assume that the opponent is acting in his/her own best interest. That's why masters sometimes hate playing against tyros -- the master almost always wins, but the rookie's unpredictability makes for a trying situation.

The German general staff strongly advised against the Ardennes offensive, but Hitler overruled them. The Germans lost the Battle of the Bulge, but our military game theorists weren't expecting it and it turned into a long hard slog. The outcome was never in doubt, but by disregarding (actually, not knowing) optimal game theory actions, Hitler threw a lot of sand into the Allied military machinery.

Similarly, game theory suggested the best place to invade was Pas de Calais. That's what the German's expected and planned for, and the Allies used deception to make the Germans think it would happen. The Normandy invasion succeeded partly because the Germans thought it was a diversion.

Game theory, at least as applied to military strategy, is usually unreliable past one level of "my opponent think I'm going to do A, so I'll do B instead," and almost always useless after two levels. However, the more knowledgeable the opponents, the more predictable they tend to be.

After two levels, optimum choice is usually not much better than a random selection. That's why Sun Tzu recommended not just unpredictability, but unforeseeability. Wesley proved the worth of that with Vizzini.
I'm not the defending Jeopardy! champion. But I have played one on TV.

User avatar
alietr
Site Admin
Posts: 5951
Joined: Thu Jun 30, 2011 1:20 pm
Location: Bethesda, MD

Re: Monday, November 5, 2018 Game Recap and Discussion (SPOILERS)

Post by alietr » Tue Nov 06, 2018 1:58 pm

AFRET CMS wrote:
Tue Nov 06, 2018 1:50 pm
After two levels, optimum choice is usually not much better than a random selection. That's why Sun Tzu recommended not just unpredictability, but unforeseeability. Wesley proved the worth of that with Vizzini.
Westley.

User avatar
Bartleby
Loyal Jeopardista
Posts: 217
Joined: Sat Apr 22, 2017 12:55 am
Location: Joplin, Missouri

Re: Monday, November 5, 2018 Game Recap and Discussion (SPOILERS)

Post by Bartleby » Tue Nov 06, 2018 2:24 pm

35,200 coryat. Ran Sister Cities, Rhymes with Mug, Poker Face, and Mars.

Instaget final. Through the first part of the clue, I briefly considered the Pevensies from the Chronicles of Narnia, but switched to the right answer at "this character."
I would prefer not to.

User avatar
opusthepenguin
The Best Darn Penguin on the Whole JBoard
Posts: 6070
Joined: Thu Aug 11, 2011 2:33 pm
Location: Shawnee, KS
Contact:

Re: Monday, November 5, 2018 Game Recap and Discussion (SPOILERS)

Post by opusthepenguin » Tue Nov 06, 2018 3:29 pm

AFRET CMS wrote:
Tue Nov 06, 2018 1:50 pm
Golf wrote:
Tue Nov 06, 2018 8:32 am
And yes, there are optimal wagers in Jeopardy. That's what game theory is all about.
Unfortunately, "optimal actions" in classical game theory assume that the opponent is acting in his/her own best interest. That's why masters sometimes hate playing against tyros -- the master almost always wins, but the rookie's unpredictability makes for a trying situation.

[snip]

After two levels, optimum choice is usually not much better than a random selection. That's why Sun Tzu recommended not just unpredictability, but unforeseeability. Wesley proved the worth of that with Vizzini.
Take my challenge offered above. Play with the numbers. Then see if you still want to say this. If you can find even one instance where Mary Ann's bet wins the game for her and mine doesn't, I'll concede. Along the way you'll discover that there are MANY instances where MY bet wins the game for her and hers doesn't. If the objective is winning the game, my bet is statistically better than random selection, and it's better than Mary Ann's.

User avatar
jeff6286
Watches Jeopardy! Way Too Much
Posts: 4306
Joined: Fri Jul 01, 2011 7:34 pm
Location: Indianapolis, IN

Re: Monday, November 5, 2018 Game Recap and Discussion (SPOILERS)

Post by jeff6286 » Tue Nov 06, 2018 4:00 pm

opusthepenguin wrote:
Tue Nov 06, 2018 3:29 pm
AFRET CMS wrote:
Tue Nov 06, 2018 1:50 pm
Golf wrote:
Tue Nov 06, 2018 8:32 am
And yes, there are optimal wagers in Jeopardy. That's what game theory is all about.
Unfortunately, "optimal actions" in classical game theory assume that the opponent is acting in his/her own best interest. That's why masters sometimes hate playing against tyros -- the master almost always wins, but the rookie's unpredictability makes for a trying situation.

