Do you think that Arthur Chu is going to change how the game is played?

[Magna]

He changed nothing.

This is the kind of thing people yell while shaking their fists at the sky.

[Golf]

He changed nothing.

This is the kind of thing people yell while shaking their fists at the sky.

Hey, I wish he had. I love seeing the game played with perfect strategy.

Now get off my lawn.

[kristinsausville]

Or maybe some contestants analyze Arthur's strategy and decide it isn't for them, for whatever reason. After all, Julia Collins won almost twice as many games as Arthur without using it.

Maybe, but we all know the vast majority of contestants have zero clue on how to play optimally, or even that such a concept exists.

And yes, Julia won a ton of games while playing sub-optimally. Had she played optimally she might have won 40, or 4, or whatever. But regardless, she did not give herself the best chance of winning. We've seen contestants play perfectly optimal and lose, and also play downright awful and win. Such as throwing away a lock on the final clue before winning on a FJ TS with zero wagering sense.

I know you had all this knowledge going into your taping, and probably devised some sort of strategy on how you wanted to play. Which is much more than most contestants.

Arthur Chu played near perfect in terms of DD and FJ wagering. Since his run we've seen the continued horrid DD wagering as well as tons of poor FJ wagers. Anybody that thinks he changed how the game is played needs to realize this. He changed nothing.

No, I'm saying that "playing optimally" isn't necessarily an objective thing that applies equally to all contestants. I knew that that strategy would do nothing but throw myself off as much as anyone else, so it wasn't optimal for me.

Besides that, it doesn't work effectively unless you have really good timing on the buzzer. And if you have really good timing on the buzzer, do you really need a special strategy?

[Rex Kramer]

Besides that, it doesn't work effectively unless you have really good timing on the buzzer. And if you have really good timing on the buzzer, do you really need a special strategy?

*Ding!* We have a winner!

Rex

[dhkendall]

Or maybe some contestants analyze Arthur's strategy and decide it isn't for them, for whatever reason. After all, Julia Collins won almost twice as many games as Arthur without using it.

Maybe, but we all know the vast majority of contestants have zero clue on how to play optimally, or even that such a concept exists.

And yes, Julia won a ton of games while playing sub-optimally. Had she played optimally she might have won 40, or 4, or whatever. But regardless, she did not give herself the best chance of winning. We've seen contestants play perfectly optimal and lose, and also play downright awful and win. Such as throwing away a lock on the final clue before winning on a FJ TS with zero wagering sense.

One would think, then, that success in Jeopardy! is dependent on a whole lot of luck, then ...

[MDaunt]

Or maybe some contestants analyze Arthur's strategy and decide it isn't for them, for whatever reason. After all, Julia Collins won almost twice as many games as Arthur without using it.

Maybe, but we all know the vast majority of contestants have zero clue on how to play optimally, or even that such a concept exists.

And yes, Julia won a ton of games while playing sub-optimally. Had she played optimally she might have won 40, or 4, or whatever. But regardless, she did not give herself the best chance of winning. We've seen contestants play perfectly optimal and lose, and also play downright awful and win. Such as throwing away a lock on the final clue before winning on a FJ TS with zero wagering sense.

One would think, then, that success in Jeopardy! is dependent on a whole lot of luck, then ...

And timing on the buzzer...

But mostly timing on the buzzer.

[Vermonter]

One would think, then, that success in Jeopardy! is dependent on a whole lot of luck, then ...

At least two lots, I would imagine. Possibly a third.

[BigDaddyMatty]

Arthur Chu played near perfect in terms of DD and FJ wagering. Since his run we've seen the continued horrid DD wagering as well as tons of poor FJ wagers. Anybody that thinks he changed how the game is played needs to realize this. He changed nothing.

All of this is irrelevant. Not only does one need not to have changed everything to have changed something, but the claim in question at the moment is

The answer to the thread title is yes, right? Arthur literally changed how the game is played, with regards to the tiebreaker rule.

It has been established that a) shortly after Arthur offered several ties, b) several other contestants also offered ties, followed by c) TPTB changing the rules of the game to disallow ties. The only open issues are 1) whether a caused b and 2) whether b caused c. No one seems to be questioning #2, which leaves only #1. Unless the increase in tie offers post-Arthur was mere coincidence, it is a fact that Arthur's gameplay permanently changed the game of J!

[Vermonter]

Here is a list of all games since Season 29 in which the leader (1) did not have a lock and (2) wagered to have exactly twice second's score if correct

(Revised to incorporate additional data and to correct a formula)

non-lock 2B tie offers S29-31 v2.png (76.33 KiB) Viewed 4454 times

[lieph82]

Cool stuff, Keith. Do we know approximately the date when we can start assuming that most contestants would have seen Arthur's games before playing themselves?

