Let’s talk strategy as the corporation. There are two ways to win as a corporation: grind agendas, or kill the runner. Depending on your choice of corp, you’ll have an easier or harder time achieving these paths to victory. First I’ll talk about general strategic principles, then I’ll discuss tactics for scoring agendas, then I’ll talk about tactics for killing the runner, and wrap up with some general thoughts on card use and tournament psychology.

Corporation strengths and weaknesses

Jinteki

Jinteki has one of the better assassination games among the corps. Its core set identity, combined with its traps and damaging ice, can set up a kill with a little preparation. Its expansion identity is good as well, but plays better with an attrition game plan.

Weyland

Weyland has a very good assassination vector due to the scorched earth doupletap. Some of the ice it has access to can be strengthened without using cards, which can be advantageous - it’s usually better for the corp NOT to draw cards to minimize the risk of getting stuck with too many agendas in hand. Weyland also has access to a fairly strong economy - Beanstalk royalties are in-faction, and hostile takeover can net fast cash.

Weyland has 0 traps in-faction, so it’s almost essential to get them from other corps.

NBN

I consider NBN one of the stronger factions. In my opinion it has some of the best ice in the game, because there are often consequences whether you break the ice or not. NBN excels at playing the attrition game - it has multiple ways to tag the runner, and can seriously mess with the runner’s resources. Free resources during traces is very powerful with the expansion cards, many of which use the trace mechanic.

NBN’s ability to win through flat-lining the runner is somewhat questionable, unless you spend all of your influence bringing in cards from other corps.

Haas-Bioroid

Haas is a bit odd for a corp, in that it gets stronger late game. Early on, its proprietary ice can be bypassed without the use of an icebreaker, which makes trying to score early points almost impossible. It’s only when you stack up enough ice that the runner MUST break your ice that the game turns in your favor. To compensate for a bad early game, Bioroid ice generally has a larger strength, and the consequences of not breaking their routines are extremely unpleasant.


Runner cards to be aware of:

Criminal

Inside Job - this card can ruin your best laid plans. Early game, you must live by the rule, “my server is not safe until I have two pieces of ice.” Just repeat that to yourself a few times so it sticks. In the same vein, it might be best not to put your most powerful piece of ice on the outside. You might consider playing your high cost ice before you can actually pay for it just to dodge inside job, but beware card #2:

Forged activation orders: This card pretty much says “don’t lay it unless you can play it.” It also means that you should think carefully before auto-rezzing a card in front of your hand or deck if the runner makes a run with clicks left. If you rez one piece of ice, you may be making another piece of ice protecting a more valuable server vulnerable. Don’t lose your Tollbooth / Wall of Thorns in a remote server because you wanted to rez something small in front of your deck or hand. Remember: running your hand/ deck is a gamble. Running a server with an agenda in it is not.

Account Siphon - You will hate this card, and it’s just another reason to protect your hand more than you otherwise would. The more the runner has to pay to get into your hand, the less valuable Account siphon becomes. You probably still won’t completely negate its effects, but it’s something to keep in mind.

Shaper

Tinkering - there’s really nothing you can do to counter this card, which is why it costs 4 influence to include in other decks. You need to be aware of this card because it means that the runner may be able to hit a server protected by 1 piece of ice even though their icebreaker doesn’t match.

The maker’s eye - This card makes it more likely that the runner will go for your deck - not a bad idea to lay down a little extra protection, just in case.

Other than those two cards, the shaper’s deck is pretty conventional.

Anarch

Liberated account - this card must die. As Weyland and NBN, you have agendas that can tag the runner during your turn. It can sometimes be useful to hold off on scoring these agendas so that you can score them and trash resources, including this card.

Wyldside - This card lets the runner find the cards they need quickly, can makes the runner more resilient to damage, both of which are useful.