[snip]

After two levels, optimum choice is usually not much better than a random selection. That's why Sun Tzu recommended not just unpredictability, but unforeseeability. Wesley proved the worth of that with Vizzini.
Take my challenge offered above. Play with the numbers. Then see if you still want to say this. If you can find even one instance where Mary Ann's bet wins the game for her and mine doesn't, I'll concede. Along the way you'll discover that there are MANY instances where MY bet wins the game for her and hers doesn't. If the objective is winning the game, my bet is statistically better than random selection, and it's better than Mary Ann's.
Winning the game is not always the only goal. (In theory the only time this is the case is in a tournament semifinal) It certainly should be the primary goal but how much that is the case could vary from person to person. If children’s literature is her dream category she’s certainly well within her rights to go for that extra $9k. Say I’m a four-day champ and just need one more win to lock up a TOC berth, put “Seinfeld Quotes” up there for Final Jeopardy and I’m gonna have a really hard time turning down the opportunity to go for an extra $10k over the MSB, Even if it does open up the possibility that I blow the game due to over betting and miss the chance to both keep playing and ensure a shot at $250k in the TOC.

User avatar
AFRET CMS
Jeopardy! Champion
Posts: 723
Joined: Thu Jan 26, 2017 2:48 pm
Location: Colorado

Re: Monday, November 5, 2018 Game Recap and Discussion (SPOILERS)

Post by AFRET CMS » Tue Nov 06, 2018 11:48 pm

opusthepenguin wrote:
Tue Nov 06, 2018 3:29 pm
AFRET CMS wrote:
Tue Nov 06, 2018 1:50 pm
Golf wrote:
Tue Nov 06, 2018 8:32 am
And yes, there are optimal wagers in Jeopardy. That's what game theory is all about.
Unfortunately, "optimal actions" in classical game theory assume that the opponent is acting in his/her own best interest. That's why masters sometimes hate playing against tyros -- the master almost always wins, but the rookie's unpredictability makes for a trying situation.

[snip]

After two levels, optimum choice is usually not much better than a random selection. That's why Sun Tzu recommended not just unpredictability, but unforeseeability. Wesley proved the worth of that with Vizzini.
Take my challenge offered above. Play with the numbers. Then see if you still want to say this. If you can find even one instance where Mary Ann's bet wins the game for her and mine doesn't, I'll concede. Along the way you'll discover that there are MANY instances where MY bet wins the game for her and hers doesn't. If the objective is winning the game, my bet is statistically better than random selection, and it's better than Mary Ann's.
The challenge to which you were referring:
opusthepenguin wrote:
Tue Nov 06, 2018 10:53 am
Here's the situation going into FJ:

Jonathan Greenan: 14400
Mary Ann Borer: 19200

To make things easy (because this is actually a special situation where one can think strategically about tying, but let's ignore that) I'll recommend that Mary Ann bet $9601.

Mary Ann actually bet $19000.

The challenge: Find me a situation where Mary Ann would have won with her bet but lost with mine.
Nope, not gonna argue -- I wasn't disagreeing with your contention about this particular situation; I was disagreeing (at least slightly) with Golf's contention about game theory offering strong guidance.

For this situation, first-order optimization would be $9601 for Mary Ann to cover Jonathan's potential double -- as you recommended.

Classical game theory would suggest then than Jonathan NOT bet to double, but to instead wager $4800 to win on a double stumper.

Knowing that, Mary Ann's second-order optimization would then be to assume that Jonathan will do that and achieve $19,100 if he gets it correct -- and she should therefore wager $99 to win if he gets it right and she gets it wrong.

And if Jonathan assumes she will do THAT, then he should wager $4900 to win if he gets it right, regardless of her answer.

Then if Mary Ann assumes he does that, her third-order optimization would be...... and then Jonathan would .....

At some point, somebody has to drink from a goblet.

Game theory strategies can evolve into a regressive series -- but it assumes each person knows the optimum strategic response for each of the opponent's actions. Jeopardy is replete with examples showing that is not true. Add in various comfort levels with different categories, and game theory will only take you so far.

Were I in that game, I wouldn't go past first-level optimization UNLESS I had seen my opponent in several games before I went up against him or her. The point I was originally making is that game theory is not a code of behavior -- "not rules, more like guidelines" (to switch movie references).

And finally, thanks to alietr for correcting Wesley to Westley. Long time since I saw the movie.
I'm not the defending Jeopardy! champion. But I have played one on TV.

Post Reply