Home | About | Tournament Winning Decklists | Forums

How many credits does it take to win a game?


There are multiple ways to look at this question, both from the Corp side and from the Runner side. From a corps perspective, while they possess much of the hidden information in the game, the amount they have to spend depends a lot on what the runner does, so for that reason I prefer to (at least for now) look at the question from the runner side. What’s the average cost to win a game? Given that you have to install breakers, run a certain number of times, and steal a minimum number of agendas. This will vary based on what the corporation does, how they protect their servers, and other considerations of course, but there would still be an average amount. So my questions are three-fold:

  1. @db0 Is it possible to configure OCTGN to track credits earned and credits spent? Ideally, it would be granular enough to track how much each icebreaker uses, but even a sum total for each game would be helpful.

  2. Failing that, would it be possible to have people volunteer their game logs to be parsed?

  3. Is it possible to theory craft this number, making reasonable assumptions?

For example, assuming I’m running fixed breakers at an average cost of $4 each:
Breakers: 3 breakers at $4 each = $12
Runs: 10 runs at $5 each = $50
Misc: 4 Hardware and Events at $3 each = $12
So maybe a theoretical credit cost to win a game anywhere from $74 (bare minimum) to infinity.


It’s too matchup dependent for there to be a meaningful average.

You could determine the average cost for Prepaid Kate vs NEH, for example but that’s going to vary greatly than even the same Kate vs RP Glacier.

Consider that against shell PE, you usually DON’T have to install breakers. If you are Noiseshop with Hades Shard, theoretically you don’t have to make any runs.


I disagree. I think it’s worth looking at. And you can stratify by ID within the OCTGN dataset.


I think @kiv is right on this one - the runner side varies so wildly depending on matchup. I’ve played many a game against PE where I probably spend about 12 credits total - conversely, Nasir Atman vs. HB glacier might easily spend 150+ and still not have won yet. These kind of numbers are so ridiculously disparate that an average between them is almost a meaningless value.

The Corp side is actually much easier to calculate: at the end of a game, add up rezzed card costs, install costs, and total advancements on agendas, then throw in costs of non-econ cards you played that are in the archives.

Regardless, I’m not sure what the value in this kind of info is: to determine your econ suite? In this case, do you account for recurring credits and discounts from IDs? how do you factor in an ID like Nasir or Whizzard who spend ID credits more because they can than because they needed them to win? If you’re looking for detailed, math based findings on runner Econs, definitely check out the QuantANR atricle on the main stimhack page; I don’t think the “total credits to win” number can really be said to exist, and even if it did, it wouldn’t help you much.


Plus traces, psi, cards that got cycled back w/ jackson/archived, blue sun bounces, etc…


The entire point of Netrunner is that it’s asymmetric and that different decks play differently, even within the same ID.

No offense intended, but your question is exactly as meaningless as determining the average tempo of all songs, or the average color of all paintings.


@kiv @voltorocks

You’re both entitled to your opinions. But I didn’t ask whether or not it you thought it would be meaningful or worthwhile doing. I’ll let my own not insubstantial experience be the judge of that. Rather, I asked if it were possible to acquire the means to answer the question.


While I can make OCTGN record credits gained and spent, people modifying the counters manually will throw it off. And then there’s the issue of taking back moves etc. It would be very unreliable data.


It might be more meaningful to track clicks instead of credits.


Appreciate the response. My field is statistics, not computer science, but is it possible to record the state of the tracked variables at the end of each turn? e.g. When F12 is pressed, it records whatever the credit counter is set at? That’s probably leaving out too much information. What about after each click?

Even if manual changes don’t get recorded, you might still be able to get signal through noise, given the large numbers of observations in the OCTGN dataset.


There is no averege click/credits in this game, no matter how hard you try to find it, or probably you find a number, but it is far away from reality.

Before asking someone else to do your work why don’t you start to track your game and come here with useful information? How many credits do you use to win ?


It’s a bit of work I’m not sure I have time for. Do leave a feature request in github and I’ll get to it if I can


I like this question. I think it would be worth seeing the dataset to see what’s there. I also agree with Madman that the dataset from OCTGN is probably large enough to present some value even with the problems within the data.

