Should you sit at max troops — and when the buffer wins

In the v0.34.24 build, the old way of parking troops behind a non-bordering attack that quietly rebuilt them past your cap is closed. That means the only two live options left are the two you are weighing in this thread: hold your standing army at the full cap, or keep a deliberate buffer below it. Hold at the cap when you are the side that must field the largest army and can afford to spend troops on the frontline; keep a buffer when the economy, the regen curve, or your opponent’s timing rewards staying out of the regen-dead zone. The cap itself is set by maxTroops, the regen rate is a single smooth curve that peaks near 42 percent of cap, and the patch is a retreat rule in AttackExecution — all of which is what makes the buffer-versus-cap trade real again instead of a fixed answer.

How the v0.34.24 patch changed the answer

The reason this question needs a fresh answer is that the version of the game you are playing right now is not the one most of the Reddit threads and YouTube videos were written against. In the build before v0.34.24, an attack did not require the target to be adjacent to land you already controlled. That single fact is what powered the exploit this whole guide is about: you could launch an attack at a tile you did not border, and if there were no tiles left for that attack to actually conquer, the troops were refunded and the attack effectively did nothing — but the refund cycle, repeated fast enough, let a player hold troops at the point of maximum regeneration and, in the worst case, push the total above the cap.

The fix landed in src/core/execution/AttackExecution.ts. The relevant guard now reads, in effect, “if the set of conquerable tiles is empty, the attack retreats.” When toConquer.size() === 0, the attack returns to base and never lands, so the refund-and-refill loop that let you farm troops at peak regen is no longer available. The team shipped a dedicated test in tests/Attack.test.ts to lock this behavior in: it sets up an attack, spams it across 600 ticks, and asserts that the total troop count never climbs above maxTroops. The test passes in v0.34.24.

The practical consequence is that the buffer-versus-cap question is no longer a “pick the bug” question. It is a genuine decision between two real postures, and it is the decision this guide exists to settle. One posture — hold at the cap — maximizes the standing army you present on every border and accepts the slowest possible refill at the very top of the pool. The other posture — keep a deliberate buffer — gives up a slice of standing army in exchange for sitting on the rising side of the regen curve, where any troops you lose come back faster than they would at the top. Neither posture is universally right; which one wins depends on the economy, the map, the mode, and your opponent’s timing, and the rest of the guide works through exactly those variables.

What that means for you as a player is that the “free extra troops” lever is gone. You can no longer bank troops above your cap by spamming a non-bordering attack, and you can no longer sit at a deliberately high fraction of cap and let a no-op attack hold them there for free. Your troop pool is now exactly what the cap and the regen curve give you, and it moves in real time. Every decision in this guide — hold at cap, keep a buffer, when to spend, when to pull back — is a decision about the cap and the curve under those fixed rules, not about a bug. If you came to the game during the exploit window, some of the habits you picked up there no longer apply, and the next section gives you the exact numbers so you can rebuild the intuition on solid ground.

The exact troop cap and regen math you need to hold a line

Two numbers from src/core/configuration/Config.ts control everything you do with troops, and both are worth knowing because they change how fast you can recover from any mistake.

The first is the cap itself. maxTroops() computes a base value from your territory: roughly 2 × (tilesOwned^0.6 × 1000 + 50000), then adds 250,000 per completed City level. A few important modifiers follow from the same function: bots get only a third of the standard cap; the AI Nation’s effective cap is scaled by difficulty, from half the standard value on Easy up to a little above the standard value on Impossible; and in team games each team’s cap is the sum of its members’ caps. This last point is why team games feel different from the solo ladder: the cap is no longer a personal ceiling but a shared pool, which changes when it is worth holding at the full value.

