Jeopardy! Champ Alex Jacob and Game Strategy

Even aside from daily doubles, starting at the bottom can still be a good strategy. If there’s one category that I know I’m really good at, why wouldn’t I start at the bottom? I’m not guaranteed to run the whole category: Someone might steal an answer from me, and time for the round might run out. I might as well go for the big money first, when I have the chance.

As an added bonus, if the category is my specialty, starting at the bottom decreases the chance that someone might steal it from me: The higher-value questions are harder, which makes it less likely that someone who isn’t an expert will know them.

Well, that settles that then. One more old discussion of mine rendered obsolete. :smiley:

I thought that Alex was a lot of fun to watch and I was rooting for him once I noticed what seemed like obvious stalling in his first game. It was a runaway late in double jeopardy but there was a lot of money still on the board. He slowed down his selection so much that I came here right after the show to see if anyone had started a thread yet and kept checking for the next couple hours.

I loved watching him and his various strategies but it’s not the kind of thing I’d like to see all the time. But only because it makes it harder for me to play along. Every once in a while it’s fun to watch someone work the system.

I was visiting with my dad and his wife after his second to last show and we were talking about how fun he was to watch. Dad’s wife mentioned how nice of a guy he seemed, sometimes only betting $100 on a daily double because he wasn’t greedy and didn’t want to embarrass the other contestants. She also insisted he was slowing down play because it wouldn’t be right to keep playing so fast when the other players weren’t as good as him. I didn’t spoil her view but wtf? Anyway, it was fun while it lasted.

SDMB poster Nuveena was on Jeopardy! earlier this year and she reported that it was a new rule.

[QUOTE=Nuveena]
FYI - It hasn’t played out on an episode yet, but Jeopardy has gotten rid of ties. If there’s a tie for first place, there will be a sudden death question to determine the champion. I assume if there’s a tie for second, the person who had more money going into Final Jeopardy gets second, and the other player gets third.

I don’t have a written cite, but the J Board fan site has been talking about it for awhile and the contestant coordinators confirmed it when I taped an episode in January.
[/QUOTE]

I thought he just had a touch of stage fright. First time at the free throw line, so to speak. I never noticed any stalling. Maybe I just didn’t see it.

Almost 40 Dopers have been on Jeopardy! to date.

I was on in 1991, and no, as best as I can recall, the contestant coordinators made no recommendations or suggestions about strategy or game play.

The closest anyone came to that was the stage manager who, as we were about to go into FJ said, “If you can’t come up with anything, go for funny.”

I was on earlier this year and the contestant coordinators spent a lot of time on various tips (when to ring in, you can stop and correct yourself at any time before Alex rules you incorrect, etc). This was after Arthur Chu’s episodes aired, so I’m sure the topic came up. I think they probably said the clues were designed to go from top to bottom, but I don’t remember being “strongly encouraged” to play it that way.

The returning champion on my episode was a bit of a DD hunter, whereas I have some sort of OCD need to make the board look “neat” for lack of a better word. So rather than relying on some sort of strategy, if the bottom clues were gone, I would work my way up. If the top clues were gone, I would work my way down.

I found it interesting that after six days of Alex Jacob using the Chu strategy of going through all the 3rd/4th row clues for the Daily Doubles, that the woman who defeated Jacob immediately went back to the “start at the top of the category then go down” way of playing next episode.

The winner’s strategy the next day is as you describe. The winner was a man, Todd Lovell. See for example Show #7051 - Monday, April 20, 2015 and Fikkle Fame’s Final Jeopardy: Business.

Two Plus Two Magazine did a detailed analysis of wagering strategy here:

http://www.twoplustwo.com/magazine/issue116/devin-shelly-optimal-daily-double-jeopardy.php

The author generally agrees that players are too conservative, especially in the earlier stages of the game. Near the end of the Double Jeopardy round, strategic considerations may dictate caution.

Difficulty level is an important consideration that players tend to disregard, based on the data presented. In the Double Jeopardy round, a Daily Double appearing on the $800 level has a very favorable 74% success rate, but on the $2000 level, only 57% of Daily Doubles are answered correctly.

Is there a breakdown for daily doubles found early in the round versus late in the round? I assume that earlier in the round, most people are selecting in the categories that they are likely to do well in. At the end, they may be stuck with whatever is left, and thus be in a bad category for them.

Interesting question, but no, I’m not seeing that information compiled anywhere, although it could be calculated from the data archive.

I think it was almost discussed in the analysis of WATSON computer betting, here:

But to be sure, I would need parts of it translated to “dummy” :frowning:

You’re right, I was one game off - the person who beat Alex Jacob (Todd Lovell) played the same way, but the woman who beat Todd next episode started at the top of the category and then everyone proceeding to play the top-to-bottom way.

