Print

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

[Davio]

Anyone else getting a feeling this is being overanalyzed?

Venture > Silver.
And Venture > Gold at some point.
But probably never Venture > Platinum.

[WanderingWinder]

Yes. And like, you almost always want to play venture because you need the cash anyway.

Though venture CAN be better than platinum (albeit extremely rarely, and not in a deck you're likely to get in a real game)

[Davio]

Hence, the probably, meaning 'realistically'.

You could always make a deck with 10 Ventures. One Venture in hand guarantees a Province. With Plat your chain could break.

[ecq]

It was a long thread, but I think it's important to understand what cards really do.  For me, it doesn't impact when I'd play Venture.  I'd play it nearly 100% of the time.  It may impact when I'd choose to gain Venture, though.  It'd have a much bigger influence on when I'd choose to play or gain Farming Village.

[DG]

Just to put an another old argument to rest, my proof also holds true for loans. The mean value of the remaining cards in the draw deck is the same after playing a loan as before playing the loan (for a deck of victory cards and basic treasures, at least 2 treasures).

[blueblimp]

It was a long thread, but I think it's important to understand what cards really do.  For me, it doesn't impact when I'd play Venture.  I'd play it nearly 100% of the time.  It may impact when I'd choose to gain Venture, though.  It'd have a much bigger influence on when I'd choose to play or gain Farming Village.

This is where I'm coming from too. Also I like math. Hey, I'm a mathematics student, so that's my excuse. =P

[NinjaBus]

This doesn't seem right. Either I'm misunderstanding what you're arguing or you're claiming that a deck full of farming villages is equal to an identical deck of regular villages. That can't be true.

Take a basic deck of 10 coppers, 5 estates. Now add 5 farming villages. If these were regular villages, your deck would not improve. (In theory, by manipulating shuffles you can actually beat the odds slightly.) You will draw your deck, worth 10$, every three turns. However, farming villages muck that idea up. Since you have a chance at skipping over estates, you will actually be drawing through your deck faster than that. Potentially, you could improve your deck to the point of drawing 10$ in two turns, not 3. That's a significant increase.

Every time you skip over a dead card in your deck you are playing your treasures at a faster rate than you normally would. You're not magically making your future draws better, but your deck is obviously better for it. If you switch out farming villages for ventures the results are identical.

[mnavratil]

This doesn't seem right. Either I'm misunderstanding what you're arguing or you're claiming that a deck full of farming villages is equal to an identical deck of regular villages. That can't be true.

I believe the argument isn't that ventures are equivalent to silver; just that Ventures help your current hand, but do nothing for your future hand(s).
The same applies to farming village. It's better than Village because it helps the current hand. The above proof just shows that it does nothing (on average) for your next hand.

[Deadlock39]

Ninja'd a bit, but here is my attempt at explanation:

The Farming Villages absolutely improve your deck compared to an identical deck with regular Villages.  The point made in this thread is that playing them does not improve the expected value of your next hand.  What they do is draw you better cards on average this turn.  The average value of your draw deck is unaffected from turn to turn, but the average value of a hand from the Farming Village deck is definitely higher than the average value of a hand from the regular Village deck.

[Davio]

Still I think the conclusions are faulty or too simplistic and the main reason is: shuffling.

A deck with 10 Farming Villages will shuffle more often than a deck with 10 Villages.
As we all know, shuffling more often is a good thing. While I don't have hard proof I think the added value of more shuffles in the early and middle game outweighs the negative effects of shuffling in the greening stage.

This makes the whole "the average value of your future hands will not change" a bit moot. It's impractical. You're not buying nothing every turn. You buy stuff and increase your deck power. The easier you see the new stuff, the better.

To summarize I don't agree with the conclusions that have been posted because I firmly believe that the average value of both the current and future hands goes up with Farming Villages or Ventures, just because you shuffle more often. It may not be by much, but it certainly isn't 0.

[ecq]

Clearly Venture improves future hands if cycling improves future hands.  The question was whether Venture has any filtering effect or if it's purely a cycling effect.

I showed on the previous page that when you take into account shuffling, it does have a miniscule filtering effect, however that effect is not always good for your deck depending on what other treasures you have and what your target hand value is.

To sum up the conclusions:
  - Playing Venture adds $1 plus whatever it draws to your current hand, which is good.
  - Playing Venture draws some cards, increasing the rate at which you cycle, which is usually good.
  - Playing Venture has complicated filtering effects which you can safely ignore.  The only tangible impact it has on your draw pile is making it smaller.

[RiemannZetaJones]

This thread has brought to my attention, at least, and to others I suspect, the observation that if you play a card that draws into hand until some trigger is reached (Farming Village, Venture, Margrave etc) and does not return cards to the deck, and if you know that you are not going to deplete your deck to <5 cards by playing it, then it does not change the probability of drawing any of the possible next hands in your draw deck.

[carstimon]

This discussion came up again recently in the best $5 card list.  Because of this I wrote a small simulator which plays super basic big money (Prefer to buy province, then gold, then silver).  I also replaced Gold with "Venture3", a treasure which, when played, does exactly what venture does but ignores the treasure; it's always worth $3.  It costs $6.  I also made there be 32 provinces, so that any "cycling ability" of the Venture3 will be exaggerated.

I simulated 5000 games with buying gold, and buying venture3.  Playing with gold took 80.7062 turns on average, playing with Venture3 took 80.847.

Python code attached.  Sorry for the dead thread revival, but I wanted to post this and this seemed like the best place.

[ftl]

Oh, ok. Well, that settles it for me then, venture chains are only special when they chain a lot, not from individual ventures cycling a little, I was wrong before. 

[blueblimp]

This discussion came up again recently in the best $5 card list.  Because of this I wrote a small simulator which plays super basic big money (Prefer to buy province, then gold, then silver).  I also replaced Gold with "Venture3", a treasure which, when played, does exactly what venture does but ignores the treasure; it's always worth $3.  It costs $6.  I also made there be 32 provinces, so that any "cycling ability" of the Venture3 will be exaggerated.

I simulated 5000 games with buying gold, and buying venture3.  Playing with gold took 80.7062 turns on average, playing with Venture3 took 80.847.

Python code attached.  Sorry for the dead thread revival, but I wanted to post this and this seemed like the best place.

Nice experiment! I haven't looked at the code yet. My intuition is that 5000 games might not be enough for the difference to be statistically significant.

[WanderingWinder]

You should probably build up extra rather than going so province happy so early, if you're trying to drain 32

[carstimon]

Nice experiment! I haven't looked at the code yet. My intuition is that 5000 games might not be enough for the difference to be statistically significant.

Thanks.  I did it some more and got
venture3: 80.803
gold: 80.5682
venture3: 80.9496
gold: 80.7292
So using the made up "do it many times three times" statistical test it's definitely somewhere in that range.

You should probably build up extra rather than going so province happy so early, if you're trying to drain 32

Probably but I just needed something simple to compare.