[whatisbishkek]

So now it's about contestants offering the tie as opposed to actual tie games?

Are you suggesting that there's no relationship between how often the leader offered a tie and how often a tie occurred?

Yes.

Can you honestly look at this score and say, with a straight face, the leader -offered- a tie?
http://www.j-archive.com/showgame.php?game_id=4714

Wow.

Let me rephrase that to make sure we're on the same page.

You are saying that if the leader always offered a tie, ties would occur at the same rate as if the leader never offered a tie?

The fact that ties can, on rare occasions, occur when the leader does not bet for a tie doesn't change the fact that ties are more likely when the leader does bet for a tie.

I mean, I am really baffled that I have to explain this. The assertion you appear to be making by answering "yes" to my question is completely absurd.

I don't think you've established your premise: that more ties were offered post-Chu than pre-Chu.

Thanks to Vermonter for providing the stats, which do show the trend I expected.

Edit: by adding the numbers, it looks like pre-Arthur there were 10 ties offered over the course of about 287 games. That's 3.48%. It would be interesting to see an even larger sample, but I'll assume that was roughly the normal level. Post-Arthur, starting from the first tie offered (the long gap undoubtedly corresponds to the time period after Arthur's games filmed but before they aired), there were 14 ties offered in about 60 games, or 23.33%. Using a binomial probability calculator, the probability of getting at least 14 successes out of 60 trials when the probability of success in each trial is 3.48% is less than 1.5*10^-8. That is, the probability of the increase in ties offered being the result of random variance is about 1.5 in 100 million.

2nd edit: the table was updated while I was writing this, but my conclusions don't meaningfully change.

[Vermonter]

I just updated the table above with additional data – thanks to dhkendall for the suggestions and catch of incorrect date

[Vermonter]

Cool stuff, Keith. Do we know approximately the date when we can start assuming that most contestants would have seen Arthur's games before playing themselves?

Julia said she taped wins 6-10 on February 4, after the Chuphoria had begun. Those aired starting April 28.

[MDaunt]

I concede the point.

[Vermonter]

