"Elementary, My Dear Wilson!"
- Famous Netrunner Stacks -
#14: Preying Mantis/Poisoned Water Supply

by Jens Kreutzer
using material by Byron "Neal" Massey
with support by Richard Cripe

"Playing Mantis ... ain't easy."
- Scott Dickie, February 15, 2001.

"I still think it's the most boring deck on this planet (and probably on most others too)."
- Erwin Wagner, later that day.

A little while after the self-mutilating Faked-Hit stack surfaced on the Netrunner strategy map, the other type of deck that could win by Bad Publicity counters and without any interaction with the Corp started to rear its ominous head on the Netrunner-L newsgroup. This time, the idea was not to play seven Faked Hits and somehow avoid a premature death by brain damage, but to use Poisoned Water Supply as the main theme instead. This Proteus rare reads:

Poisoned Water Supply - Cost: 4
Prep - Bad Publicity
Play only if you have at least two connections in play. Trash two connections. Give the Corp 1 Bad Publicity point. If the Corp has 7 or more Bad Publicity points, it loses the game, even if it fulfills victory conditions at the same time.

The question about which kind of connections should best be used was quickly decided. While any 0-cost connection (like Databroker or Smith's Pawnshop) offers a cheap solution, and while generally useful connections like Broker or Crash Everett might help with the preparation before being trashed, it became clear by 1997 at the latest that Preying Mantis seemed to be custom-made for the job at hand (i. e., getting poisoned). It's non-unique, installs for free, and speeds things up by offering extra actions. Since speed is of the essence in this no-interaction plan, Poisoned/Mantis has the potential for an uncanny synergy that comes very close to the horror that is Psycho Tycho, as would gradually become apparent over the course of the next couple of years. The race was on.

An early concept by Leonard Blado (September 8, 1996) tried to win in an average of 6-8 turns by combining Arasaka Owns You, Faked Hit, Loan from Chiba, Militech MRAM Chip, N.E.T.O., Poisoned Water Supply and Preying Mantis. The idea to use Arasaka Owns You as the main draw/bit engine was nice, but since it cannot be played when the Runner suffers unpreventable damage (Faked Hit, Preying Mantis), the concept was flawed - back to the drawing board.

On June 1, 1997, Byron "Neal" Massey made the first serious attempt, relying exclusively on the Poisoned/Mantis combo. The deck he suggested would contain nothing but Preying Mantis, Poisoned Water Supply, Loan from Chiba and Bodyweight Synthetic Blood, winning in about 8-11 turns (a variant with Top Runners' Conference is also feasible). This is really the deck concept in its purest incarnation, albeit much too slow for being competitive. But since this approach didn't get much mileage out of its Preying Mantes (probably using them only in the last turn), there was still a lot of room for improvement and speeding-up.

On the next day (June 2), Leonard Blado showed us a way of dealing with the brain damage that results from no-holds-barred Mantis abuse: Emergency Self-Construct (ESC).

Here is how it works: Each round, the Runner installs and uses as many Mantes as possible, while keeping one less card in hand than the number of times a Mantis was activated. The last point of brain damage would normally flatline the Runner, but ESC then kicks in, preventing the flatline and convenienty removing all brain damage as well. The Runner starts the next turn with zero cards in hand and one less action, but with an armada of Mantes installed, this is a very minor annoyance. Since it was ruled that multiple uses of ESC do not reduce the action count further (instead re-setting it to 3 each time), using it each round is no problem at all. The hand size does keep decreasing, though, but that is irrelevant since there are never any cards left in the Runner's hand at the end of the turn anyway. This is Leonard's idea of a 3-turn win with this strategy:

Turn 1, Actions 1-3: Install Preying Mantis
Turn 1, Action 4: Install Loan from Chiba
Turn 1, Action 5-6: Play Bodyweight Synthetic Blood
Turn 1, Action 7-9: Install Preying Mantis
Turn 1, Action 10: Install Emergency Self-Construct

Turn 2, Action 1: Draw
Turn 2, Action 2-6: Play Bodyweight Synthetic Blood
Turn 2, Action 7-14: Install Preying Mantis
Turn 2, Action 15: Install Loan from Chiba
Turn 2, Action 16: Install Emergency Self-Construct

Turn 3, Action 1: Draw
Turn 3, Action 2-4: Play Bodyweight Synthetic Blood
Turn 3, Action 5-7: Install Loan from Chiba
Turn 3, Action 8-14: Play Poisoned Water Supply

The problem with this is that there is about a 1-in-10,000 chance of drawing the right cards at the right time. Especially the seven Poisoned Water Supplies at the very bottom of the deck (you are in trouble if you draw them before the last turn) are just improbable. Thus, the theory was formulated - a 3-turn win is possible! -, but what kind of deck should you build in order to pull this stunt off reliably? That was a tough question.

