Skip to main content
⌖ OF Intel

GUIDES

Stop the Small-Probe Grinder: Attrition Defense in OpenFront

A neighbor sends 20k-30k probes on a timer so you never get a counter window. Hold at max, layer posts, and counter only when their committed force dips below your standing force.

Tutorial Difficulty · Intermediate Published Nov 21, 2026 Updated Nov 21, 2026 Reviewed by OpenFront Intel editors #defense#combat#attrition#troops#strategy

A neighbor of roughly equal size does not commit a full attack. Instead they send 20,000 to 30,000 troops on a timer, fast enough that your forces never refill before the next wave lands. You have no clean counter window and you are slowly bleeding to death. This guide answers exactly that situation: how to make the probe cost the attacker more than it saves them, and when to finally turn the fight around.

Direct answer: hold at max, layer posts, counter only on the dip

For v0.36.x, verified 2026-11-21: keep your standing force at or near maximum so your own regeneration runs at its fastest, put one or two Defense Posts at the front so every probe meets a 5x combat and 3x speed penalty, and only counter the moment their committed force drops below your standing force. Do not drop to a partial force for a full counter; the partial-force trap is exactly how the grinder wins. The 40-word version: stay at max, build the post at the front, let the 5x multiplier starve their probe, and hit back only when their committed force is smaller than your standing force.

A concrete operating sequence, so the three parts are not just a slogan: keep your standing force at a round, defensible number (130,000 in the example below), place the post one tile behind the border so its build finishes before the next contact, and write down the single trigger you will act on - “their committed force at home < my standing force.” Between probes you do nothing except let regeneration run; the post is doing the absorbing for you, and your force stays at the level where regeneration is fastest. Worked in numbers, against a 120,000 neighbor probing with 20,000: each probe lands on your 130,000, the post applies the 5x defender multiplier and the 3x slow, the probe spends itself, you close the gap over the thirty to fifty seconds, and their 120,000 baseline bleeds by the 40% floor every wave. By wave four or five their home force is under your front, and the trigger fires.

The version boundary matters because the attrition math is stable here. The v0.36.0 through v0.36.5 formal Releases changed maps, keybinds, clans, UI, FFA rating, and replays; they did not change the combat or balance code. So the numbers you will act on below - the 0.6 loss floor, the terrain multipliers, the defense-post cost and multiplier, the ~42% regeneration peak - are the same values the engine applies in a v0.36 match. You can trust them for this game series.

The single decision that separates winners from the bled-out here is refusing to leave max for a full counter. The grinder’s entire strategy depends on catching you mid-regeneration. If you hold a full standing force, every probe you absorb leaves you at, say, 95,000 of 130,000, and your regeneration is now running fast because you are below the peak. The post makes the next probe cost them more, so the gap they leave you closes before the wave arrives. The counter is only safe when, after they commit, their remaining force at home is smaller than yours on the front. Until that moment, holding is the correct, profitable move.

Why the small probe actually grinds you down

The mechanism is regeneration, not raw strength. OpenFront troop regeneration peaks near roughly 42% of your capacity and slows down toward both the very low end and the full end. A grinder who holds you at a partial force keeps your front starved: every time a probe strips 10,000 or 15,000 off your standing force, you spend the next thirty to fifty seconds recovering, and the next probe lands before you are whole again. You never sit long enough at a healthy level to rebuild the margin.

The second mechanism is the attack-loss floor. When the attacker fields roughly twice the defending force on a tile, the engine applies a 0.6 loss floor, which means the probe loses about 40% of its committed force on impact. At 2:1 and 3:1 the attacker still loses around 40%, so the floor is not an absolute shield - but it is not a free hit either. The grinder accepts that 40% because their committed force is small in absolute terms: 20,000 in, 8,000 lost, 12,000 still doing damage. The absolute number they bleed is tolerable to them and unbearable to you, because the loss you take is in regeneration time, not just troops.

The third mechanism is the defense multiplier. When your own Defense Post is near the tile being attacked, the defender-side combat magnitude is multiplied by 5 and the advance-speed term by 3. A probe that would strip 15,000 from an uncovered front now has to fight a 5x response and move 3x slower through the post’s radius. The attacker’s per-probe cost climbs sharply, and the more of their force they have to spend to get the same result, the less they can bank for the next wave. That is the whole game: raise the price of the probe until the gap between what they commit and what they can recover shrinks below your regeneration window.

