What is your formula for your calculations? It doesn't match mine and I've done pretty thorough testing on my calculator. Here is what my calculator is showing for TLV.

These numbers all assume three copies of a card in the deck. In my sheet Target represents drawing that many or more copies of a given card. Mulli is the probability for a given target when mulliganing for that specific target. The numbers at the top of the percentiles represent the number of cards drawn. The Normal calculation is 3 copies of a card in a 60 card deck. The TLV calculation is 3 copies of a card in an 85 card deck. The percentiles are all cumulative probability of drawing.

You'll note that your cumulative numbers are significantly higher than mine are.

And why post here? I tried signing up for CardGameDB and it failed. It signed me up, but the account didn't activate. I never got an urge to try again.

mdc273 said:

I think our numbers are identical

Edit: And apparently that's a formatting nightmare. Anyway, our formula are identical (I only represent it as 1-hypgeomdist(0,drawtodate,3,85). The non cumulative forms are 1-hypgeomdist(0,draw,3,85-drawtodate))

The difference is you're showing the absolute chance, and in my graphs I'm showing it as relative efficiency (eg. 23%/32.5% = ~73%, and so on). There actually was a small math error on my non-cumulative graph (Figure 1), but the differences were about 2% and I submitted a new graph to replace it this morning.

I shouldn't have skipped math class methinks.

dcdennis said:

LoL. You should see my spreadsheet. It's kind of silly. I had to edit it at one point because I wasn't using formulas and actually having excel create thousands fictitious draw sets.

@Istaril - Cool. I would've never guessed it from the graph. Good to know I got my math right, lol.

mdc273 said:

Haha - ouch!

The word "Relative to 60 card deck" in the title and the label on the Y axis weren't enough? *grins*.

-Istaril said:

I dunno. I didn't mean anything against you. The text and the image just didn't mesh for me. I'm not sure what the relative means. Does one of those lines represent a TLV deck and one represents a 60 card deck? And to me, turn 0 is setup and turn 1 is the first plot. I couldn't really make heads or tails of it, but it's probably just me.

Believe me, I'm glad someone else has started to legitimze the math of the game. I've had this spreadsheet for like 2 years now I think and most of my meta have asked me about odds at one time or another. My sheet gets a little silly. I can actually calculate the value of drawing cards, what the odds of a given set up are, and all sorts of other stuff.

Like here is the chart the way I think of it:

Differet folks, different strokes or something, lol.

Fundamentally, we're answering different questions with the same dataset. You want to know what the chance of drawing a given card is, and I want to know how much less likely I am to draw a given card in an 85 deck compared to a 60 card deck - a comparison of efficiency. Each line on my graph represents how much worse your chance to draw a card (in an 85 card deck) is.

I did quite a lot of math-based tweaking of my worlds deck last year, and I find the math aspect quite rewarding. I suspect my spreadsheet is as much of a mess as yours, but there's quite a lot of information hidden in there for those that know how to look at the data.

Your definition of 0, 1 etc make sense. When you say "plot 1" do you mean "draw phase of round 1"? If so, we're treating it identically (the question being asked is how likely am I to have card X by the marshalling phase where I can play it), although for events it can be more relevant to ask whether you'll have it by the plot phase that round. Either way, I agree that setup would be better represented by 0 and not 1, and that's just an unfortunate choice on my graphs. I think I'll adopt that convention from now on.

-Istaril said:

I have spent so much time tweaking my decks based on the math it's silly… Shockingly I don't do it for Netrunner. I have no idea why it's so different.

For the numbers, this is what I think of it as:

Turn 0 - Setup, before drawing back up. (so 7 cards total)

Turn 1 - Post draw phase (so assuming 4 card flop, this is a total of 13 cards in a normal deck without any draw)

Turn 2 - Post draw phase (15 cards now)

Yadda, yadda, yadda.

I could quickly do an event one, but ugh do I hate it. You have to assume you'll be able to hold onto the event through a whole challenges phase. There's enough randomness that I think it makes knowing the stats a negative. You might know you have a 95% chance of getting a battle. You might build a deck knowing you can get 1 battle every turn only to have every battle knocked out during challenges phase every time you play it. I always wanted to try a Battle deck and then realized how fragile it would be even if I COULD consistently draw the **** events.