Time passed, until Neal (with the help of Erwin Wagner) tried in earnest to solve the puzzle, hoping to play his newly developed strategy in the 1998 U. S. Northwest Championships. As a quirk of fate, he got the days for the Sealed and Constructed tournaments confused and couldn't play his new innovation after all, because there wasn't enough time to tweak it, but he really got the ball rolling with his post to the Netrunner-L on 12 November, 1998. Part of this effort was to disprove the claim that there were no new deck types to be found in Netrunner anymore, which was heard time and again during those pre-Classic days. It seems that Byron Bailey had also been working on the Preying Mantis/Poisoned Water Supply deck, with a similar approach, but hadn't taken it quite as far as Neal and Erwin.

Was the three-turn win possible after all? Let's hear it from Neal himself: "The third turn is where I want to stop, applying the winning lock with my sixteen Preying Mantis actions. In practice, I haven't been able to engineer it. You start each turn with no cards, and usually no bits. My brother convinced me that you should spend the third turn just like the first two, installing Preying Mantis until you have the magic number (twenty-two) on the table." And so, in practice, four turns would become the realistic target.

Here are Neal's ideas in a nutshell: "Draw cards and install as many Preying Mantes as possible each turn. Since each Mantis immediately grants another action, they can be installed at no real action cost, greatly speeding up the process. Each turn, you must install an ESC and have exactly one less card in hand than the number of Mantis actions that were used. The accumulated brain damage will flatline you at the end of turn, but ESC lets you go virtual and immediately remove all brain damage. You start each turn with no cards in hand, so you must begin anew with drawing. The number of available actions increases rapidly each turn since more and more Mantes are being installed. Do this for three turns, and in the fourth, play misc.for-sale on all the Mantes. You now have an incredible amount of bits and actions and should be able to win in various creative ways, one of them being to instead sell all Mantes but twelve and play six Poisoned Water Supply, followed by one Faked Hit.�h

However, the problem was still to get the �ewinning cards�f, a. k. a. winning kit, at the right time. Also, the question of an affordable draw engine needed addressing. Neal put a lot of thought into these issues and suggested many interesting avenues that might be taken. In the end, it appears that two archetypes of the stack evolved with time. Neal never really made his final results public, because he wanted to put his work to good use in Constructed tournaments, but based on the comments that he let slip, several other players tried to emulate his strategy, notably during the Finals of World Domination 1999. Up till the present, an aura of mystery still surrounds this stack that nobody but Neal and Erwin has ever seen (with the exception of Douglas Kaufman, who judged the famous IRC match challenge between Neal and Sean Harvey in January 2000).

Apparently, the question of draw engine/bit engine proved too difficult for Neal's original plan. In one, the most successful archetype, he settled for Loan from Chiba as bit engine and N.E.T.O. as draw engine (this deck list was created by Richard Cripe, based on conjecture):

Preying Mantis/Poisoned Water Supply
(Loan-from-Chiba Archetype)

15 Preying Mantis
7 Poisoned Water Supply
9 Loan from Chiba
1 Emergency Self-Construct
3 Sneak Preview
8 N.E.T.O.
2 MIT West Tier

Sneak Preview cleverly combines with N.E.T.O. and Emergency Self-Construct, fetching the latter from the trash or the stack as needed. Incidentally, to illustrate the guesswork that was involved in approximating Neal's deck, his complaints on the NR-L about the ruling which states that Sneak Preview doesn't bring back ESC from the trash at the end of turn served as a hint that Sneak Preview was part of the Poisoned/Mantis stack. Richard Cripe wrote a detailed comment about this reconstruction:

"This deck is based off what I know of Byron "Neal" Massey's PWS deck. He may have more to say about it, having more experience and knowledge of the deck. You could probably play with one less MIT, but if you draw it in your opening hand, you're screwed. You could probably play with one less Sneak Preview, but trying to scramble to get the ESC down is a real pain and often leads to flatlining. The goal here is to install every LFC and PM you get your hands on while not flatlining yourself.