The fourth mechanism is compounding, and it is why the fight resolves instead of stalling forever. Every wave they probe, they spend a bank of committed force and regenerate it at the slow end of their own curve, because a fresh commit leaves them near full for a moment and then starves as the probe is spent. You, by contrast, sit at max between waves and close each gap on the fast end of the curve. So the per-wave cost to them rises while your per-wave recovery is flat and fast: after enough waves, the gap you close each wave is larger than the damage they do each wave, and the attrition flips to your side even though you never initiated an attack. That compounding is what makes holding at max a winning posture rather than a delay.

The decision framework: hold, counter, or escalate

You need a named trigger for each of the three states, or you will default to the wrong one under pressure. Hold is the default: you are at or near max standing force, the post is in place, and their committed probe is smaller than your standing force. Holding is profitable because the 5x multiplier and the 0.6 floor make each probe expensive for them while your regeneration, running below the peak, closes the gap they leave. You hold even when it feels passive, because the patience is the weapon.

Counter fires only when their committed force at home drops below your standing force on the front. The grinder sends a probe, it bleeds out, and their home force dips below yours - that dip is the window. Counter into that dip with a force that is larger than their remaining force, not a force that matches it. If you counter with less than they have left, you are repeating their own logic against yourself and you are the one who gets ground down. The counter is a one-time reclaim of the tile and the margin, after which you return to holding at max.

Escalate fires when their committed force approaches or exceeds your standing force - the probe is no longer small, it is a full commit. At that point the 0.6 floor no longer protects you because the ratio is no longer attacker-favored, and holding becomes a slow loss. Escalation means either bringing a bigger force to the front (a banked boat, a held reserve) so the next probe meets a 2:1 defender advantage, or hitting their home with an attack that forces their standing force down. You escalate when the attrition ratio flips against you and holding at max no longer closes the gap.

When two triggers fire at once, counter outranks escalate: a clean dip where their committed force is below yours is the cheaper win, so take it before you commit a banked reserve. And size the counter by the dip, not by habit: if their home force dipped to 95,000 against your 130,000 front, a 110,000 counter force is larger than their remaining 95,000 and reclaims the tile and the margin; a 60,000 “just to push back” force is not, and leaves you at 70,000 where the next probe finds you starved again. The escalation decision has a hard exit: you escalate only while your standing force is still above the committed probe, because the moment you drop below it, the 0.6 floor no longer protects you and you have become the one being ground down.

Scenario A: equal-size grinder at 120k probing your 130k front

Assume you are at 130,000 standing and a neighbor at 120,000 starts probing with 20,000 troops every thirty to fifty seconds. They are slightly smaller, so their probe is a 20,000-of-120,000 commit that leaves them at 100,000 at home. Your standing force is 130,000. You are holding at max. Place a Defense Post on the front tile that their probe must cross, one step behind the border so it finishes its construction before the next contact.

Each 20,000 probe now fights a 5x defender response and a 3x slower advance through the post’s radius. Against your 130,000 standing force the probe is a small absolute number; even without the post it strips maybe 10,000 to 15,000 from your front, leaving you at 115,000 to 120,000. Your regeneration, now below the ~42% peak region, is running fast, and the thirty to fifty seconds to the next probe is enough to close that 10,000 to 15,000 gap. You hold. The probe gets more expensive for them every wave because the post is still there, and their 120,000 home baseline starts to erode as they bank less and less. After four or five absorbed probes, their home force dips below your 130,000 front, and that is the counter window: reclaim the tile with a force larger than their remaining force and return to holding.

Watch the bank across waves: their home force is 120,000, 118,000, 114,000, 109,000, 102,000 as the 40% floor and the post eat into it, while your front holds at 128,000 to 130,000 because the post absorbs most of each probe. The first wave where their home force drops below your 130,000 front is the counter window. Bank a small reserve the whole time - a 10,000 to 15,000 boat on the nearest sea lane - so the counter can be a combined landfall rather than a bare land push, which is safer and faster.

