Goldfish Draft 2: Final Score

I worked through one sequence of play for the previous goldfish draft, giving a lower bound to the possible score.  I'm not fully satisfied with it, but it has become too cumbersome to work through the possibilities.

The basic idea is to get the [mtgcard Glint-Eye Nephilim] down, give it double strike with the [mtgcard Silverblade Paladin], use [mtgcard Gisela, Blade of Goldnight] and [mtgcard Anthem of Rakdos] to increase the damage, use the card draw from the first attack to pump the [mtgcard Glint-Eye Nephilim|Nephilim] for the second attack and [mtgcard Niv-Mizzet, the Firemind] for damage.  Throughout, [mtgcard Elbrus, the Binding Blade|Withengar] keeps growing, and most the extra cards get turned into land to fuel the card dump, and also to allow [mtgcard Ghave, Guru of Spores] to shuffle counters around, allowing the [mtgcard Goblin Sharpshooter] to do damage.  Finish off with [mtgcard Soul's Fire] and [mtgcard Sword of the Ages] for further damage.

The resulting score was 4,139,209,534 (details are available in this spreadsheet); this looked like it would be the best of the group for some time until one of the others found an alternative line of play which bumped his score up to 27 billion, a target which seems sufficiently far ahead that I'm not motivated to find a better line of play.  There are some obvious areas for potential improvement, however; starting a bit slower might allow [mtgcard Elbrus, the Binding Blade|Withengar] to grow earlier.  It should also let [mtgcard Chandra Ablaze] use her ability in the last round to re-cast [mtgcard Soul's Fire] and [mtgcard Titanic Ultimatum], with the latter also opening up the possibility of cycling a lot of damage through the [mtgcard Crypt Rats].  Certainly options worth considering, and I suspect that they would lead to an overall increase of score... but I'm rather doubtful that it could reach the required 27 billion mark.

Goldfish Draft 2

I participated in a Goldfish Draft this week.  This is actually the second that I was part of, but the other one had nothing much of interest as far as my results went.  One of the other participants, though, managed a comfortable win thanks to a serious combination that ended up in him gaining over 10¹⁹³ extra turns!

(I know I haven't made the post about the previous Goldfish Draft that I said I would.  I got bogged down in too many options, and frankly it looks unlikely that I will ever get back to it.  Oh, well.)

The draft went passably well, but the results never quite managed to hit high gear.  There's some nice combination potential but the key cards to make it all come together (a [mtgcard Stolen Identity] would have done very nicely) never quite materialised.  In retrospect there were a couple of poor drafting decisions also, although not that many.

Here is what I ended up with; first the A deck:

[mtgcard_image Fastbond] [mtgcard_image Glint-Eye Nephilim] [mtgcard_image Gisela, Blade of Goldnight]
[mtgcard_image Anthem of Rakdos] [mtgcard_image Chandra Ablaze] [mtgcard_image Elbrus, the Binding Blade]
[mtgcard_image Liege of the Tangle] [mtgcard_image Niv-Mizzet, the Firemind] [mtgcard_image Timberwatch Elf]
[mtgcard_image Liliana of the Dark Realms] [mtgcard_image Borborygmos] [mtgcard_image Titanic Ultimatum]
[mtgcard_image Soul's Fire] [mtgcard_image Norin the Wary] [mtgcard_image Goblin Chieftain]

Quite a lot of gold in that selection!  And this was the B deck:

[mtgcard_image Wheel of Fortune] [mtgcard_image Silverblade Paladin] [mtgcard_image Black Lotus]
[mtgcard_image Crypt Ghast] [mtgcard_image Sword of the Ages] [mtgcard_image Szadek, Lord of Secrets]
[mtgcard_image Extraplanar Lens] [mtgcard_image Ghave, Guru of Spores] [mtgcard_image Goblin Sharpshooter]
[mtgcard_image Underground Sea] [mtgcard_image Bayou] [mtgcard_image Goblin Bombardment]
[mtgcard_image Sarkhan the Mad] [mtgcard_image Searing Meditation] [mtgcard_image Crypt Rats]

I'm still working out the play sequence; I believe I'll exceed a million, and am hopeful of getting past the billion mark.  I've no idea how much higher, if at all, I can go.

Risk

I played my first game of Risk today.  Well, it was actually some computerised knockoff, but as far as I can tell the rules were the same and the map almost identical (although the names differed).  It's never been a game I really looked at before; I'm mostly uninterested in multiplayer (i.e., more than two player) wargames as a result of early experiences with Diplomacy, and one even worse one with Machiavelli that would have been the last time I played games of that kind.  I'm not temperamentally well-suited to games where breaking alliances is a key part of the play.

(I also have issues with games where much of the play lies in convincing players to adopt the courses of action that you wish them to.  I'm all for setting up a situation through gameplay that shapes the desirability of their options, but it ceases to be fun for me when such facets are overshadowed by the persuasiveness of others (or myself).  I enjoy tactics and strategy, but not negotiation and persuasion.)