Stimhack - this is a late game trump card that can throw off all your calculations. If you’re playing Jinteki or Weyland, it could also mean near instant death for the runner, so it’s risky, but it can also be worth it. If you’re winning in resources, but the runner is close to winning through points, it can be a good to play around this card. It’s much better to make the runner take his chances with your hand than to put the winning agenda in a server that the runner could access if only he had +9 creds.

Medium - the inclusion of this card makes protecting your deck a priority. If the runner is able to run your deck for cheap, he can gather resources, then run your deck 4 times for increasing amounts of cards. This only gets worse if he’s able to use demolition run before the last run, netting him a large chunk of cards accessed from your deck. It’s entirely possible for a runner to just flat-out win on a good turn with medium.

Strategic Concerns

Game flow

Generally speaking, the corporation is most powerful at the beginning of the game, and the runner has an advantage late game. This happens for a number of reasons:
1) The runner starts with no icebreakers, and must find and play specific icebreakers in order to get through. Even if the runner starts with a generic icebreaker (Crypsis), he may not have the resources to use it until you’ve already scored your first agenda.
2) It costs less in the short run to lay ice than it does to break ice. Icebreakers become more cost efficient over time, particularly when combo’d with cards that reduce the effectiveness of ice, but early on they’re more of an investment.
3) It takes the runner time to gather resources to break ice, whether through cards or money. Conversely, playing and advancing an agenda is cheap.

This means that you should act fast, but with reservations. Keep in mind who you’re playing and what they could do with the cards they’re likely to include. Don’t try to rush an agenda vs. a criminal by laying out one piece of ice. Maybe place some initial ice on your deck or hand before going all-in on a remote server. You can pretty much guarantee that the runner will run your hand or deck if you leave either unprotected (which is also something to keep in mind when evaluating your opening hand).

When to play traps vs. agendas

Generally speaking, you should play and score agendas when the runner:
1) doesn’t have a matching icebreaker (but be careful of the runner’s cards); or
2) doesn’t have the resources to break your ice

You should play traps and decoys when the runner
1) has matching icebreakers; and
2) has the resources to bypass your ice

Play this pattern until your opponent understands that this is how you’re playing. If the runner starts leaving advanced cards alone when he has the resources to run your remote servers, THEN you can switch it up and play real agendas even though the runner could steal them. However, you should not do this until you have established a pattern of play with your opponent. It’s better to let the runner take his chances with your hand or deck then to lay out an agenda you know he can steal. Keep drawing cards and laying ice until the runner is low on resources.

Agenda Scoring Tactics

I’m going to discuss four different tactics:

1) Decoys
2) Traps
3) Rushing
4) Bluffing

These tactics all assume that the runner has sufficient resources to break through your ice. If the runner is low on resources and can’t break through to a legitimate agenda, you may as well play it and score it without resorting to any of these tactics, assuming it’s otherwise safe to do so (see “Runner cards to be aware of,” above).

Decoys

A decoy is a card placed in a remote server that is not an agenda. You may use an asset or an upgrade for this purpose. You should play a decoy in a remote server that is protected by ice, preferably ice that will cost the runner resources to break, or that has consequences upon being approached (Data Raven and Tollbooth are very good for this purpose).

The purpose of a decoy is to cost the runner time and resources running a server that does not contain an agenda. Ideally, this will sap the runner’s resources enough that you may play an actual agenda on your next turn, advance it, and then score it on a subsequent turn after the runner fails to steal it due to lack of resources.

Decoys are sometimes not as effective against runners as a trap because placing a card in a remote server and then not advancing it may signal to the runner that you have placed a card that can’t be advanced.

Traps

A trap is a card that can be placed in a remote server and has negative consequences when accessed by the runner. Like decoys, the purpose of a trap is to sap the runner’s time and money, for the purpose of setting up conditions to score an agenda. Traps can sometimes be fatal in their own right (an advanced Project Junebug, for example), or can set up combinations of cards for the following turn (ghost branch into Psychographics).

