Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Robz888]

The game itself was a bit of a mess, as Black Market games tend to be: http://dominion.isotropic.org/gamelog/201210/30/game-20121030-131013-6ddf67b2.html

I pulled Witch, he got King's Court, so you know how it is.

The interesting thing is the last turn. I have $7 in hand, plus Witch and Militia. If I play Militia, I have $9 to buy the last Province and tie him, 36-36. If I play Witch, he takes a Curse, and as long as I draw at least $1, I can buy the Province and win 36-35.

I assume he would be able to buy the Province (I'm dreadfully wrong, but I don't know that), so not drawing $1 with Witch seems like it's going to be a loss for me. What would you do?

In the end, I take the tie. The fact that I had trashed a couple Coppers with Spice Merchant (from the BM) is what swayed me. Odds were too high I draw Estate-Duchy, or something.

[ednever]

You been tieing all over the place these days Rob...

My general rule is to take the tie- unless you are very sure you can get the win or it's a tournament setting where a tie and a loss mean the same for some reason.

Ed

[shark_bait]

I know what you're thinking. "Will I play my Witch or my Militia?" Well, to tell you the truth, in all this excitement I kind of lost track myself which would be better. But seeing as this is a Laboratory style Black Market construed deck, the most powerful type of deck in the world, it's guaranteed to blow right past you, you've got to ask yourself one question: Do ya feel lucky? Well, do ya, punk?

I bet this was going through your head as you deliberated your decision 

[shark_bait]

[Robz888]

You been tieing all over the place these days Rob...

My general rule is to take the tie- unless you are very sure you can get the win or it's a tournament setting where a tie and a loss mean the same for some reason.

Ed

Oh, I meant to say: Sorry I disappeared on you! My landlord barged in shouting about battening down the hatches in preparation for the storm.

[Grujah]

My general rule is to take the tie- unless you are very sure you can get the win or it's a tournament setting where a tie and a loss mean the same for some reason.

I agree.

[Insomniac]

According to the log you would have drawn Lab, Courtyard.

[Beyond Awesome]

I think taking a tie is the wise decision unless you have kept careful track of your opponents deck and know for sure they can't get it. Still, it is always a risk.

[Davio]

To be able to say anything meaningful about this it's important to know how much ranking points a win would gain you.
I don't know the actual algorithm, but let me give an example.

If someone offered you $50 immediately or a 50/50 shot at either losing $100 or winning $250, you should take the 50/50 shot, as its EV is $75, $25 more than taking the $50.

So if you know what a win is going to do to your rating, you can work backwards to determine what your chances of winning should be at minimum to disregard the tie and have a positive EV.

In general, taking the tie is probably a good idea, it's just that because of your ranking there are extra variables in play.

[Ozle]

Death or glory!!!

Go for the win!

[Axxle]

Death or glory!!!

Go for the win!

QFT

I would only go for the tie if a win is impossible, or extremely improbable.

[ConMan]

If someone offered you $50 immediately or a 50/50 shot at either losing $100 or winning $250, you should take the 50/50 shot, as its EV is $75, $25 more than taking the $50.

Except that in terms of expected utility, for most people the guaranteed $50 is worth more than the chance of losing $100, never mind how much they stand to win. So the question then becomes "How much is a win worth to you?" which, I agree, can be put in terms of ranking points (as well as the value of gaining versus losing those points).

[Davio]

If someone offered you $50 immediately or a 50/50 shot at either losing $100 or winning $250, you should take the 50/50 shot, as its EV is $75, $25 more than taking the $50.

Except that in terms of expected utility, for most people the guaranteed $50 is worth more than the chance of losing $100, never mind how much they stand to win. So the question then becomes "How much is a win worth to you?" which, I agree, can be put in terms of ranking points (as well as the value of gaining versus losing those points).

This principle is also used in those TV shows where you have to open suitcases and the bank offers you some money: "Deal or no deal". In essence, not taking the bank's offer is always going to be +EV. But EV only makes sense if you can repeat the same experiment over and over again.

Because it's a once in a lifetime opportunity, the bank can always offer way less than the average of remaining unopened suitcases.

[Ozle]

You can't apply ranking in a game to real life money because other situations actually affect that real life money choice.

It's a game, go for the win!


Unless there is a prize on offer here, which there wasn't, then go for the win. The object of the game is to win, and I know there is a 'shared victory', but Donalds only really put that in there so the game doesnt tell you you are both losers!