This was a three-person game, and fortunately (from my point of view) there was no real attempt to do much negotiation.  We had a random start, each claimed a continent (more or less), then fought out the rest.  I was fortunate enough to win as a result of a couple of things, in particular being able to claim two continents early on with a very small combined border (this was where the map differed from the official one, and I think it is a flaw in the version I played).  The other part was an overly cautious player adjacent to one side of that, enabling me to get away with glass cannon tactics far longer than I should have.

I don't want to comment on the quality of the game so much -- although obviously it plays fairly well -- but rather on the difference that I feel the computerised version made.  None of us had played it before (whether physically or on computer) so we were all on even status there.  We each got caught out by some interface issues, not understanding the attack rules and troop transfer mechanism at first.  If we have been playing a boardgame as such we could have simply worked out what was going on and resolved it, but since here the computer was in control we had to live with the consequences of these errors.  This may well have been ultimately significant, as that player I described as being overly cautious was, I believe, caught out by this several times, finishing their move early as a result of selecting troop movement.

But the difference that struck me afterwards -- and maybe I'm wrong about this, but it feels right -- is how little real investment there was in the results of battles.  It was pretty much just "click-click-click", with win or loss following.  Whereas if we were clustered around the board, dice clutched tight as we tried to infuse them with winning energy, making each individual roll... I think it would have been a lot more engaging.  By streamlining that process away the game was quicker and easier to play... but I think it also lost a good deal of player involvement as a result.  There was still the overall board position to consider and play for, but the narrative of the battles was lost.

I'm not sure that anything could be done about this, mind you.  But it feels like something to keep in mind when producing a computerised version of a board game: What aspects of the experience are being lost, and if they are important, is their some way to provide a similar experience?

Goldfish Draft: Tinkering

In my previous post I gave a possible score that I might have had if presented with the same choices as Andrew in his Golfish Draft (as detailed in his posts starting here).  This post follows up on some issues related to that; namely: What was I alluding to with the [mtgcard Vesuvan Doppelganger] shenanigans, and how would things have gone if I had drafted the key card in the deck (the [mtgcard Yew Spirit]) after all?

There's a further question of what would be the best options with the benefit of hindsight (or prescience, if one prefers).  I certainly can't pretend to have a definitive answer to that, but I will give one possible result in the next post.  As a teaser, I invite you to consider the opening play of [mtgcard Black Lotus], [mtgcard Show and Tell], [mtgcard Omniscience], [mtgcard Serra Avatar], and [mtgcard Garruk, Primal Hunter].  Sure it's a five-card combination, but it pretty much allows you to play 23 of your other 25 cards at once.  That may or may not be optimal, but it's certainly a powerful way to start.

As for the other matters... not choosing the [mtgcard Yew Spirit] was an extremely poor decision on my part (one could argue that there were several others, but that was the worst of them).  It was so broken if it could be made to work that it was not sensible to ignore it that second time, particularly given that I did not expect to make effective use of the [mtgcard Stream of Life] that I chose instead.  Anyway, with that in hand the play would essentially go as before except for the final turn; this time [mtgcard Ajani, Caller of the Pride] would summon the lions between the two combat damage stages (if you want to be technical about it, after the first damage resolution and associated affects and before the second combat damage phase begins) and then the [mtgcard Yew Spirit] would be pumped as much as possible.  I make the end result somewhere in the vicinity of 10^10^30.8, or just 30.8 in Andrew's revised scoring metric.

Now for the [mtgcard Vesuvan Doppelganger|Vesuvan Doppelgangers]: What I had been thinking earlier was that rather than having just copies of [mtgcard Roiling Horror] around, it might be feasible to use [mtgcard Trostani, Selesnya's Voice] during upkeep, then switch a [mtgcard Vesuvan Doppelganger|Doppelganger] to copying [mtgcard Trostani, Selesnya's Voice|Trostani] instead.  That new copy could then also create a token creature, and so it would go; each time the life total more than doubles, at the expense of being able to attack with one less creature that turn.  That turns out to be definitely worth it provided there are always at least three [mtgcard Roiling Horror]s that will attack, and there are more than that in this case.

That first plan ran into the stumbling block that [mtgcard Trostani, Selesnya's Voice|Trostani] is legendary.  But even then I thought it was still worthwhile acting under the new "Legend Rule" -- we could choose to get rid of the original and so still use the new one, and the doubling machine runs happily along.  The end creature count is the same so the difference is simply between front-loading the doubles or attacking with more creatures.  The doubling is always better, and the limiting factor in how well this works is the availability of white mana for the token creation.  It's a lovely generation engine, really, as the [mtgcard Stolen Identity] keeps us supplied with both tokens and mana (thanks to [mtgcard Gaea's Cradle]) so that we can keep pace with the other factors.  In this case I think it ended up being around 30 extra doubles, which would probably push the end result over 10^10^36.

However, that new Legend Rule only comes into play starting July 13, 2013.  This draft, and indeed this post, happened before then so this cannot work as envisaged.  It is possible that a variation of it still can -- if two [mtgcard Vesuvan Doppelganger|Doppelgangers] both copy [mtgcard Trostani, Selesnya's Voice|Trostani] then the first to resolve will trigger the legend auto-destruct on both.  What I don't know is whether the second resolution then fizzles as the creature to be copied is no longer present, or whether the creature data is "already set".  I suspect that it ends up fizzling based on similar rulings, but I'm not sure.  I also haven't worked out whether this is worth it; it's a process that loses creatures, unlike the other version, so I imagine it is not.