The assumption here is a single front and a slightly smaller neighbor. If the neighbor is larger, the probe is a bigger absolute number and the post is even more essential - see Scenario B. If you are in team play and can bank a boat, the counter can be a banked landfall rather than a land push, which is covered in the mode section. The numbers are from the v0.35 engine, stable through v0.36.x.

Scenario B: the grinder banks a larger wave and you are below them

Now the grinder is at 140,000, you at 90,000, and instead of 20,000 probes they bank a 60,000 wave. A 60,000 probe against your 90,000 standing force is no longer a small number; it is a 1:1.5 commit that the 0.6 floor no longer softens, because the ratio is not attacker-favored enough. Holding at 90,000 against a 60,000 wave means you lose the tile and bleed the regeneration on top. This is the escalation state.

The correct move is a controlled retreat behind the second post, not a last-stand. You have a second Defense Post one step deeper - you should have built it while the grinder was probing, because the probe phase is exactly when the 50,000 Gold is cheap relative to the value it will buy. Fall back to that line, let the 60,000 wave spend its force on the second post’s 5x/3x penalty, and do not counter on the retreat. You trade the front tile for regeneration time: behind the second post, at a lower standing force, your regeneration is running at peak speed, and the attacker’s 60,000 has been cut by the 0.6 floor and the post multiplier. After the wave breaks, their home force is down, your standing force has recovered to near max, and you now face the same hold-counter window as Scenario A, one tile deeper.

If you have no second post and no depth, the escalation is a banked boat: ship a force to land on their home or a valuable asset while the 60,000 wave is committed to you. In FFA there is no hard cap on attacking troops, so a banked landfall plus a land reserve can meet their next wave with a 2:1 defender advantage. The assumption: you have water access and the grinder does not have a warship on the same sea lane. If they do, the boat is at risk and the retreat must be deeper still.

How much depth you need is set by the wave size, not the map: a 60,000 wave against your 90,000 standing force, slowed 3x and cut by the post’s 5x on the second line, will break within one to two tiles of it, so one tile of depth between your front and the second post is enough to survive, but two tiles of depth is enough to survive and then counter. On a map with only one tile of room behind your front, the controlled retreat is not viable and the banked boat is the primary answer; on a map with a long border, the retreat is the primary answer and the boat is the reinforcement. Read the tile count behind your front before the first big wave lands, because by then the second post must already be built and the boat must already be banked - you cannot afford to improvise escalation in the same wave it is needed.

Failure modes and how the grinder answers each one

The most common failure is dropping to a partial force for a full counter. You see the probe, you decide to hit back hard, you commit 40,000, and you are now at 55,000 of 95,000 - starved. The next probe lands on the 55,000, not the 95,000, and the 0.6 floor no longer helps you because you are the smaller force. The grinder has done nothing different; you handed them the window. The fix is the counter trigger: only counter when their committed force at home is smaller than your standing force, and only with a force larger than that.

The second failure is panic spam-clicking to remove the probe pixel by pixel. Clicking fast empties your front, which keeps your regeneration at the slow end of the curve and ties your whole force to a single tile. The attacker keeps a much larger force at home while your entire front is committed to deleting one pixel. The fix is to let the attack play out against the post’s 5x response; the post is doing the deleting for you, and your force stays at max where the regeneration is fast.

The third failure is defending only at the capital, not at the front. A grinder who probes the capital instead of the border means you have no post on that route, so the probe meets no 5x/3x penalty and strips your standing force directly. The fix is to place the post on the route they actually use, not the tile you consider most important. The fourth failure is ignoring the banked-boat read: the grinder’s force at home says 50,000 but they have 30,000 on a boat. When the boat lands and they click again, the real attack is 80,000, not 50,000. Read the boats, not just the standing number, and hold your reserve for the landfall.

A fifth failure is escalating too early, out of frustration. You feel ground down, you commit the banked boat and the reserve to a big counter before their force has actually dipped below yours, and now you have spent your escalation on a wave that would have broken against your 130,000 front by itself. The counter is a one-time reclaim, not a catharsis; the moment you burn the reserve, the next big wave finds you at the low end of the curve with nothing left to escalate with. The discipline is to hold the reserve until the trigger fires - their committed force below your standing force - and then commit it fully.