The second is the regen rate. troopIncreaseRate() is a single smooth curve: 10 + troops^0.73 / 4, multiplied by (1 − troops/maxTroops). There is no worker branch, no separate multiplier, and no cap at 100. That last detail matters because older community write-ups described a different formula, and if you are still planning around a flat or worker-scaled regen you will mis-time your buffers. The curve has two properties that drive the entire strategy. First, the (1 − troops/maxTroops) term means the closer you are to your cap, the slower you refill: at the cap the regen is exactly zero, which is why a player sitting at 100 percent of cap is the slowest to recover from any hit. Second, the troops^0.73 term means that at very low troop counts the absolute regen is dominated by the flat +10, so a nearly-depleted pool refills faster in raw terms than a pool sitting just under the cap. The two effects together produce a single peak somewhere in the middle, and the next section pins down exactly where.

The two properties above produce the full shape of the curve, and the table below pins it down against a 120,000-troop cap so the numbers are concrete. The “regen at that fraction” column is the absolute troops gained per tick while holding at that fraction of cap, and the “side of the peak” column tells you whether you are on the fast-rising side or the slow-falling side:

Fraction of cap	Standing army (of 120,000)	Regen at that fraction	Side of the 42% peak	What it buys you
5%	6,000	fastest (flat +10 dominated)	far below	fastest recovery, smallest army
20%	24,000	rising	below	fast recovery, small army
42%	~50,400	peak	at the peak	best single refill tick, large army
60%	72,000	falling but still strong	above	large army, good refill
80%	96,000	falling fast	above	very large army, slow refill
100%	120,000	zero	at the top	largest army, no refill until you spend

Read the table as the whole trade in one glance: the left column is how fast you recover, the middle is how big your standing army is, and the two trade against each other with the 42% row as the hinge. Holding at 100% gives you the biggest army and the slowest (zero) recovery; holding at the 42% peak gives you the fastest single refill tick; and everything in between is a deliberate blend of the two. That table is the entire buffer-versus-cap decision compressed into six rows, and every section that follows is a way of choosing which row you want to be on given your specific match.

The practical reading is that “how fast I recover” is not a constant — it is a function of where you are sitting right now. Holding at 100 percent buys you the largest standing army but costs you the slowest possible recovery. Holding at, say, 60 percent buys you a smaller army but a materially faster refill when the frontline swings. The buffer decision is, at bottom, a decision about where on this curve you want to be, and the numbers above are what let you make that call with real values instead of a gut feel.

What the 42 percent peak actually does to your reserve

If you plot the regen curve against the fraction of your cap you are holding, it rises from the flat +10 baseline at very low troop levels, climbs to a single maximum, and then falls back to zero as you approach the cap. The maximum sits at roughly 42 percent of your maximum troops. This is not a rounding artifact and it is not an estimate a video guessed — it is the argmax of the curve the game actually runs, and it was independently confirmed by both the source in v0.34.24 and the community write-up that called the “42 percent troop economy” a win condition.

This single number is the heart of the whole decision. Below 42 percent of cap you are on the rising side: every troop you lose and then regenerate gives you back regen that is faster than you had before, so a low reserve is a self-accelerating recovery. Above 42 percent of cap you are on the falling side: as you hold more troops you get a larger army but a slower and slower refill, and at the very top the refill is zero. The 42 percent point is the boundary between “losing troops buys me back faster” and “holding troops costs me recovery speed.”

The way to use this in a match is as a target for your reserve, not as a fixed number. If your goal is to keep a buffer and recover quickly, you are usually trying to keep your standing army somewhere on the rising side of the curve — below the 42 percent peak when you want fast recovery, and up near it when you want the best single regen tick. If your goal is to hold at the cap, you are deliberately accepting the falling side all the way down to zero regen, trading recovery speed for the largest possible standing army. The exploit that made this less of a trade is gone, so the falling side is now real: sitting at the cap genuinely means a zero-regen pool until you spend troops, and that cost is what makes the buffer a legitimate answer instead of a mistake.

There is a useful way to frame the entire guide in one line: the peak at 42 percent is the hinge, holding at the cap is buying standing army at the price of the slowest recovery, and the buffer is buying recovery speed at the price of a smaller army. Every scenario, failure mode, and counter in the rest of the guide is just that trade played out against a specific map, mode, and opponent. The how to time your attacks to the troop bar guide pairs directly with this one, because the moment you decide to spend is exactly where the regen curve and the attack-timing curve meet.