I’m getting serious deja vu with this thread. Wasn’t there a nearly identical thread within the last year?

Ahh, here it is - regarding player Arthur Chu:

I disagree with this because you are betting money that you didn’t walk in with, but is still technically yours.

I always bet ridiculously conservative in the times I’ve gone to a table in Vegas, and sometimes a card combo will come up that would win me money if I bet on the “money for card combos” thing. The dealer always chided me for it since it was money I could have won…but since it was never mine to begin with, I didn’t care.

On the other hand, when I place a bet on something that I think I know, and I lose, it really pisses me off. My kitchey phrase for it is that I’m never mad at losing out on money I never had, but I’ll be mad at losing money that is mine.

Extrapolating that to Jeopardy, I would be hesitant to bet 3000 of my 6000 winnings on a DD unless I was sure I knew the category

The game is much more edited than you might think. If a contestant goes into the tank to an extent that’s problematic, they can just cut some of it out.

Frankly, the “Forrest Bounce” strategy is just going to become normal, eventually, because it obviously works. It’s also going to become normal to see highly variable betting on Daily Doubles (Arthur Chu, famously, would either go big or bet next to nothing, depending on his familiarity with the topic; this is, IMHO, clearly the right thing to do.) These things work, and what works always wins out. The reason we’re seeing more people trying it is that in the information age, the fact it works is more easily heard about and objectively proven.

[QUOTE=Sir T-Cups]
Extrapolating that to Jeopardy, I would be hesitant to bet 3000 of my 6000 winnings on a DD unless I was sure I knew the category
[/QUOTE]

You have no winnings at that point. You have nothing at all unless you win.

The primary objective is to finish first; the precise amount isn’t as relevant. Obviously it would be nice to win with more, but winning with $18,000 is infinitely better than finishing with $19,000 when an opponent has $19,001. Winning not only banks that money but offers you a chance to play another game and win that money, too.

In 99% of cases your betting strategy has to be focused on the mathematics of winning (the exception being if you have total confidence in the subject or none at all; if the subject was Baseball I would bet everything.) If you’re sitting near the end of the game at $6000, your opponent has $13000, and you hit a Daily Double with just $1800 of questions left on the board, you really have only two rational options; if you are very confident in the categories remaining for the $1800 left on the board, bet it all (or $5900, anyway, if you want to ensure you can’t finish with nothing) and try to surpass your opponent with the remaining questions to you’re leading going into Final. If you have little confidence in the remaining categories, bet $1400 so your opponent can’t have a runaway even if she gets the last two questions (she can max out at $14800, so you need $7400 to prevent a runaway.) I’m discounting a third opponent but that can change your calculus too; if a third opponent has $12500, there is no possible runaway and you may be safer to sit tight at $7000 if you don’t like the remaining categories and are better off hoping for a Final category you’re good at.

I could design ten different scenarios, but the long and short of it is that Jeopardy! is sort of like the chips in a winner take all poker tournament; their values do not matter unless you’re the guy left with all of them. The game is about surviving to fight another day. A huge payday is great, but the big winners are all the ones who won many games. I don’t remember Julia Collins having any ridiculously huge one day winnings, but she won a truckload of money by relentlessly winning day after day.

RickJay already gave the definitive response, but casino betting strategy is a purer decision because you’re not competing against other players. It’s similar to Final Jeopardy with an insurmountable lead. If you see a difficult category and assume that you’re unlikely to answer correctly, then your only reasonable bet amount is zero.

In a casino where the odds are against you, the most sensible bet amount is zero, otherwise table-minimum. Always consider the credibility of people giving advice in a casino. Dealers are not trained strategists.

I don’t know why it seems people get hung up on the semantics. Yes, you don’t have any “winnings”. But that’s not what he’s saying.

Treat the dollar values as points scored in a game. The one with the most points at the end wins both the game and a dollar value prize equal to the number of points.

But I still wouldn’t wager half my “points” on a category I didn’t know. If I lose my points, I don’t win the game. So betting large because “it’s not my money” means there’s a good chance none of it will be my money when I lose.

After all:

And I can’t win if I don’t have any “points”.

It depends entirely on the situation, but there are two structural factors that encourage aggression:

(1) the payout structure is nearly winner-take-all, and
(2) statistically, daily doubles are usually answered correctly

In the rare instance that you believe a question will be uncommonly difficult, you may need to bet big anyway to have a realistic chance of winning.

Each situation must be analyzed individually, but the historical evidence appears to suggest a broad tendency toward inappropriate caution.