
Seven games in seven days across seven different modes, which is the first clean week this desk has filed since June. I have spent the weekend trying to name what they have in common. Four of the seven were wins on paper. What they share is that in six of them the thing that decided the game was taken off the table rather than won on it.
At Punta Vela a Breacher put a demolition charge on the leg of the catwalk he was standing under, rolled a natural one, and wounded two of his own. Four activations later the Pressure Clock fired, the Climax came up Structural Collapse, and the game dropped that same catwalk for free across all three objectives. Nobody crossed the centreline again. Eastern Compact took it five to one and neither commander scored a single point after round four.
On the Raze the cask was gone before round two. In the Ferghana corridor a splice killed the lead vehicle on the only bridge in the valley, corked a three-vehicle convoy exactly where they wanted it, then spent four turns on the wrong side of their own roadblock. Weir put a Type-3 one wound from dead and lost the priority die in ten rounds out of eleven. Circuit cleared four gates on borrowed sixes and died in the fifth doorway with the door already open behind him.
So: the denial issue. The word arrived from the errata rather than from me. The Skirmish patch that went in this week gives Breach Tokens a third spend and calls it DENY. Cancel one no-roll ability, declared before it resolves, once per operator activation. It is a small rule and it is also the first thing in the line that lets a player do on purpose what the board has been doing to everybody by accident all week.
There is a real hobby argument underneath this and it is worth having out loud. A game that only rewards taking things is a game where the interesting decisions all happen in the first four turns. A game that also prices closing something, a lane, an ability, a crossing, a scoring window, keeps deciding things right up to the cap. Every fight in this issue was still live on the last activation. The clearest case is Wreckage Rights on the Terrace Grid, where a salvage consortium lost three of its seven machines to the Coalition's one and won by three points, because the mission counts standing bodies near a wreck pile and never once asks who is winning the fight.
Quill Adair has built you a gantry that falls over. Rio Solveng went looking for why the maps in this world are mostly wall with a few doors in them. Tib Marlow has seven things to do about it on Wednesday.
Read the patch notes. They run long this week and they are load-bearing.
Bragin spent a demolition charge trying to drop a cable gantry onto the middle marker, rolled a natural one, and put two of his own operators down instead. Four activations later the Pressure Clock tripped, the Climax Table came up Structural Collapse, and the game dropped the same gantry across the full width of the yard with all three phase cabinets underneath it. It earns the week because both commanders then played four more rounds correctly and neither of them could score: this is a clean record of a table that stopped being winnable and two players who kept working anyway.
Wreckage Rights counts unbroken units within twelve inches of a wreck pile and scores nothing at all for kills, so seven cheap machines beat five good ones while losing three to one. Read the victory condition before you write the list. If the mission counts bodies, buy bodies.
BLACK MARKET 8 // PRC 5 // T5A five round clock ended in two because the Sparrows went straight at the cask and contested nothing else. Most of that scenario's budget sat on the table untouched, so build your Raid Racers events assuming the package leaves early and put the complication in the exit rather than the approach.
SPARROWS EXTRACT // T2 OF 5The Irregulars reached actionable proof in a single session and still lost the murder, because two natural ones inside the same break let the syndicate close the file its own way. Cold Case pays for routing, so spend soft leads as threshold drops and check that a declared special is legally spendable before you build a character around it.
PROOF FILED // MURDER UNATTACHEDWeir lost the priority die in ten rounds of eleven and finished one wound off a kill. In solo play losing priority takes half of everything you own that round, which is why the desk's own errata now exempts a one-model side from Priority Pressure. Build a hunt plan that survives going second twice running.
CONTRACT FAILED // SPECIMEN DOWNSTREAM // T11Slow Match killed the lead vehicle on the only crossing in the valley and corked the convoy perfectly, then spent the rest of the game unable to reach the man they came for. Before you block a road, work out which side of it you need to be standing on afterwards.
PARTIAL // ASSET OUT, STRANDED // T6Impossible deals one operator, and the deck handed a five-gate rifle problem to a hacker who rated Good on exactly one of them. The arithmetic was settled before the first card turned. Check the gauntlet against your one body's proficiencies and take the difficulty your deck can actually answer.
OPERATOR KIA // C5This week on the table, the rules moved. Everything the desk changed or confirmed, logged in one place.
Twenty five fixes clearing the Cold Case and Raid Racers backlog, covering the investigation loop, the Evidence Chain, the Burn Counter, the Zone ladder, possession contests and the security drone.