Why “sit at max troops” is the dominant default

Most of the top answers in the Reddit threads you are reading — including the one asking why people sit at max troops during an invasion — converge on the same default, and it is worth stating the default clearly before you look for the exceptions. In the majority of matches, the best default is to build to and hold near your cap, and to spend from the cap rather than from a permanent buffer. The reasons are structural, not stylistic.

First, the cap is the number the game measures your pressure by. An army at or near the cap is a standing threat on every tile you border, and the pressure it applies forces your opponent to allocate defense they otherwise could use elsewhere. Holding at the cap is a way of making your entire frontline a continuous cost for the opponent to pay, and that pressure has value even when you are not actively attacking.

Second, the regen curve, while it falls on the upper side, is still fast enough that a cap-holding player who spends a meaningful chunk of troops recovers to useful levels quickly. The zero-regen state is only the exact moment you are at the very top; the instant you commit troops to an attack or an invasion, you drop off the top and the refill begins in earnest. So the “slowest recovery at the cap” problem is largely a non-issue for a player who is actively spending troops, which is what a cap-holding player is doing.

Third, and this is the point the invasion thread is really making, sitting at the cap during an invasion is a defensive statement: it tells your opponent that you can absorb their push and still refill, which deters the exact timing of the invasion they were hoping for. If your standing army is at the cap and they commit to a push, they are walking into a pool that will refill the moment their push spends troops, and the attacker’s own regen on the falling side works against them. The attacker’s regrow advantage after a drain guide explains the offensive mirror of this, but the defensive logic here is the same: the cap-holding side is usually the side that is not bleeding regen.

The fourth and final reason is the most underrated one, and it is the reason the buffer is a real exception rather than the default: the cap-holding army is a continuous, visible signal, and visible signals are read by humans, not just by the game. When your opponent can see that you are holding near the cap, they are making their own buffer-versus-cap decision against the information that you are the larger standing army, and most players respond to that signal by over-allocating defense on the border you are pressing, which is exactly the over-allocation the cap-holding player is trying to force. The cap-holding army is a pressure you apply by existing, and that pressure is worth real value even on the tiles where you never actually commit troops.

The default is not unconditional, though, and the next section is entirely about the situations where the buffer beats the cap-holding default.

When the buffer is the right play instead

The buffer wins in a specific set of situations, and you should reach for it when one of them is true rather than out of habit. The first and most common is when you are the slower economy in the match: if your opponent is refilling their frontline faster than you are, holding at the cap means you are always chasing their timing, and a deliberate buffer that keeps you on the rising side of the regen curve lets you recover faster than your opponent’s frontline can bleed. In that match the buffer is not a smaller army, it is a faster one in the places that matter.

The second is when you are playing to outlast and out-regenerate a single long push rather than to win the timing contest. If your win condition is “make their attack spend more troops than they can refill,” you want to be on the rising side of the curve at the moment their attack lands, because that is the moment you can give back troops the fastest. Holding a buffer at or just below the 42 percent peak is exactly that posture, and it is the posture the “sending troops to the frontlines is criminally [bad]” thread is really defending for that kind of match.

The third is when the map or mode punishes a standing frontline. On some maps the frontline is long and exposed, so a cap-holding army is a long row of troops that your opponent can pick off tile by tile; on maps where the frontline is short and the rear is safe, the same cap-holding army is cheap to defend. When the frontline is expensive, a buffer that keeps your army in the safe rear and only commits on the rising side of the regen curve is strictly better than parking the whole pool at the exposed front. The how to place Workers guide is the natural companion here, because where you place your Workers changes where your cap-holding army can safely stand.

The fourth is the team-game case, where the cap is the team’s shared pool. If you are holding your personal share at the cap while your teammates are the ones actually fighting, you are locking your share of the shared pool at zero regen, which costs the team recovery speed it needs on the tiles where the fight is happening. In team games the buffer is often the better contribution to the shared pool, and the team game pressure guide covers how the Nation’s shared pressure changes this further. These four situations — slower economy, outlast-the-push, expensive frontline, and shared team pool — are the full set of cases where the buffer beats the cap-holding default, and the scenarios in the next section put numbers on the first two of them.