"Best first turn:

a1: N.E.T.O.
a2: Preying Mantis
a3: Preying Mantis
a4: Draw 4 cards with N.E.T.O.
a5: Preying Mantis
a6: Loan from Chiba
a7: Loan from Chiba
Leaving you with 0 cards and 25 bits.

�gApproximate turn two (12 actions in total):

- Draw 16 cards (Don't draw the second MITT West Tier or the third Sneak Preview!)
- Play Sneak Preview to get Emergency Self-Construct
- Play 5 Preying Mantes and 2 Loan from Chiba
- Cards discarded to brain damage should bbe N.E.T.O. or Poisoned Water Supply, maybe a Sneak Preview or a Loan from Chiba.

�gApproximate turn three (15 actions in total):

- Draw 24 cards (but not the last MIT)
- Play the rest of the Preying Mantes, as many Loans as you can safely, and a Sneak Preview. You may need to let a Loan or two go for damage here.

"Turn four:

- Draw and play MIT
- Draw the seven Poisoned Water Supply, annd any Loans you may need to pay for them.
- Win by playing the seven Poisoned Water Supply."

So - why N.E.T.O. instead of Bodyweight Synthetic Blood? For once, it solves the problem of getting the right cards at the right time, because you can simply choose not to take that MIT into your hand when you see it early on. Also, since it doesn't count as drawing, N.E.T.O. evades City Surveillance, which would have been a serious danger otherwise. As it is, but two dangers remain, namely Blood Cat and Underworld Mole, which spell quick doom if aimed at the Loans. As Richard Cripe remarks, one Sneak Preview and one Preying Mantis might be switched for two Access through Alpha as a countermeasure, but since this slows down things a lot, it's perhaps best to take one's chances, hoping that even if the Corp packs Cats and Moles, they won't show up right away. Backdoor to Netwatch is also worth consideration, but it is only of use against Underworld Moles. The Deck has also been discussed in this context.

Let's now take a look at one of the other conjectured archetypes of Poisoned/Mantis (a. k. a. Poisoned Dealer Supply), which has misc.for-sale as its main feature. The idea is to install more Preying Mantes than needed for the Poisoned Water Supplies, and sell the rest off to misc.for-sale during the last turn. In order to make do without bits until the final turn, Jack 'n' Joe is used as draw engine. This is Scott Dickie's version, which he played (without much success) in World Domination 1999 - as a variation on the seven Poisoned Water Supplies, one Faked Hit reduces the number of Preying Mantes that need to be left after misc.for-sale to twelve:

Preying Mantis/Poisoned Water Supply
(misc.for-sale/Jack 'n' Joe Archetype)

22 Preying Mantis
4 Emergency Self-Construct
6 Poisoned Water Supply
1 Faked Hit
15 Jack 'n' Joe
2 MIT West Tier
1 misc.for-sale
4 Crash Everett

With 55 cards, Scott's deck seems rather big, and though Crash Everett can be used to 'float' crucial cards on top of the draw stack, N.E.T.O. might still be the better choice. In fact, there are still more possible permutations of the theme. Richard Cripe reports that "an entirely alternate method of playing the deck revolves around using a Jack 'n' Joe engine, not using any Loans until the turn it is ready to win. It is less vulnerable to Blood Cat or the Mole, but it is more vulnerable to City Surveillance. The Corp can then start trashing your Preying Mantes, but the ESC pretty much stops meat damage. I believe this deck is a turn or two slower than the N.E.T.O. version as well as it just doesn't draw cards as fast and has less throw-away cards to soak damage.�h

This would imply a combination of Jack 'n' Joe and Loan from Chiba, but no misc.for-sale. Richard once more: "The misc.for-sale is not essential at all; it just boosts the already sick amount of bits from the Loans." Unfortunately, there is no decklist for this variant, so you have to figure it out yourself. Actually, Neal himself hinted at still another version on January 13, 2000 - one that combines N.E.T.O. with misc.for-sale: "As far as playing the stack, the secret is to get to a spot where you have 14 Preying Mantes installed, then end your turn with an MIT waiting to be drawn. You N.E.T.O. for the MIT (with good planning it should happen on the first action of your fourth turn). Then you play it, use N.E.T.O. to draw four cards per action until you get all the Poisoned Water Supplies in your hand, play your misc.for-sale, and then all seven Poisoned Water Supplies." As you might have been expecting by now, no decklist here either.