Twenty one fixes in one day: eight approved balance and design calls across Iron Line, Raid Racers, Threat Response, Big Game Hunts and the solo core, ten Skirmish clarifications covering terrain, the Climax Table, melee, the Pressure Clock, stealth detection, grenades, the Medic and cover, and three approved Skirmish design calls on Breach Token generation, a new Deny spend and the Recon Drone.
Forty five daily-desk playtest clarifications across Iron Line, Raid Racers, HVT, Threat Response and Big Game Hunts, including new vehicle rules, a new Raid Racers traversal mode, and a full stat line for solo enemies.
The Punta Vela game ended under a gantry and the gantry was a length of foamcore on two soup tins. I know that because I built it, on a Tuesday, in about eleven minutes, and I have been quietly embarrassed about it since Marlow's report went up. The piece that decided the most interesting fight of the week could not do the thing the fight was about. It could not fall down.
The errata this week gave Skirmish an OVERHEAD terrain row and a full paragraph for Climax result 3, and between them they describe a very specific object. Standing, an overhead piece grants Elevation to anything on it, Light Cover to anything at ground level inside its footprint, and blocks no line of sight at all. Collapsed, it becomes Difficult Terrain across that same footprint, still grants Light Cover, still blocks nothing, models on it drop to the ground with no falling damage, and anything caught underneath is displaced to the nearest legal ground. Those are two different pieces of terrain occupying the same eighteen inches of table, and the whole point of the thing is the moment one becomes the other. So build both, and make the swap take four seconds.
Core is a length of ten millimetre styrene I-beam, cut to eighteen inches and left in one piece. Deck it with brass mesh or the etched grille off any cheap photo-etch fret, two inches wide, glued along the top flange so a 32mm base sits flat with room either side. The legs are the fiddly part and the answer is to make them not fiddly: four uprights of 8mm square tube, each glued to its own 40mm round base, each with a shallow notch filed into the top that the beam sits in. Not glued. Sits in. A gantry that lifts off is a gantry you can collapse without touching a model.
Then build the wreck. Same footprint, one inch tall instead of five, and it goes faster than the standing piece: a strip of 2mm foamcore cut to eighteen by two, the top skin scored and torn open in three places, the leftover I-beam offcuts laid across it at bad angles, a handful of cork bark, and every join buried in sand and PVA. This is where the offcuts from the standing build go, so cut the beam long and keep everything.
Failure point is the notches. File them too deep and the deck sits low enough that a model on the ground beneath cannot be picked up without lifting the whole span, which defeats the purpose. Too shallow and the beam walks off the legs the first time somebody leans on the table. You want it sitting about a millimetre into the notch, held by gravity and nothing else, and you want to test that with the heaviest model you own standing on the deck before you paint anything.
Paint both halves in the same session, same primer, same wash. A wreck that reads as a different piece of terrain from the thing that fell over is worse than no wreck at all. Grey primer, a heavy sponge of dark rust across the deck and the leg joints, brown wash into everything, then the one detail worth the time: paint the trunk line. A single conduit running the length of the beam in dull yellow, and on the wreck version, that same conduit torn open with the cut ends in bare bright metal. That is the entire story of the piece told in two colours.
Four of the seven tables this week had exactly one way across them, and in three of those cases the players did not notice until it mattered.
That is not a coincidence of scenario design. It is what the ground looks like now. The Ferghana corridor where Slow Match took the convoy has a side channel running its whole length, cut by the same water management scheme that was going to make the valley farmable in the 2030s, and one bridge. There were eleven. Nine came down in the first two years of the contested-zone period, dropped by people who were absolutely certain they would not need them again, and the tenth silted its approaches. What survives is a single span that four separate armed parties have now spent three decades not destroying, because the corridor's entire value to any of them is that things can still be moved through it.
The same arithmetic runs the other direction in Samut Prakan. The Terrace Grid is a drowned aquaculture belt, hundreds of square kilometres of shallow pond wall, and the dikes are the only ground in the district that carries weight. Nobody defends the water. Everybody fights on the raised lines, and both commanders in the Wreckage Rights game deployed on a dike because there was nowhere else a two thousand point formation could stand. Punta Vela is the third version of the same fact: a wind farm on a shelf where the katabatic wind never stops, so every structure is bolted twice and the only thing that spans the yard is a cable gantry sunk four metres into shale.