The exploit that let people hold troops past cap — and why it is gone

Because this guide is titled around a decision that only became a real decision after a patch, it is worth being precise about what the exploit actually was, who it helped, and exactly which line of code closed it. The exploit is documented in the source and in the new test, and the honest summary is that it was a small but real asymmetry in the pre-patch build.

The mechanism was a consequence of attacks not requiring an adjacent tile. In the old build you could queue an attack at a tile you did not border, and if that attack ended up with no conquerable target, the troops returned to you and the attack effectively did nothing on the map. But the return-and-refill cycle was not free time — it was regen time at whatever fraction of cap you were holding. The players who understood this found that by holding their standing army at a high fraction of cap and spamming no-op attacks, they kept the regen running at peak while the no-op attacks gave them a way to manage the pool. In the worst cases the refund logic let the total climb above the cap entirely, which is why the thread “how do you suddenly increase your pop/troops” is asking about a number that, under the fixed rules, is not supposed to be reachable at all.

The v0.34.24 fix removed the premise. The guard in AttackExecution now checks whether the set of conquerable tiles is empty before the attack commits, and when it is empty the attack retreats instead of landing. That one guard kills the no-op refund loop, because there is no longer a “the attack did nothing but I still cycled my troops” path. The new tests/Attack.test.ts test is the proof of closure: it spams an attack for 600 ticks under the fixed rules and asserts the total never exceeds maxTroops, and it passes. The 42 percent regen peak is unchanged by the patch — the curve is the same — but the ability to farm the pool at the peak by spamming no-ops is what disappeared, and that is precisely why the buffer-versus-cap trade became a live decision again.

The practical takeaway is simple and important: if you built your habits around the exploit, the habits are now dead weight. A player who was “holding at 100 percent by spamming a non-bordering attack” is, under v0.34.24, just sitting at the cap with zero regen and a smaller effective army than they thought they had. The correct move after the patch is to re-derive your posture from the cap and the curve, which is the whole point of the sections that follow. The population growth and cap formula guide is the stable reference for the cap number itself, and everything else in this guide is the decision you make on top of it once the exploit is gone.

Worked scenarios with numbers

Abstract trade-offs are easy to nod at and hard to act on, so here are two scenarios with concrete numbers that you can replay against your own cap. Use your own territory number in place of the placeholder where it says “your cap.”

Scenario one: you are the slower economy and you are being pushed. Assume your cap is 120,000 troops and your opponent is spending 30,000 troops against your frontline every cycle while refilling about 40 percent as fast as you do. If you hold at the cap, you start each cycle at 120,000 and get hit to 90,000; because you are now on the falling side of the regen curve, your refill is slower than theirs, so by the next cycle you are starting lower than you ended. You are bleeding net troops into their attack and losing the timing contest. Now run the same match with a buffer: hold at 42 percent of cap, which is about 50,400 troops, and only commit from there. When their 30,000 hits, you drop to 20,400, which is back on the rising side of the curve, and your refill for the next cycle is faster than theirs. You give back the 30,000 faster than they can keep up, and the net bleed reverses. The number that decided the match is the 42 percent peak: holding at the peak is the highest reserve that still keeps you on the fast side of regen, and that is why the buffer at the peak beats the cap in a slower-economy push.

Scenario two: you are the faster economy and you are invading. Assume the same 120,000 cap, but now you are the one spending 30,000 against a static defense every cycle. Here the cap-holding default is the right call, and the reason is the falling side working for you instead of against you. You hold at 120,000, spend down to 90,000, and refill from the upper side; because you are the faster economy, your refill outruns their defense’s refill, so you keep pushing while they keep dropping. If you had instead held a buffer at the peak, you would have started each cycle with a smaller standing army and a faster refill, but in a match where you are already the faster economy the extra standing army at the cap is worth more than the extra refill speed, because the cap-holding army is the pressure that forces their defense to over-allocate. The number that decided this match is the same 42 percent peak, read the other way: when you are the faster economy, accepting the slower refill on the falling side in exchange for the larger standing army is the better trade. These two scenarios are the same curve and the same peak; the only thing that flips the answer is which side of the economy you are on, and that is the whole decision in one line.