So there you have it: Though it is easily the most notorious Runner stack since Precision Bribery/Time to Collect, Poisoned/Mantis remains shrouded in mystery. Neal (and others) claim that a rather reliable win in the fourth or fifth turn is possible with this strategy, but to come up with the optimal card combination is left to the individual player. I find this to be a very desirable situation, since it makes players think for themselves and get creative. Moreover, coming up with a clever decklist alone doesn't do the trick, since actually playing Poisoned/Mantis is reportedly no easy task: "Unlike Psycho Tycho, this deck does not play itself but rather requires concentration and memory to achieve success," Richard Cripe tells us. Accordingly, many Runners are said to have inadvertedly flatlined themselves with this strategy.

Before I give you a list of ingredients for cloning your own Poisoned/Mantis, let's hear Richard once again with his theory on the evolution of this strategy: "My guess would be that it started as a BSB/Loan engine that evolved into two separate methods of dealing with City Surveillance." He is referring to the N.E.T.O./Loan variant and the Jack 'n' Joe/play-Loans-only-in-the-final-turn variant: The former avoids City Surveillance altogether, while the latter avoids playing the vulnerable Loans until the last turn, daring the Corp to start trashing resources (as ESC protects against meat damage). And here's the toolbox for Do-It-Yourself Poisoned/Mantis stacks (combine at your leisure):

- Bit engine: Loan from Chiba, misc.for-saale, Do the 'Drine (suggested as a way to avoid playing with the dangerous Loans).
- Draw engine: N.E.T.O., Jack 'n' Joe, Craash Everett, Sneak Preview, MIT West Tier, Mantis, Fixer-at-Large.
- Core: Poisoned Water Supply, Faked Hit, Preying Mantis, Emergency Self-Construct.
- Protection: Access through Alpha, Backdooor to Netwatch, The Deck.

With Preying Mantis/Poisoned Water Supply (PM/PWS) winning in four or five (or maybe even as few as three) turns, it is easily the fastest Runner deck out there. A question that has been asked repeatedly is how it measures up to the hallmark of fast Corp decks, Psycho Tycho. Richard Cripe said (on February 16, 2001) that he believed that PM/PWS was "generally faster" than Psycho Tycho, though he'd never really done a full study. We have to consider here that the Corp starts the game, and that it is therefore enough if it can win by turn four, which should perhaps be doable with Psycho Tycho. In any case, like Richard concluded, it's a close call. But with Tycho Extension banned in the Revised Constructed Format, there is a real danger of PM/PWS becoming dominant, and there have been appeals to do something about it, maybe making Preying Mantis unique or banning Poisoned Water Supply in that format, or issuing an erratum on Emergency Self-Construct to the effect that it may only be used once per game.

Since PM/PWS is one of the no-run/no-interaction Runner stacks (like Masochism Rules), we'd rather see it go, since it tends to spoil the fun out of Constructed play - this of course isn't supposed to discredit Neal's genius in any way, but the stack does play much like Solitaire. In order to have any chance of even slowing down PM/PWS, the Corp needs to include some very specific cards, like Blood Cat or Underworld Mole, and must draw them in time.

That nothing has happened yet to curb PM/PWS has four reasons. First, it uses a mind-boggling number of rare cards, many of them from the Proteus expansion, which are notoriously hard to find nowadays. In fact, apart from Loan from Chiba, Neal's version of the stack is all rares! And Loan from Chiba is a valuable uncommon to boot. This means that only a very small number of players will be able to actually build PM/PWS and play it in a meatspace tournament (online play, like on IRC, is a different story). But a deck that only shows up once in a blue moon perhaps needn't be banned. Second, since almost nobody has ever seen the actual decklist created by Neal, PM/PWS is basically a phantom. Neal hasn't had the chance yet to really show off the merits of his stack, and to ban it outright even before it has shown its full potential seems a bit hasty, and very unfair to Neal as well. Third, much unlike Psycho Tycho, PM/PWS is quite difficult to play. There is a lot of juggling with actions and bits, and the correct way of drawing cards. This means that winning with PM/PWS is a measure of skill in the end, and that cannot really be such a bad thing. Fourth, being a direct consequence of points two and three above, players trying their luck with PM/PWS in tournaments mostly got a very bad performance out of it. Most notably, the PM/PWS clones played in the WD 1999 Finals didn't do well at all. It remains to be seen how it fares in WD 2002, however.