Like decoys, traps should be placed in servers protected by ice. The advantage of a trap compared to a decoy is that traps may be advanced, and will therefore look exactly like an agenda in the eyes of the runner. You should always advance a trap the turn you play it, at least once. Advancing a trap signals to the runner that the trap is a card that can be advanced, and traps sometimes have no effect unless advanced. The runner will likely make a run on an advanced card in a remote server, out of fear of you scoring an agenda.

Traps MAY be placed in unprotected servers before advancing them. A runner may decide to run them anyway, particularly if the runner is already ahead - unless the trap has the potential to kill them outright, they don’t lose as much by running it. Generally speaking, it’s better to get the runner to waste money breaking your ice on an unproductive run then it is to just have them spring the trap in an unprotected server. On the other hand, it can be useful to establish that you’re willing to place a trap in an unprotected server for the purpose of setting up a bluff later in the game.

A mistake that I sometimes see people make I their decks as the corporation is that they take out all of their traps. This is a mistake. Traps are an essential part of your strategy as a corporation because it allows you to more easily fool your opponent into making a disadvantageous run. Without traps, the game becomes a simple cost evaluation - “does the runner have enough resources to break through my ice? Y/N” Often the answer to this question, by the time it matters, will be yes.

Rushing

Rushing is playing an agenda under circumstances where you can score that agenda during the same turn. This is possible with agendas that cost 2 advancement counters to score (such as Posted Bounty or Breaking News), or with agendas that cost 3 advancement counters to score, in conjunction with Biotic Labor. While rushing can give you quick points, in many circumstances it is not the best option. Posted Bounty, for example, can be used along with scorched earth to double tap a runner for the win. Similarly, Breaking News can be used with Closed Accounts or Psychographics to set up more powerful turns than simply scoring an agenda. However, rushing an agenda can be beneficial if you need to clear agendas from your hand, or only need a few points to win.

Bluffing

Bluffing is placing an agenda (a real one) in an unprotected server. Conventional wisdom would suggest that no one would play a real agenda in an unprotected server, particularly when you have other servers that are protected by ice. Bluffing takes advantage of this belief. Bluffing is one of the riskier tactics, because it relies on the runner believing that either 1) you’re capable of placing a trap in an unprotected server, or 2) the consequences of running into a trap are worse than letting you score. If the runner CANNOT let you score (if you have 5 of 6 points already), or the runner is ahead in resources and points, bluffing may not be as effective.

Ideally, you want to give the runner only one turn to think about whether they want to run your bluff or not. Whatever agenda you play, you should be able to play it one turn and advance it to score on your next. It’s not a bad idea to set up this play by placing a trap into an unprotected server early in the game and advancing it twice. The runner will hopefully remember that play and believe that you are trying to do it again.

Bluffing is riskier when you have not established with your opponent that you’re capable of bluffing in this way, or if your opponent is looking only at the game state and not thinking about what you could have played from your hand. If your opponent routinely assumes that everything you play in a remote server is an agenda and will run it anyway, you should be playing traps and decoys until you can adequately protect a server.

Bluffing is also more effective when you play project Junebug. Assuming that the runner leaves your unprotected card alone for a turn, and you advance an agenda three times as your entire next turn, the runner will often not be willing to run that card, even though it’s unprotected, and even if it would win the game for him. A Project Junebug with three advancement counters will kill a runner outright - it’s just too risky for him to go in blind. You should only try this if you’ve already established with your opponent that you’re playing Project Junebug and that you’re willing to lay traps unprotected.

Killing the runner

There are a couple of ways to kill the runner, but the two main ways are scorched earth doubletap and Project June Bug.

1) Scorched Earth Doubletap

The strategy here is to tag the runner and then play two scorched earth cards in one turn. This does 8 total damage, which will kill the runner unless they have a Plascrete Carapace or similar card. Weyland is probably the best corp to run this strategy: Scorched Earth costs 4 influence, so it can be hard to fit into other decks. NBN can use scorched earth as well - it has ways of tagging a runner during the corp’s turn, and can otherwise support the strategy.