Can recurring credits be tracked in a similar way? Would they show up as “credits spent” but not “credits earned”?

Is there a way to see “free credits” from things like ABT, or oversight mechanics that allow free rezzing?

WRT the naysaying, I agree that a single number can’t possibly exist. But if a trend displays, like shapers spending upwards of 150 credits with regularity, and you build a shaper deck that doesn’t include unlimited sources like Kati/Magnum/etc, it might be worth bearing in mind.

I expect the value of this kind of information to be no different than the value of “rules” about how to play Hanabi. They create guidelines and a framework for best play but the key to excelling is in managing the situation that exists, not in abiding by the “rules” explicitly.


I’d be most interested in data that averaged the amount of credits required to beat Corp identities. It would be cool to see how much money (on average) you need to beat PE versus Blue Sun Glacier versus NEH Fastro, etc. It would also be cool to look at the correlations between Runner credits spent in a match and win percentage, and average credits spent in Runner wins across identities. I have a feeling that there is a sweet spot for credits spent; if you spend too much, you are probably running when you shouldn’t, and actually decreasing your chances of winning. There are all kinds of insights that could be gleaned from hardcore economic data like this, but I feel that it is unfeasible to ask @db0 to work on it. Maybe a Patreon goal for a big benchmark, like $300 or $400/month!


There’s the added problem that credits spent is correlated with game length. If I spent very few credits during a game, it was probably the game I hit 6 points off R&D with that turn 1 Maker’s Eye. If I spent a lot of credits, it might be that I was trying and failing to break through a rolling IT Department on turn 57.

Like Stock Mana Theory, I would imagine that the player with the most credits spent will probably win the game.


Nice one necroing this thread from half a year ago!



Whoops. Didn’t realise I clicked onto this from ‘Suggested Topics.’

I guess this is what getting flatlined by Cortex Lock feels like.


I missed this topic. This kind of statistic is really interesting to me, specifically click+draw and click+credit for runners. Why shoot down the idea? If there is data to analyze, at least we can say we looked at it and couldn’t correlate it with anything.

I’ve given a lot of thought to taking my own game logs to track my click behavior. You can take a lot of information from:

draw, draw, draw, draw,
advance, advance, advance, score.

Who can say what else we are missing?


Somehow I missed this thread as well. Thanks for clone chipping it!

I couldn’t disagree stronger with the nay-sayers here. I would agree that this information isn’t useful on its own given this high variance, but there are a number of interesting possible use-cases:

  • We can cluster the data and see what variable impact the credits required in a game. This could lead to an understanding, for example, of if this variable is mainly a function of the corp, or the runner. If it isn’t clear which, we’ll see which identities might flip the equation.
  • We can correlate the win/loss with credits expended, possibly after separating different corp/runner identities. It isn’t clear what will come of this.
  • We can correlate this data with output from simulations (e.g. QuantANR) to validate that they are reporting the proper values. Though we aren’t doing hard science here, validations from different perspectives is a required part of strong and correct scientific work.
  • We can better understand warning signs during deck construction for if an econ is insufficient/over-compensating. This will save testing time for people who aren’t deep in the meta. I live in the world of jank, and it is not trivial to understand how to tune an economy and a tempo (this is why I started QuantANR).

All of this will 1. give us an interesting framework with which to think about the game, and 2. possibly inform deck design. The act of modeling systems in science isn’t to out-do the experts that can make fine-grained decisions (which breakers are best in the current meta); it is to try and create a conceptual framework to think about the system from a higher level. One area this could be useful is to provide a set of idioms that new players can keep in mind that are “often correct” for their forays into deck building. This can increase the appeal of the game, and broaden our audience.

So @db0, @mtgred, if it is possible to get a data-dump, even if it is flawed, with credits/clicks/identities/win-condition/who won, we might be able to do something “interesting” with it. Even better would be the credit curve over the game, but that might be asking too much.

In short, if you aren’t interested in the data, that’s fine. I think that some of us can make good use of it. Experimentation and exploration of the unknown (even in the face of intuitive roadblocks) is essential for successful communities that don’t want to ossify.

That said, I’m the author of the QuantANR series, so I have an interest in getting a different perspective on this game we enjoy.