As if it wasn't enough already, though, PM/PWS has recently got another boost by a clarification that concerns the way Emergency Self-Construct functions. It used to be a worry of PM/PWS players that they had to have exactly one less card in hand at the end of turn than the total number of Preying Mantis brain damage that would be suffered, so that the last point of brain damage would trigger ESC and remove them all in the process. It is clear that, should there be any more incoming damage, the Runner would flatline again and lose if there was no other Emergency Self-Construct installed for each additional point of brain damage past the limit - and this gets deadly real soon.

But there was also a perceived danger at the other end of the spectrum, which was, strangely enough, not enough brain damage. If the Runner has, say, twelve cards in hand at the end of turn and suffers eleven points of brain damage, he or she would 'survive' this, but would then end up with a handsize of -6, which means a lost game. But this effect of having to discard 'more than you have' is counted as flatlining as well according to the clarification. This means that when the Runner attempts to 'discard' to -6, ESC triggers and removes all brain damage, and the handsize is then reduced by just one, which should bring the Runner out of the danger zone. This makes playing PM/PWS much easier, since any number of brain damage < or = (cards in hand + 1) at the end of turn will do. This aspect of playing PM/PWS was pointed out by Douglas Kaufman and others in January 2000. As noted above, the current understanding seems to be that the handsize reduction incurred by multiple uses of Emergency Self-Construct is cumulative, as opposed to the one lost action, but this doesn't really hurt PM/PWS, since it wins before ESC-induced handsize reductions can result in a negative handsize that is not due to brain damage.

Another issue that saw discussion in the context of PM/PWS is the interpretation of N.E.T.O. Here�fs a reminder of N.E.T.O.�fs game text:

N.E.T.O. - Cost: 0
Resource-BBS
Action: Look at the top four cards of your stack. You may bring any prep or resource cards among them into your hand. Pay one bit for each card taken in this way, and show those cards to the Corp. Shuffle the rest back into your stack.

The question about this wording is the "pay one bit" part. Since it doesn't appear in front of the colon like the action cost, it isn't written in compliance with the usual cost:effect syntax. But this seems to imply that paying a bit for each card taken into hand is a penalty, not a cost, and according to the rules, penalties can be ignored if they cannot be fulfilled. This in turn would mean that even if there are no bits in his or her pool, the Runner could still get four cards per action for free! With a supercharged draw engine like this, Neal estimated that he could win in three turns with PM/PWS. Tom Wylie ruled on November 12, 1999, that paying a bit is to be considered a cost and therefore must be paid in order to get the effect, but this issue is still controversial, since it contradicts the card text.

In January 2000, Neal gave us a hint on how to play PM/PWS with style: "When I am playing the Poisoned Dealer Supply stack, I turn each Preying Mantis sideways when I declare that I am using his action (a.k.a. 'tapping' in Magic). On my fourth turn, I only discard copies of Preying Mantis that are already turned sideways when I play a copy of Poisoned Water Supply. Because my base actions are down to three, it has never been even a minor inconvenience to play with this convention. Fifteen or sixteen of the eighteen or nineteen actions taken on the fourth turn require the use of Preying Mantis. Only seven of those actions are used to play Poisoned Water Supply. As it turns out, there are always two Preying Mantes available for destruction. If your version of the stack runs in to trouble (I can't imagine how, but ... ) you can always play with a single Faked Hit and six Poisoned Water Supply. Then you only need to kill off twelve dope dealers, an easy thing when you are running a drug empire."

As is fitting, Neal gets the final word on the metagame implications of his creation: "If played properly, Poisoned Dealer Supply should win on turn four. That means the Corp gets twelve actions to tag and kill a Loan. That might be possible in some fraction of games. But when you add in the fact that the Corp must also prepare itself for Runners with no resources, and non-Loan bit engines, plus TagMe plans, it's not realistic. Blood Cat, though ... ." I guess we can expect Blood Cat to show up in quite a few Corp decks in the future, should Preying Mantis/Poisoned Water Supply become really popular with players - which fortunately hasn't happened yet.

"Oops, looks like I broke the game, sorry."
- Neal

"Oops, guess we have to add Neal to the banned list then."
- Benjamin Ford