The habit of mind this breeds in the people who live there is worth stealing for the table. An operator working the Eurasian corridors does not read a map for cover. They read it for the number of ways out, and they price every piece of demolition against the possibility of coming back the same way. Blowing a crossing is understood in that theatre as an admission that you are never returning, and it gets treated socially the way a burned bridge deserves: a statement about you, made permanent, in front of witnesses.
Slow Match made that statement by accident. They killed the lead vehicle on the bridge because the ambush window said kill the lead vehicle, and they did it well, and the wreck corked a three-vehicle convoy exactly where they wanted it. Then they found out the asset they had come for was on the far bank with a working cargo truck and a technical, and the only crossing on the table was full of the thing they had just shot. Partial victory. Asset out, team stranded on the wrong bank of their own success.
The table consequence is one line long and it will change your next three games. On a 24 by 24, stop building the symmetrical board with cover on both flanks. Build one line of movement across the short axis that both sides need, make it destructible, and put the objective on the far side of it. Then say nothing about it. The players will find it in round two and spend the rest of the evening arguing about whether to close it.
Breach Tokens got a third spend this week and it is the best one on the list. DENY cancels a single no-roll ability for that operator's activation: a Medic's Field Dress, an adjacent auto-success Override, any class ability that resolves without a die. Declare it when the ability is announced and before it resolves, once per operator activation. Hold one token back from round three onward. The abilities that never touch a die are the ones your shooting has no answer to.
Structural Collapse got a full paragraph this week and it came straight out of Punta Vela. The trailing side chooses the piece, the wreck is Difficult Terrain granting Light Cover and blocking no line of sight, models on it drop with no falling damage, and a piece that carries or blocks a scenario objective may not be chosen at all. That last clause would have saved the Crossfire game. Read the paragraph while you are setting up, not when the clock trips.
The terrain chart gained an OVERHEAD row: a raised passable structure gives Elevation to models on it, Light Cover to models at ground level inside its footprint, and blocks no line of sight. That combination is unusually generous, so run one per table and put it where standing on it costs something. A catwalk over the centre objective rewrites every approach on the board without turning the game into a staring contest, because it blocks nothing.
Threat Response and Big Game Hunts now exempt a side fielding exactly one model. A squad absorbs that tax across four bodies. A lone hunter loses half her round to it, which is what happened to Weir for ten rounds out of eleven and is the single reason that contract failed. Losing priority still costs the solo side the initiative, which is plenty. If you are running an older printing, house-rule it tonight and write it on the sheet.
Apply the single best cover modifier only. This is the most common quiet error at a table with good terrain, because a model behind a wall inside a ruin under a smoke cloud feels like it should be collecting three things. It collects one. Announce it while the terrain is going down and the argument never happens. Announce it with a tape measure already on the table and you are negotiating, not adjudicating.
Tokens now generate only from a Flesh Wound inflicted by an enemy attack: a shot, a melee, a grenade, a demo charge belonging to the opposing player. Terrain, Climax results, falling and your own side's fire generate nothing. Without that fix the Punta Vela collapse would have minted a fistful of tokens for two players who had done nothing to earn them. If your group has been paying out on terrain, you have been running a much swingier game than you thought.
Active detection now names a point within 12 inches and line of sight, and every Stealthed model within 3 inches of that point rolls TACT or is revealed. It is an area, so stop pointing it at a model you think you have spotted and start pointing it at ground an Infiltrator has to use. Doorways, the approach to an objective, the near end of your one crossing. Two Ghost Infiltrators walked Punta Vela unopposed and the Compact never swept the only lane they could have taken.
NUR HABASH plays almost entirely alone. Two nights a week in a converted signal box outside the Saharan Oasis freight yards, mostly Big Game Hunts and the solo rules, and she has finished eleven campaigns without a second player ever sitting down at the table. She keeps what she calls a denial ledger, a running column of every turn the game took away from her, and she has kept it for four years. Kose called her the morning after the Weir report went up.
I have a group. I play with them on Sundays and I like them. A solo game tells you the truth about a rule in a way a shared game cannot, though, because there is nobody across the table being generous with you. If a rule is bad, it is bad at me, in the dark, and I write it down.
One line for every turn where the game stopped me doing the thing I had already decided to do. Not failure. Failure is fine, failure is the dice doing their job. Denial is when I never got to make the roll at all. Lost the priority die and the specimen moved out of arc. Natural one on the wrong action, lost the activation. Terrain went perilous and I paid a toll to stand still. Four years of that is about nine hundred lines.