The first thing you need to do is draw two copies of scorched earth. If you’re using this strategy as a your main goal, you should try to include the Anonymous Tip card in your deck - it will help you draw what you need to win. Don’t be too concerned about placing agenda cards in your discard pile if you have to on your way to drawing the two scorched earths. If you need to unload cards, you can always lay ice on top of your discard pile as additional protection. Once you assemble your combo, it won’t matter how many points the runner has anyway.

The second thing you need to do is have the runner tagged on your turn with at least 6 credits and 2 clicks left. This can be accomplished by scoring and forfeiting Posted Bounty, if you’re Weyland, or Breaking News, if you’re NBN. Sea Source can also be used if the runner made a successful run, but that will require additional resources to run the trace. Some ice will also tag the runner, but the runner will only keep the tag if he made the run as his last click, which may not happen very often. If the runner is going for your hand, it is possible to trick him by not rezzing, say, a data raven, draco, or hunter on a first run, but then rezzing it if he makes his last run. You should particularly consider this tactic if you don’t have any agendas in hand (you might even say something sneaky, like “aw, I can’t rez that now”).

Note: private security force

Private security force will probably not win you the game. Unless the runner is being very unwise about his number of cards in hand, you just won’t be able to kill him. Even if the runner is tagged and doesn’t have enough resources to get rid of the tag, the runner will still be able to take one extra action per turn to get credits and eventually ditch the tag. Simply put, there’s no way to “lock the runner out of the game” with private security force. If the situation does arise, it’s not a terrible idea to take advantage of it. The runner will be forced to discard potentially useful cards, and if you have a pad campaign, you’ll be passively generating income while the runner is not. On the other hand, private security force pairs well with the snare card from Jinteki - if the runner makes an unwise run with his last click, it's possible you can win outright if you have previously scored private security force.

2) Project Junebug
An advanced project Junebug will kill a runner outright. Possibly the best way to use Project Junebug is outside of Jinteki, where the runner may not expect it. It only costs 1 influence, which is a steal for a card that can win the game for you.

In order to win with project June bug, the runner must believe that it’s an agenda. The more points have already been scored in the game, the more likely the runner will go for it. If the runner must make a run or risk losing the game, he has a tough choice to make. Most people would rather make a run and lose to a trap than not make a run and lose to a scored agenda when they could have stolen it.

In order to get the kill, you will need Project Jungebug to have at least 3 advancement counters on it. One tactic is to play project Junebug in a server when the runner can’t access it, and leave it with no advancement counters. Once the runner has had a turn to gather more resources/ play icebreakers, you can take an entire turn to advance it three times. The runner may assume that you’re trying to rush out an agenda now that he can break through your ice, and make a run. This works best when you install Junebug on one turn and then advance it on the next. If you leave it out there for longer, the runner may wonder why you didn’t advance it before when you had the chance.

Card Use

Ice that traces

Some ice contain subroutines that use the trace mechanic. This is often useful because you can sap the runner’s money if they want to avoid tags or other consequences. However, you need to be careful when playing cards like these. On one hand, they can be extremely powerful, particularly when you’re ahead on resources. On the other hand, they can be broken without any icebreaker by letting the subroutine happen and then paying for the trace.

For example - when you rez Draco, you need to keep in mind that draco’s subroutine doesn’t say “give the runner a tag and end the run.” It says “trace2: if successful, give the runner a tag and end the run.” So you can’t just lay out all your money and say “you can’t break draco, so I win.” If you just used all your money, the runner can just pay 2 credits and the trace fails.

