It’s worth pointing out that you’re talking about NEH as if it’s still a competitive deck that you need to plan for. You could very easily cut a TME or Indexing and have a better deck if you ask me. There are lots of cards that could be cut or moved around, and the deck is not perfect.
If you care so much about NEH, make cuts for a Clot and play 45. 46 will hinder you against NEH and every other deck in the field.
I’m using the old metagame as an illustration to show that 46 was correct back then. Doesn’t matter whether 46 is necessary in the post-clot meta since I just need to prove my point for one meta rather than all of them.
Even so, I think 46 may also be correct post-clot. Won’t divulge why until after the SSCI. 
As a free thought exercise, not constrained to netrunner in particular, but couldn’t one consider adding cards in a deck cause an increase in power level, not seeing diminishing returns until beyond the minimum?
Consider a 10 card ANR deck, where adding 5 cards increases power (same old thing, for account siphon, shutdown etc). Imagine the card quality is so good that 15->20 increases power level. At some point power increase outweighs inconsistency and finding cards you need, when you need them.
With tutoring and massive draw opportunities in netrunner, at what point does not running the minimum matter? What if there were a runner that was 20 minimum, would this be the best ID in the game if devoid of an actual printed ability?
I’m not arguing either side, I’m curious how it would shake out regarding what is the most optimum unconstrained deck size.
I think its pretty clear deck size matters less than we make it out.
Here’s to all the Silhouette and Chaos Theory decks that won worlds!
While I agree that deck size matters less than a lot of people make of it, two IDs is just small sample size and I doubt we can draw any meaningful conclusions there.
If you have a perfect read on the meta, then sure you can cut the right card.
Playing these silver bullets is cheap insurance that can give you a huge win boost in specific matchups - like bblum says the opportunity cost is generally a fractional click per card with the amount of draw power in this deck.
Aside from basically scooping to flatline decks, such an identity would be forced to play non-burst economy and therefore its power level is based on how good Opus and similar are in the meta.
I think clearly a 40 card Andy would be better than 45 card Andy. Silhouette’s ability is just garbage compared to Andy’s.
Shaper just has way more tutor and draw than Criminal has, minimizing the consistency loss of the extra cards.
I don’t think that the argument completely holds and I do still believe there’s a cut that’ll give you a higher winrate than no-cut. These cards are all useful, but there’s absolutely going to be one that’s the most useless.
There’s more to 46 cards than just a bonus click, as I pointed out, it gives 3-ofs a 1% lower chance to be seen in your opening hand which is actually quite a high and important amount. Just because you can’t immediately see a cut doesn’t mean that the deck wouldn’t be better with 45 cards.
It’s so bizarre, people obviously realise that Andromeda, CT and Silhouette have an upside and that Valencia has a downside, and you don’t see 41 or 51 card decks, but for some reason 46 is seen as acceptable? The change in consistency is actually lower the more cards that are already in the deck, so I mean why not 47? It’s obviously wrong, right? People do understand that… I think.
Andromeda only has 36 cards in her deck after opening hand. Yeah. Consistency matters. Andromeda won Worlds.
Those identities don’t plan on drawing 95% of their deck, levying, and drawing another 40% before the game ends. Plus the shaper tutoring kiv pointed out above.
I like 47 in regmax. Same arguments apply as for calim kate. The deck is better with 3x imp, 2x david, and 2x medium, but that package is too fat to fit in 45.
Have you ever played the Chaos Theory Doubles deck? That plays two Levy and like three Quality Time or something.
The only reasonable argument in book for including more is that you find yourself losing unavoidably because you ran out of cards on a somewhat regular basis. Everything else is horseshit.
This is really the thing. There is probably some optimal ratio of draws + tutors to power cards. I’ve often wondered if “draw” cards should be given a different “weight” in deck building. Though all cards in a given deck probably have a different weighting/power based on individual match-up.
I didn’t say the deck was better than Kate, but the deck certainly plans on at least one Levy and draws far more than Calimsha Kate does, and would be worsened by adding a card. That’s all we’re talking about here, whether the deck is better or worse with more cards, and the answer is always worse.
In short: the argument that states “it’s fine if you plan on drawing it all anyway” is stupid. I pointed out a counter example, and though the deck is obviously worse than prepaid Kate, that doesn’t mean it’s immune from any and all critical discussion.
Sorry I wasn’t making a point, I just thought of a funny image macro and had to make it.
THIS JUST IN: Scientists reveal unified theory of horseshit; all other renowned deckbuilders concede to flawless argument!
Funny thing is, this will actually come up more if the field glacierizes. I didn’t like >45 in the NEH meta but if everything is RP the extra cards often end up being pure value, (in maxx).
It’s always trash in Kate.
If i’ve to be fair, I’m gonna say that RP (and to a lesser extend, Minh’s PE) won Worlds, not Andromeda.
@Xenasis : CT Double isn’t a deck who try to be flexible. It’s exactly the opposite : you want to have a shittons of cash and win just by having more cash than what the corp can manage. With this kind of gameplan, it’s easy to see why having the bare minimum of cards is necessary. It’s a pro-active way to play and not a reactive one. Kate (or RegMaxx) on the opposing side isn’t playing the all in your face brute-force way to win the game. Both decks are flexible and reactive deck who adapt their own gameplan to what the corp is trying to do. With a reactive gameplan, you can more often than not just sit back, build your board-state / econ and draw for what you need. In this kind of scenario, having to draw one more card won’t really change what’s going on and won’t disrupt that much your gameplan because your gameplan isn’t as fixed as Andy when you (almost) don’t plan to draw at all when your rig is set-up or decks who need a specific cards to make his gameplan works (like Oracle May / Hostage in Double CT for instance).
For some reason, I feel that a lot of players under-evaluate clicking to draw and almost feel dirty to do it.
I like to build a shaper deck at 45 cards then go “shit i forgot to add diesel” and then add diesel for 48 cards.
But…but…muh click-efficiency 
For the record, there are worse things you can do than play 46 cards.
Like play 47.
I am going to start a pythagorean-style cult where we dogmatically worship the number 45.