Grand Game Strategy?

[Jordanar18]

Would it be reasonable to save an easy GP for your last pick in Grand Game?  If you are unsure about one, you can pick it third to be sure you win $100.  Seems like no one has taken $1,000 in ages.

[Mallory16]

It would be reasonable, yes.  It's just not something that would occur to most contestants on stage, but if you have the presence of mind while up there, then sure, nothing wrong with playing it that way.

[Axl]

But why would you want to make it harder on yourself to get to $1,000 in the first place?  Deliberately sacrificing your bailout option just for a "sure" $100 means you're potentially costing yourself $900.

[noahproblem]

But why would you want to make it harder on yourself to get to $1,000 in the first place?  Deliberately sacrificing your bailout option just for a "sure" $100 means you're potentially costing yourself $900.

Well you would be a rare person who would even consider bailing out at $1,000 today (obviously back when the game first started $1,000 was worth a lot more than it would be today, and it would be a tougher decision to pass that up).

[WooWho]

Would it be reasonable to save an easy GP for your last pick in Grand Game?

It's certainly what I TRIED to do when I got up there and played it...but when you look at the lineup of products I had, there really wasn't any easy option for a backup.  There were about two or three gimmes, but there wasn't a definite one that screamed "pick me last" and three others that were for sure.

These days, there really isn't any good reason to take the $1,000--the reward outweighs the risk, and even if you blow it, you could potentially get it back at the Wheel.

[pannoni1]

The obvious, more important strategy is to go for the $10,000 rather than bail, since 2:1 odds for a 9:1 payout shows that if you played this game three times and reached the $1,000 level, stopping all the time would give you $3000 while going for it all three times would give you an average of $10,000. Unlike Pay The Rent, this game isn't nearly as manipulative and its clear that picking the one of the two highest priced GP's are the danger prices.

Based on the OP's idea, the strategy would go to start with the second cheapest item, followed by the third cheapest, then the fourth cheapest, and finally the cheapest item.

[Axl]

Well you would be a rare person who would even consider bailing out at $1,000 today...

Then that seems to be a sign that people need more self-control, rather than that a whole new strategy needs to be developed to protect contestants from themselves in the hopes of salvaging a hundred bucks.

[Mallory16]

Self-control?  Why?  Even if you're picking a product at random, the risk versus reward for the game so ridiculously favors risking the $1000.  It's just so overly worth going for it, even if you have absolutely no clue which of the remaining products is under the target price.

And let's face it, the game would be far, far less interesting if most people did bail out with the $1000.

[SteveGavazzi]

And let's face it, the game would be far, far less interesting if most people did bail out with the $1000.

When did we start talking about Temptation?

[vadernader]

And let's face it, the game would be far, far less interesting if most people did bail out with the $1000.

When did we start talking about Temptation?

I believe he was talking about It's in the Bag.

[whammy007]

I thought he was taking about Spelling Bee... 

Regardless, the $1,000 bailout looked pretty good in 1980 when the game debuted (adjusted for inflation, that's around $2,800 in today's dollars). But nowadays, $1,000 isn't nearly as big of a thing, particularly when the One-Bids frequently exceed that number.

[stardf29]

The obvious, more important strategy is to go for the $10,000 rather than bail, since 2:1 odds for a 9:1 payout shows that if you played this game three times and reached the $1,000 level, stopping all the time would give you $3000 while going for it all three times would give you an average of $10,000.

I'd just like to point out that any given person is only likely to play Grand Game once, so this logic isn't really the best.

It's still not really worth it to take $1,000, especially nowadays when it's not worth that much. (Back when Grand Game debuted, $1K was actually a pretty decent sum of money and potentially not worth risking, but then inflation happened.)

[TPIRighteous]

I'd just like to point out that any given person is only likely to play Grand Game once, so this logic isn't really the best.

It's right, although phrased a bit misleadingly. Just divide by 3; if you had $1,000, and you picked a product eeny-meeny-miny-moe, you'd win $10,000 one out of three times—an average payout of $3,333.33.

Of course, we're far below that $3K+ average for the season because of forced choice through deception, which is a factor that skews the odds of winning. But if you're accounting for it in your in-game decision making, you're probably trying to beat it anyway.

[stardf29]

^ It's mathematically "right", but the reason it doesn't really work is because the average contestant isn't going to get to that point in Grandy and think, "Okay, I'll go for the $10K and if I don't get it on this 1-out-of-3 shot, I'll just try again next time I play this game." For the contestant playing the game, you either get it right and get the $10K, decide to walk away with the $1K, or get it wrong and get nothing. No trying again, no playing the game multiple times to accumulate $1K multiple times, just a one-and-done decision that you might end up regretting later.

Really, it's not even all that "right" from a mathematical perspective, because what you're trying to do is to statistically evaluate a probability distribution... with a sample size of one. The standard deviation on that is going to be incredibly large, and as a result, almost any statistical analysis on "is doing X a better choice" is going to return "data inconclusive". This applies to pretty much any game on Price; probability analysis isn't bad, but for the contestant playing the game, there's far more than just the raw numbers involved.