Here you go – Game AKJ #1 is Nancy Zerg's first win

Code: Select all

Game AKJ  	Date      	∆ games   	Player              	B all-in? 	Tie?
40        	1/24/05   	          	Ted Stratton        	          	
47        	6/2/05    	7         	tie                 	one       	
125       	11/29/05  	78        	tie                 	one       	
144       	12/26/05  	19        	Peter Rubin         	          	
168       	1/27/06   	24        	Paul Canty          	          	
286       	9/20/06   	118       	tie                 	both      	yes
332       	12/14/06  	46        	Cory Hartman        	          	
388       	3/16/07   	56        	Scott Weiss         	yes       	3-way
405       	4/10/07   	17        	David Haglund       	          	
486       	10/17/07  	81        	Linda Zell Randall  	yes       	yes
528       	12/28/07  	42        	tie                 	both      	yes
529       	12/31/07  	1         	Dan Pawson          	          	
536       	1/9/08    	7         	Dan Pawson          	yes       	
543       	1/18/08   	7         	Susan Forman        	          	
562       	2/28/08   	19        	Gwynne Ash          	yes       	yes
598       	4/18/08   	36        	Gabriel Schechter   	yes       	
609       	5/19/08   	11        	Mary Kay Schmidt    	          	
619       	6/2/08    	10        	Chris Vestuto       	          	
647       	7/10/08   	28        	tie                 	both      	yes
661       	9/10/08   	14        	Laura Novak         	yes       	
665       	9/23/08   	4         	Elza Reeves         	yes       	
730       	1/6/09    	65        	Ranjan Ramchandani  	yes       	yes
745       	1/27/09   	15        	Andy Walvoord       	          	
764       	2/23/09   	19        	Kenneth Burns       	          	
834       	6/30/09   	70        	Nina Ginocchio      	          	
849       	9/15/09   	15        	Enrique Machado     	          	
887       	11/25/09  	38        	Robert Bethune      	          	
907       	12/24/09  	20        	Stephen Weingarten  	          	
938       	2/23/10   	31        	Tom Toce            	          	
939       	2/24/10   	1         	Tom Toce            	          	
953       	3/16/10   	14        	Amanda Baber        	          	
1059      	10/25/10  	106       	Pam Jones-Pigott    	          	
1063      	10/29/10  	4         	Marie Braden        	          	
1079      	12/6/10   	16        	Christina Barley    	          	
1122      	2/3/11    	43        	Fred Cofone         	yes       	
1153      	4/6/11    	31        	Christopher Short   	          	
1243      	10/19/11  	90        	Liz Greenwood       	          	
1249      	10/27/11  	6         	Sunny Stalter       	          	
1274      	12/15/11  	25        	tie                 	both      	
1287      	1/3/12    	13        	Nicholas Campiz     	yes       	yes
1325      	3/23/12   	38        	Dennis Wright       	          	
1365      	6/8/12    	40        	Aaron Cappocchi     	          	
1369      	6/14/12   	4         	Kathy Wright        	          	
1385      	7/6/12    	16        	tie                 	one       	
1436      	11/5/12   	51        	Paul Nelson         	          	
1437      	11/6/12   	1         	Paul Nelson         	          	
1476      	1/14/13   	39        	Kristin Morgan      	yes       	yes
1492      	3/5/13    	16        	Dylan Wint          	          	
1502      	3/19/13   	10        	Lauren Girard       	          	
1523      	4/17/13   	21        	Michelle Martin     	          	
1540      	5/24/13   	17        	Mike Lewis          	          	
1549      	6/6/13    	9         	Greg Draves         	          	
1568      	7/3/13    	19        	Hunter Sandison     	          	yes
1612      	10/22/13  	44        	Bill Tolany         	          	
1672      	1/28/14   	60        	Arthur Chu          	          	
1673      	1/29/14   	1         	Arthur Chu          	yes       	yes
1675      	1/31/14   	2         	Arthur Chu          	          	
1680      	2/28/14   	5         	Arthur Chu          	          	
1705      	4/18/14   	25        	E.A. Srere          	          	
1741      	6/23/14   	36        	James Friscia       	yes       	
1752      	7/8/14    	11        	Campbell Warner     	          	
1753      	7/9/14    	1         	Campbell Warner     	yes       	
1754      	7/10/14   	1         	Campbell Warner     	          	
1755      	7/11/14   	1         	Andrew Harris       	yes       	
1757      	7/15/14   	2         	Ed Patterson        	          	
1761      	9/15/14   	4         	Elizabeth Williams  	          	
1770      	9/26/14   	9         	Alan Lange          	yes       	yes
1774      	10/2/14   	4         	Shawn Choe          	yes       	
1775      	10/3/14   	1         	Shawn Choe          	          	
1781      	10/13/14  	6         	Dan Tran            	          	yes
1783      	10/15/14  	2         	John Campbell       	yes       	
1790      	10/24/14  	7         	Matthew LaMagna     	          	
1792      	10/28/14  	2         	Bill Albertini      	yes       	yes
1801      	11/24/14  	8         	ties eliminated     	          	

[Rex Kramer]

I concede the point.

Oh, you give up too easily. The data shows that Arthur's games aired in time for Julia and all those other players between her and James Friscia to have seen them, but the frequency for betting for the tie did not increase with them. Only once Campbell Warner starts playing does the frequency increase appreciably, and the last person he had seen betting for the tie would have been E.A. Srere, almost two months after Arthur's run. I think the only logical conclusion is that E.A. Srere changed the way the game is played. (Note that Srere did not have the benefit of watching Arthur's games beforehand, so we can't even say that Arthur changed the game indirectly.)

Rex

P.S. -- Furthermore, Campbell Warner's betting for the tie probably influenced Andrew Harris to do so, since Harris was in studio that taping day watching Warner bet for the tie, and may possibly have influenced Ed Patterson, if Patterson taped his show the following day and had also been in studio watching Warner play. After Patterson, the next players who bet for the tie are all in the next season, and would have seen the spate of for-the-tie bets of Warner, Harris, and Patterson at the very end of the previous season. So if it's not Srere who is responsible for the trend of for-the-tie bets, there's even stronger evidence that it's Warner, not Chu, who prompted them.

[ACW]

The answer to the thread title is yes, right? Arthur literally changed how the game is played, with regards to the tiebreaker rule.

What was the old rule, what's the new one?

[econgator]

What was the old rule

There can be ties.

what's the new one?

There can't be ties.

[whoisalexjacob]

Serious question: in how many games was a pre-FJ leader saved (won when they would have lost by $1 with the cover bet) by betting for the tie (excluding conventional bet-for-the-tie situations)?

[jeff6286]

Serious question: in how many games was a pre-FJ leader saved (won when they would have lost by $1 with the cover bet) by betting for the tie (excluding conventional bet-for-the-tie situations)?

From Keith's chart I found:

7/3/13
http://www.j-archive.com/showgame.php?game_id=4242

10/13/14
http://www.j-archive.com/showgame.php?game_id=4694

If I'm reading the chart correctly, those may be the only instances since Ken Jennings of ties where the leader would've lost by a dollar if they made the MSB. There may be more instances where they won by a dollar instead of tying.