Anyway, that's what I was thinking, and come mid-July that engine will work quite nicely.  But not yet.

Update: An email exchange with one of the other players in the draft drew my attention to the fact that the [mtgcard Stolen Identity] is cast when the ciphered copy's ability triggers.  That means that it can interact profitably with the [mtgcard Ink-Treader Nephilim], causing exponential growth in the number of copied creatures rather than linear.  We lose the initial boost from the [mtgcard Rite of Replication] (which I had chosen instead of the [mtgcard Ink-Treader Nephilim]), but it is plausible that this is quickly overcome.

... Well, not so quickly, but inevitably.  The slowdown is actually the loss of [mtgcard Trostani, Selesnya's Voice|Trostani] -- since it is legendary, the first time the massive copying happens it disappears.  That removes some copies, but far more importantly it also removes the life gain each time a creature appears.  Still, exponential growth must win out given enough turns; in this case I make the resulting score at least 1.6*10^45, or with the [mtgcard Yew Spirit] somewhere over 10^10^40 (I've not thought about this carefully, but that feels plausible as a lower bound).

Goldfish Draft: Final score

(For context, see the previous post.)

If you've read Andrew's series of posts about the Goldfish Draft, you'll have noticed that I stopped computing my score at the end.  In part this was because it became irrelevant, but also because it was getting somewhat cumbersome to insert into a comment section.  So here's the details in this post.

Magic the Gathering: Goldfish Draft

Back in the day I used to be a pretty obsessive Magic: The Gathering player.  I first heard about it through baffling snippets posted on rec.games.board (for those of you who still remember Usenet; ah, memories...) that were tantalising in how they hinted at a rich set of unfamiliar concepts and gameplay waiting to be explored.  I was fortunate to wander into a local games shop on what they said was the first day of sale for Unlimited Edition packs in Sydney and soon roped a friend into playing some games with me.  I was, inevitably, hooked (as was he) and much damage was done to my bank account thereby.

Looking back at the release dates I can only have been playing for a couple of years; Fallen Empires was the last expansion I acquired and I never even really looked at it much; I'm not sure I even got around to separating out the play cards from the unneeded extras.  The next expansion after that was Homelands (released October 1995; Unlimited Edition was released in December 1993) but I had mostly stopped playing by then.  That coincides well enough with my starting full time work and so having much less time, but I think I'd already started to drift away from it -- the friends at university that I used to play with had mostly either left or were leaving that year, and the ongoing cost for each new expansion was troublesome.

Another likely contributing factor is that the previous two expansions at the time (The Dark and Fallen Empires) were both rather boring in their way; I see that Wikipedia's entries on them suggests that they are both regarded as rather weak expansions.  Certainly the earlier Legends expansion was much better on all fronts than both, except for the then-infamous A/B box issue.  (Due to poor collation at the factory, the uncommon cards were implicitly split into two groups in such a way that boosters would only ever have uncommons from the same group.  Worse, all boosters in a box were of the same type, and it seems that all boxes in the same... crate?  I don't know what the next step up was... were also of the same type.  So a store that ordered ten boxes (say) would very likely get all boosters of the same type.)  I think trying to overcome this distribution issue may have been my first foray into online purchasing items from overseas.

Aaaaaanyway, I have friends who play the game still (or rather, picked up the game well after I left, but do play now), and I do hear occasional snippets about it from various sources.  (Like doing away with mana burn; I swear, kids these days...)  One of those friends is David Morgan-Mar, who does a great many things (including many great things).  David recently posted about a "Goldfish Draft" that his circle of friends had engaged in, and it piqued my interest.

In particular, David linked to Andrew Shellshear's series of posts about the draft, where Andrew was kind enough to go through the draft choices he was confronted with turn-by-turn.  It was interesting to "play along" and see what choices I would have made instead; they mostly turned out pretty well, perhaps undeservedly so.  I commented on the various articles there, but the comment field format was getting a bit constricting towards the end.  Hence this blog, so that I can more easily address a couple of points in upcoming posts.  In particular:

• How my final draft choices worked out; addressed in this post.
• How a very small adjustment could have made a very large difference to my result; addressed in this post.
• With the benefit of hindsight, what is the best we can do?  I don't have a certain answer for that, but I like to think I'm doing pretty well.

In the unlikely event that someone ends up reading this then I encourage you to read through Andrew's posts and play along with us.  There's some hard decisions to be made, but it's also quite enjoyable when things start to come together and a decent line of play can be developed.

About this blog

Every so often it's convenient to have a place to air one's thoughts about something; this blog aims to serve that function for me on the rare occasions that it is wanted.  Mostly this is for situations where a comment field on someone else's blog does not feel like the best format for a response.

I have another blog dedicated to the sadly missed SBS TV show Letters and Numbers.  Although the show has been cancelled for almost a year (as of writing this post) I have been generating new "episodes" of the show for people to play.  If that sounds idly entertaining then please drop on by.