Mode and map adjustments: FFA, team, and mountain maps

In FFA there is no hard cap on attacking troops, so the grinder can bank a boat, regenerate while it sails, and land a combined force larger than their standing population. The counter is to bank your own boat in response, so the landfall meets a 2:1 defender advantage instead of your depleted front. In team play the grinder can be fed: a teammate donates troops and the probe is larger than the probing player’s own population. The counter is to target the feeder, not the prober, because the probe keeps coming as long as the feeder stays whole. Check the alliances and the donation flow before you commit to a counter.

On a mountain map the defender is favored: the terrain defense-cost multiplier is 120 on mountain versus 80 on plains, roughly 1.5x, and the troop-kill cost scales 25 versus 16.5. A grinder probing across a mountain range pays more per probe than on open ground, so holding at max is even more profitable. The anti-synergy is that the grinder will route around the mountain by sea to a flat coastal tile or port, where the multiplier is back to 80. If your front is a mountain pass, expect the probe to move to a sea lane; if your front is a flat coast, the grinder has no terrain cover and the post is the whole answer.

The map also sets your escalation line. On a small map with few tiles, a controlled retreat behind the second post has only one or two tiles of depth, so the retreat buys little time and the banked boat is the main escalation. On a large map with a long border, the depth is real and the retreat is the primary answer. Read the border length before you commit to a retreat or a bank.

In FFA specifically, the warship is the force multiplier that breaks the boat-based counter: if the grinder keeps a warship on your sea lane, your banked boat is at risk and the landfall never comes, so the hold must be deeper and the second post must already be built. On coastal or port maps, a grinder probing by sea can land a boat on a port that gives them a second front you did not expect; the counter is to contest the port with a banked boat of your own, or to accept the port loss and fight only the land front at full depth. The map decides which of retreat, banked boat, or port contest is the primary escalation; the mode decides whether a second front is coming at all. Read both before you commit, because a retreat that assumes a single front fails badly the moment the boat lands. For the full combat calculation behind these numbers, see Land combat, and for the standing-force discipline that makes the hold work, see Sit at max troops.

The leverage table: numbers that decide the state

LeverThe grounded number (v0.35, stable through v0.36.x)The decision it drives
Regeneration curvePeaks near ~42% of capacity, slower below ~20% and above ~80%Hold at max so regeneration runs fast after each probe
Attack-loss floor0.6 floor at a 2:1 attacker advantage; a 2:1 or 3:1 probe loses ~40% of its forceA probe is not free for the attacker; the post makes it cost more
Defense post multiplierDefender-side attack power x5, advance speed x3 in the post’s radiusPlace the post on the route the probe must cross, one step behind the border
Defense post cost50,000 (tier 1), 150,000 (tier 2), 250,000 (tier 3) GoldBuild the second post during the cheap probe phase, not after the collapse
Terrain defense costPlains 80 / highland 100 / mountain 120 (kill cost 16.5 / 20 / 25)On mountain, holding is more profitable; expect the probe to route by sea
Banked-boat readNo hard cap on attacking troops in FFA; team teammates can donateRead the boats, not just the standing force; hold a reserve for the landfall
Escalation signalThe committed probe approaches or exceeds your standing forceHold is no longer profitable; bring a banked boat or hit their home
Counter triggerTheir committed force at home drops below your standing forceCounter with a force larger than their remaining force, then return to hold

For the structural defense that sits behind the post, see Outgunned defense, and when the grinder is one of several fronts, see Multi-front defense. How to use the table in the moment: the regeneration row tells you when it is safe to hold (your force at max, regeneration fast), the floor row tells you how much each probe costs them (40% at 2:1), the post row tells you where to place the multiplier (one tile behind the border on the probe route), the cost row tells you when to build the second post (during the cheap probe phase), the terrain row tells you whether the map favors the hold, the boat row tells you what to read before you commit, the escalation row tells you when holding stops being profitable, and the counter row tells you when the one-time reclaim fires. If a single row disagrees with your gut during a match, trust the row; the gut is what the grinder is counting on.

The table is the reference; the sections above are the reasoning.

Related content

Continue reading
OpenFront AFK Teammate Takeover: Absorb a Disconnected Ally for Zero Troop Loss

A Team-mode decision guide for what a teammate's disconnect does to your win/loss position, when absorbing their land is free, when it is a trap, and what the 30-second mark and the ranked 2v2 rules actually change.