A Note on Tournament Psychology

Tournaments are strange play environments in that you’re only going to be playing any given opponent one time as the corp and runner. As such, it can be very difficult to feel out the level of play that your opponent is using. If your opponent is only evaluating the board state, he will likely run any advanced cards in remote servers - traps and decoys will be very effective against this kind of opponent. On the other hand, if your opponent is thinking beyond the board state, then you will be able to take advantage of his awareness of “conventional wisdom.” Here are some important indicators that you can use to evaluate your opponent’s play:

1) If you played a trap in a remote server and the runner didn’t go for it, your opponent is thinking beyond the immediate board state, and you may need to alter your strategy in order to fool him into making an unproductive run. Your opponent is thinking “no one would play an agenda that he knew I could steal” - and therein lies the trick. Take advantage of that line of thinking. Cards that allow you to advance beyond 3-4 will be effective against this runner because he may let them build up, assuming that they’re a trap.
2) if the runner makes a run on a remote server that contains an agenda, after the runner breaks your ice, put your hand on the card and say “do you want to see what this card is, or would you like to jack out?” Watch your opponent’s reaction. Tournament players are generally sticklers for rules. Your opponent may consider “why is my opponent asking me this? Is it important that I be given a choice before revealing the card? Is he asking me this because I have a chance to jack out before springing a trap?” If the runner does in fact jack out, you have saved an agenda and learned that your opponent is thinking beyond the immediate board state. If the runner does not jack out, you may or may not have learned anything about your opponent - he could be ignoring your words, assuming that they’re intended to put him off, or he could be making an informed decision about your choice in playing an agenda. There’s really no way to tell without more information.
3) Your opponent used infiltration to reveal a card in a remote server. In this case, your opponent is at least thinking that you could have a trap. You may need to set things up carefully to fool him, but there’s also a chance that your opponent will simply evaluate the board state if he can’t reveal your cards.

Note that if you’re the corp first, and the runner is being cagey about when to run, he’s most likely the kind of player that will use traps, so you should be equally wary as the runner. On the other hand, if your opponent is simply evaluating the board state, he may not run traps in his corp deck, or may play them in an odd manner. As the runner in that situation, the corp player may fall victim to surprise cards from your hand. Holding on to an icebreaker until you can play it and use it in the same turn may catch the corp player off guard, allowing you to score an agenda that he didn’t expect. Likewise, the corp player will probably get blown out by inside job and tinkering.

I read the whole post, bookmarked it to read again.

Thanks for the effort, well done. Good reminders here.

Good job. :-)

On bluffing: Another kind of bluff in the game is the installation of ICE that the Corp can't actually rez with its current resources. This mostly happens during the early game when the Corp is still setting up. If you have undefended or weakly defended servers and you only have ICE you can't afford to rez in your hand, put them down anyway. The Runner will have to guess if the Corp can rez it or not; he may think that you can turn it on and that it's actually something nasty, thus choosing not to run. This deterrance can buy you time to raise credits or score an agenda. This is better than not putting down said ICE and explicitly telling the Runner you can't defend those servers.

Very well done. I learned quite a bit and I've been playing this game since the first time around in the 1990s, LOL!

Hi Brigaldio,

Super cool post, really enjoyed reading it, thank you. First time posting here - just wanted to ask one quick thing - I just started playing Netrunner recently and wanted to get the rules right.

So in the rules, it says that the corp can boost your bid for trace strength with any number of credits. Then the runner gets a chance to outbid you to increase their link strength and outbid the corp. If he/she doesn't, then the corp successfully puts a tag on he runner.

In the part where you say this:

For example - when you rez Draco, you need to keep in mind that draco’s subroutine doesn’t say “give the runner a tag and end the run.” It says “trace2: if successful, give the runner a tag and end the run.” So you can’t just lay out all your money and say “you can’t break draco, so I win.” If you just used all your money, the runner can just pay 2 credits and the trace fails.

I'm not sure what the meaning of the last statement is: "if you just used all your money the runner can just pay 2 credits and the trace fails"

From my understanding of the rules - the runner can pay 2 credits and one click as an action to remove the trace. however, that wouldn't stop the trace during the run to be successful, which includes putting a tag on the runner and also ending the run. The runner can then use their next action/click to pay 2 credits and remove the tag, but they can't do it in the middle of the run.

Is that accurate?

