Pricing Game Odds...

[tvmitch]

This is a question that Mandel might be most qualified to answer, but I think it makes for good discussion:

While watching the Spelling Bee game on today's TPiR rerun, the thought came to mind: are these pricing games designed so that the smartest contestant will always have a winning advantage? In other words, when the games are designed, is there a certain set of odds that the designer is supposed to keep in mind for different stages of the game...so if the contestant plays perfectly, s/he should win x% of the time?

For instance, in Spelling Bee, the contestant only got 2 of the 3 products right and had only 4 of 5 cards. She ended up winning the game (naturally, getting C-A-R spelled out). But if a contestant played the game perfectly and got all 5 cards, are the odds in favor for them to win? This spurs a (fascinating, for me) side conversation about probability in all these games.

When creating Let 'Em Roll, for example...at least 90% of contestants get three rolls, I think it's fair to say. I don't have the time at the minute to calculate the expected odds of winning, but are those odds in line with all the other games?

Secret X is a blatant exception (there may be more) where the contestant can play the game perfectly and still lose. Half Off is another that's based completely on state-lottery game show luck when it comes down to the final choices...and I know that Mandel created that one, and I would be interested to hear if that topic was brought up in deliberations about the game.

In comparison, Spelling Bee seems to me like a different kind of game in this vein because you always have the choice to take some money, some kind of out.

I apologize if this post isn't coherent! Discuss.

[gwarman2005]

I don't know about the odds, lots of times it seems that the contestant is so nervous that it all boils down to dumb luck.  And we all know it's better to be lucky than good.

Personally, I'd discuss more but I'm still giggling over the trip to New Orleans being offered in the showcase.  Holy bad timing Batman!

[JasonA1]

Personally, I'd discuss more but I'm still giggling over the trip to New Orleans being offered in the showcase. Holy bad timing Batman!

While this whole post really just deserves a tsk tsk and "..." in response, I couldn't let this go: they're in freakin' reruns. For another week.

-Jason

[clemon79]

[quote name=\'mitchgroff\' date=\'Sep 8 2005, 08:44 AM\']When creating Let 'Em Roll, for example...at least 90% of contestants get three rolls, I think it's fair to say. I don't have the time at the minute to calculate the expected odds of winning, but are those odds in line with all the other games?
[/quote]
We figured the odds of winning Let 'Em Roll with all three rolls in this thread, and concluded that the chances of winning are over 50%. 1/2-Off, is, of course, right at 50%, and Secret X is at 66%. Five Price Tags is 80%, and is a car game. So I'd have to say it fluctuates pretty wildly.

[Neumms]

[quote name=\'clemon79\' date=\'Sep 8 2005, 01:15 PM\']We figured the odds of winning Let 'Em Roll with all three rolls in this thread, and concluded that the chances of winning are over 50%. 1/2-Off, is, of course, right at 50%, and Secret X is at 66%. Five Price Tags is 80%, and is a car game. So I'd have to say it fluctuates pretty wildly.

[snapback]96143[/snapback]

[/quote]

One should also consider how difficult the small prize games are. The true-false part of Five Pricetags is much harder than Let 'Em Roll's. Plinko, despite the player's choice being somewhat similar to true-false (if the price seems about right, the second digit must be the wrong one), seems way more forgiving.

The question is, how consistently is the difficulty manipulated? Is Let 'Em Roll always set up to have fairly obvious answers, or if it's won a few times in a row, will they choose a more challenging set of groceries?

[Steve Gavazzi]

[quote name=\'Neumms\' date=\'Sep 8 2005, 03:06 PM\']The question is, how consistently is the difficulty manipulated? Is Let 'Em Roll always set up to have fairly obvious answers, or if it's won a few times in a row, will they choose a more challenging set of groceries?

[snapback]96152[/snapback]

[/quote]

I've seen pretty much every playing of Let 'em Roll, and I'd say that the grocery portion has been truly hard maybe five times.  Furthermore, Roger has stated that they generally make the pricing portion simple on purpose, so it seems like they're more concerned with making it interesting to watch than balancing out the win-loss record.

[Strikerz04]

I wrote a paper senior year of high school, where my friend and I described the basis of some of the pricing games (I wanted to add "let em roll", but we have time constraints).

We got a 100 on it, and if you want to see it, PM me. It was basic as it gets, my be slightly off, but then again, I'm not a math major.

-Dave

[Steve Gavazzi]

[quote name=\'Strikerz04\' date=\'Sep 8 2005, 06:22 PM\']We got a 100 on it, and if you want to see it, PM me.

[snapback]96173[/snapback]

[/quote]

Before anyone tries to do this, someone should probably point out that the Private Message feature is disabled on this forum.

[chris319]

when the games are designed, is there a certain set of odds that the designer is supposed to keep in mind for different stages of the game

The answer to your question is no. The games are designed principally for entertainment value. By manipulating various aspects of a game it is possible to make it easier (better odds for the contestant) than a random pick, making odds calculations moot. They have a naive idea of how winnable games are, and I know that they do not calculate the expected odds when a game is created, but instead keep records after the fact. Not surprisingly, playing time is given more consideration than you would think. There are times when they want games which play quickly, where there are only one or two prizes to describe and the contestant only has to make one decision.

Most Expensive is a good example of a highly stackable game. The random odds of winning are 1 out of 3, but it can be stacked so that anyone who does not have a metal plate in their head can win with almost 100% certainty.

Having said all of that, I'm sure that when the insurance policy was taken out for the million-dollar spin, the game and its odds were analyzed with a fine-tooth comb before the policy was issued.

[Robert Hutchinson]

I always seem to be late to the TPIR odds discussions. If I did my math right, and it's entirely possible that I didn't, having all five cards in Spelling Bee gives you a 31% chance of winning merely by finding a CAR card.

I think one distinction that should be made, especially considering the aspects Mitch was  asking about, is that even without the producers "stacking" a game, the odds always vary when pure luck is replaced by rewarding knowledge. Someone with perfect knowledge of all prices, ever, will lose Joker 20% of the time. Someone with perfect knowledge of all prices, ever, will never lose 5 Price Tags, since they'll know the price of the car.

I'm pretty sure the game that is most rewarding in the reverse situation (no pricing knowledge, perfect odds knowledge) is Barker's Markers. Throw darts at the prices, always give up the $500, and you'll win 75% of the time.

[Dbacksfan12]

[quote name=\'Robert Hutchinson\' date=\'Sep 10 2005, 07:15 PM\']Someone with perfect knowledge of all prices, ever, will never lose 5 Price Tags, since they'll know the price of the car.
[/quote]
Couldn't you make that argument for other games as well?  (e.g. Buy or Sell, Double Prices....), or is their some kind of mathematical equation that I'm not catching on to?

[clemon79]

[quote name=\'Modor\' date=\'Sep 10 2005, 10:12 PM\']Couldn't you make that argument for other games as well?  (e.g. Buy or Sell, Double Prices....), or is their some kind of mathematical equation that I'm not catching on to?
[/quote]
Yes, you could. The point is, you CAN'T make that argument about Joker or Secret X or Plinko because you are dealing with the vagaries of raw chance and/or gravity.

[Steve Gavazzi]

More specifically, by just their base odds, there is no difference between Joker and Five Price Tags, the two games Robert was comparing...but when you factor in pricing knowledge, they become two totally different animals.