That most of the games I lose, I lose in the first two rounds in a way I do not notice until round nine. When I read the arithmetic on the Weir hunt I recognised it instantly. Ten rounds out of eleven is a freak, that is once in a hundred and seventy games, but the shape of it is completely ordinary. You do not lose a solo hunt to the specimen. You lose it to two hundred small subtractions and then the specimen is just there at the end to sign for it.
I never plan more than one round ahead and I always have a second use for the round I am standing in. If I lose priority, what am I doing with the half round I still have? If the answer is nothing, I built the turn wrong. People hear that as a truism. At my table it is a checklist, I say it out loud on every activation, and it has saved more hunts than any piece of kit I own.
Solo players over-read the threat and under-read the map. I did it for two years. You will spend twenty minutes on a specimen's anatomy chart and thirty seconds deciding where you are standing, and the specimen is not the thing that kills you. The floor plan is.
I roll the priority die for the whole game before I set up. Eleven rounds, eleven dice, left to right along the edge of the mat, and I look at them. Then I sweep them up and roll properly, honestly, one at a time. It is not a system. It is so that I have already seen what a bad run looks like before it is happening to me, and I do not start bargaining in round four.
Nothing this month. They just gave the generic specimen types a stat line and I have wanted that since the book shipped, so I am going to enjoy it for a few weeks. Ask me in December. There is always something.
What is the smallest amount of terrain that still makes a 24 by 24 skirmish table play properly?// Two Boxes And A Book // Low Coast
Six pieces, and one of them has to matter. Two blocking pieces big enough to hide a 32mm model completely, two light cover pieces, one raised piece, and one line of movement across the short axis that both sides want. Six. The mistake at a thin table is spreading four boxes evenly and calling it cover, which gives you a board where everything is a little bit protected and nothing is a decision. Punta Vela was a 24 by 24 with an impassable hall, a transformer bank, a gantry, a spoil berm, some reel racks and two pylon feet, and it produced eight rounds of hard choices before the game buried its own centreline. If you own two boxes and a book, the book is your raised piece. Put it in the middle.
How many objective markers is too many before the table stops looking like a place and starts looking like a board game?// Chip Farm // Reykjanes
Three, and it is one of the few numbers I will defend in writing. Three is a decision, because you can hold two and reach for the third. Five is a chore. The reason a table starts reading as a board game is rarely the count anyway, it is that the markers do not belong to anything. A phase cabinet is an objective. A poker chip on a warehouse floor is a marker. Build the objective into the terrain, keep its scoring on a card instead of on the table, and three will never look like a bingo hall.
Our group agreed a house rule mid-game and now nobody can remember exactly what we agreed. How do you keep rulings straight across a campaign?// Ruled It Twice // Gorodskie Ekho
Write it on the scenario sheet, in the moment, in the words you actually used, and initial it. Not a document, not a wiki, nobody is going to read either one. One line on the sheet for the game you are in, and at the end of the night the sheet goes in the box with the mission. The next time that scenario comes up the ruling comes up with it, which is exactly when you need it and the only time you will remember it ever existed. This desk keeps a changelog for the same reason and it is the least glamorous file we own. Look at the patch notes in this issue. Dozens of rulings, none of them invented at a desk, all of them somebody writing down what they did at the moment they did it.
When the firing arc came in, half my models lost their good shots. Do I rebase everyone to show facing or just agree a house marker?// Which Way Is Front // New Geneva
Do not rebase anybody. A rebase is four hours of work to solve a problem a two millimetre notch solves in ten minutes. File a small nick in the front edge of every base, or glue on a sliver of plasticard, or if you want it reversible, put a dot of enamel on the rim and let it wear off over a year. What matters is that the whole table can read facing from standing height without leaning in, and that the answer is identical on every model in the game. The argument is never really about facing. It is about whether the two of you agreed the facing before or after somebody picked up a tape measure.
I lost the priority die nine rounds out of eleven in a solo hunt and it stopped feeling like a game. Is there a legitimate way to hedge that or do I just take it?// Ten Of Eleven // Indo-Pak Line
Take it, then build for it. Two things. First, check your printing: a side fielding exactly one model is now exempt from Priority Pressure in both Threat Response and Big Game Hunts, and if your book predates that ruling you have been paying a squad's tax with one body. Second, and this is Nur Habash's answer elsewhere in this issue and it is better than mine, never build a turn that only works if you go first. Ask what the half round is for. A run of nine is a genuine freak and you have my sympathy, but those nine rounds only cost you the hunt because every one of them had a plan in it that needed the other half.