Thanks

You seem to be conflating Tags and Traces. They are often related, but are two distinct things.

A Trace is a game mechanic, which happens as described with the open bid mechanic. If successful, you do whatever it says on the card, which in the case of Draco, is give a tag and end the run.

A Tag is a token the runner takes, which gives them the condition of 'being tagged'. Tags can be removed for 2 credits and a click - this has nothing to do with the Trace mechanic.

Now, Draco also has an extra ability, which is when you rez it, you may pay any number of credits to put that many power counters on Draco which give it a Strength increase (+1 per counter).

Taken all together, this is what Brigaldio is saying in his post:

Assume a Runner runs into your Draco. You decide to rez it and spend a ton of money boosting it to Strength lots. So much in fact, that the Runner has no hope of breaking the subroutine! They just don't have enough resources to break a Sentry of such ridiculous Strength!

Because they can't break it, the subroutine resolves. This subroutine initiates a Trace, which, if successful, will tag the Runner and end the run. Fantastic! However, you just spent all your cash on power counters, and so have nothing left to boost the actual Trace strength. It initiates at a measly Trace(2).

The Runner, seeing your error, pays a paltry 2 credits, boosting his link strength to 2; matching your trace value, making the trace unsuccessful. Because the trace is unsuccessful, you neither take a tag nor end the run.

Does that clear things up?

I'm suprised that ICE stratergy hasn't really been covered as it's a big part of keeping the runner on the back foot and away from agenda cards. When build a Corp deck a lot of my time and enegy goes into making sure that I've got the right ICE for the deck. I catagorise my ICE into three types. First turn, annoyance and extortion.

First turn ICE is stuff that you want to see in your starting hand. This ICE is cheap to rez and should have an end the run subroutine. Good examples of this type of ICE is Ice Wall or Quandry. It's job is to get played down on the first turn in front of either R&D and HQ and keep the runner out until they manage to get the relavent Ice breaker into play by which point you've improved your defences.

The second type of ICE I use I call annoyance ICE. Annoyance ICE isn't there to keep the runner out but to sting them on the way past. Stuff like Pop-up window, Pup and Data Raven are good examples of annoyance ICE. Annoyance ICE is best placed protecting assets in remote serves. You can run to trash that PAD campaign but I'm going to take my pound of flesh on the way past. Annoyance ICE should be fairly cheap as you don't want to be breaking the bank to rez it.

Lastly there is extortion ICE which is there to rip credits out of the runners account. This is the stuff the runner is going to have to throw down plenty of credits every time they want to get past it. Extortion ICE includes stuff like Tollbooth, Hadrian's Wall and Wall of Thorns. These are to be layered over the remote server you want to advance your agendas in forcing your opponant to bankrupt himself is he wants to run it. Extortion ICE also can be placed over central servers to help protect them after your opponant starts getting out his ICE breakers.

An example of ICE from an NBN deck I use looks like this...

3 Pop-up Window

3 Ice Walls

3 Draco

3 Caduceus

3 Viper

3 Data Raven

3 Tollbooth

My First turn ICE consists of Ice Wall, Draco, Viper and Caduces. That's 12 cards in a 49 card deck meaning I should get at least 1 in my starting hand. Although Draco, Viper and Caduces need a trace to end the run I am using the Making News NBN identity so I get free credits to use on them.

Annoyance ICE is covered by Pop-up Window and Data Raven. Pop-up Window is a two credit swing every time you pass it (bar Yog) and Data Raven gives my opponant a tag (with the potential for stuff like Closed Accounts and Psychographics sitting in the deck) so that costs credits and clicks if he wants to get rid of them.

Extortion ICE comes from the Tollbooth and Draco. Tollbooth takes 3 credits from the runner before they even have chance to start pumping their ICE breakers and breaking subrotines. Even with a solid Decoder like Gordian Blade it's going to cost the runner 7 credits to get past. A pumped Draco can be equally expencive for a Killer. Seen as the Deck uses cards like NAPD Contract and Red Herring I don't feel the need for as much Extortion ICE as I normally would.

Basic ICE tactics -

So as I've already said I'm looking for first turn ICE in my opening hand. My starting five credits should be enough to rez them if needs be and I'm ideally looking for two pieces. One for HQ and one for R&D. If I've not got an agenda in my hand then I might be happy with only playing one first turn ICE over R&D (but running the risk of drawing an Agenda in my first turn). I could also bluff and put down an Annoyance ICE especialy if it's cheap to rez like Pop-up window. I could also bluff my opponant by not putting ICE in front of HQ tempting him into a run on HQ if I don't have anything on hand or rely on the fact that I have 5 cards in hand to hide the Agenda if I do have one in hand.

I also want to look at what type of deck my opponant could be using in order to decide what I want to do with my ICE in later turns. If my opponant is playing Noise then I want some Extortion ICE down over Archives. If I'm playing against a Criminal then I need to be ICEing HQ to prevent an Account Siphon and I want Annoyance ICE out in front of my Extortion ICE to protect against Inside Job.

When I set up a remote server I want to decide on what job it's doing. If it's going to be used to advance Agenda then I need to pour resources into it. I need Extortion ICE, I need ICE that ends the run, I need to build it deep and I need to have the cash to agressively rez of much as it as I can do in a single turn. The other remote server is the support server which hosts cards like PAD Campaign and Marked Accounts which really should only have a single piece of Annoyance ICE in front of it.

Lastly I want to keep an eye on my opponants stratergy. If he put's down an R&D interface then I might want to ICE up R&D a bit more. If he doesn't seem keen on taking tags I might be able to suprise him by hiding an 3 cost agenda in a support remote server behind a Data Raven and quick scoring it next turn.

I'd also say it's a good idea to keep a bit of ICE back in your hand. The first thing it does is provide cover for any Agenda you have in your hand while being fairly trash proof (just be aware of stuff like Imp and Demolition Run). Secondly it allows you to respond if your opponant switches tactics.

Kahadras,

Your post on ICE tactics was informative, and included some good things to consider. It appears you change up mid-way and possibly confuse R&D and Archives. I certainly wouldn't worry about ICE on Archives first turn unless I had three in hand and suspected Datasucker or Sneakdoor (and my other two cards are agendas).

Your three catgeories of ICE are a decent way of thinking of things, but I also like the categories of Tax vs. ETR and 'Binary' vs. 'Analog' (coined in an article by MagicDave), Almost all ICE can be described with those two characteristics. ETR can end the run, and is the main way to stop a runner until they have the right breaker. Taxing ICE doesn't stop the runner, but imposes a cost of some sort (like you mentioned in your post). Binary ICE basically shuts down in the face of the right breaker (so it's either On or Off), whereas Analog ICE continues to provide resistance and effect even when the runner can break the subroutines.

For example, 1-3 strength Code Gates are generally Binary in the face of Yog.0. Stronger Code Gates are Analog, as they are pretty expensive to break through. Data Raven is Analog because you take a tag to run through it, even if you can break the trace subroutine. Tollbooth is great Analog ETR ICE, whereas Static Wall would be an example of Binary ETR. Eli 1.0 is probably one of the best Analog Taxing ICE in the game (while it has ETR subs, you can always get through it, even without a breaker - but for a cost).

Binary tends to be used for early ICE, counting on slowing or stopping the runner until they can get breakers out. Analog ICE has better long term value, and choice over using ETR versus Taxing really depends on what you want to do. Generally Taxing is best for centrals, discouraging multiple accesses in one turn, whereas ETR ICE in a remote is usually the best way to protect and score agendas. These things vary based on the deck and plans for winning, but knowing what role each of your ICE plays, and the balance you have in your deck is key to establishing the right ICE suite - one of the most important components of a Corp deck.

I did mix up the two so I've gone back and edited (at least I didn't start taking about hand, library and graveyard)