Failure modes and the counter

The two strategies fail in predictable ways, and knowing the failure mode is what tells you when to switch. The cap-holding strategy fails when you are the slower economy, when the frontline is long and exposed, or when your opponent is playing to outlast a single push — in all three of those cases you are on the falling side of the regen curve at the moment you need to recover, and you lose net troops into the hit. The counter to each of those is the buffer: drop to or below the 42 percent peak so that the moment you are hit you are back on the rising side, and the refill outruns the hit. The specific counter depends on which failure you are in — a slower-economy buffer sits near the peak, an outlast-the-push buffer sits just below it, and an expensive-frontline buffer sits in the safe rear and only commits on the rising side — but the direction is always the same: get off the falling side and onto the rising side before the hit lands.

The buffer strategy fails in the opposite direction, and it fails when you are the faster economy, when the frontline is short and the rear is safe, or when your opponent is not actually outlast-you but is instead out-pressuring you with a larger standing army. In those cases your smaller reserve is not saving you, it is just a smaller army that never reaches the pressure threshold that would make your opponent over-allocate. The counter is to stop holding a permanent buffer and start holding at the cap, because the larger standing army is the pressure that flips the match to your side. The failure mode here is the subtle one, because a buffer feels “safe” and “disciplined” and is easy to keep holding long after it has stopped being the better play; the signal that you have over-buffered is that your frontline is not applying enough pressure to force the opponent’s hand, and the fix is to give the buffer back to the pool and hold at the cap.

There is also a third failure mode that affects both strategies, and it is the one the exploit patch is most relevant to: failing to update your mental model after the version changed. Players who held at the cap by spamming a no-op attack were, before the patch, holding at an effective army that was larger than the cap and refilling at peak; after the patch the same behavior is just holding at the cap with zero regen and a smaller effective army, and nothing on the screen tells you that the behavior changed. The counter is to re-read the numbers every time the version changes — the cap formula, the regen curve, and the 42 percent peak — and to re-derive your posture from those numbers instead of from muscle memory, because the muscle memory is from a build where a now-dead lever was available. The attacker’s regrow advantage guide is the offensive version of the same mental-model update, and it is worth reading alongside this one so the full offensive-and-defensive picture is current.

Mode and map adjustments

The same cap, curve, and peak apply in every mode and on every map, but the mode and the map change which side of the trade is usually the better one, so it is worth a short pass over each. In the standard solo ladder the cap-holding default is the most common right answer, because the match is a timing contest between two players of roughly equal economy and the cap-holding army is the pressure that wins the contest; the buffer is the right exception when you are clearly the slower economy or the map’s frontline is long and exposed. In the team game the shared-pool rule changes the whole picture: your personal cap-holding decision is a contribution to the team’s total pool, so holding your personal share at the cap can be a net cost to the team if your teammates are the ones fighting, and the team game pressure guide is the right companion for that mode because the Nation’s shared pressure further shifts the balance toward the buffer.

The Nation difficulty setting changes the effective cap of the AI, which changes the pressure you are defending against and therefore the right default. On Easy the Nation’s cap is half the standard value, so the pressure it applies is light and the cap-holding default is very safe; on Impossible the Nation’s effective cap is a little above the standard value, so the pressure is heavier and the buffer is more often the right answer because you need the faster refill to survive the Nation’s bigger pool. The how Nations put you under pressure guide is the direct reference for this, and it is where the numbers for the Nation’s cap and pressure come from. On the map side the rule is simple: short safe frontlines favor holding at the cap, long exposed frontlines favor the buffer, and maps with a lot of water favor the buffer on defense and the cap on offense, because the water changes where the frontline is exposed and where the rear is safe. The map and water choice logic in the adjacent guide covers the water side of this, and the troop bar attack timing guide covers the moment you commit from whichever posture you have chosen.

The one adjustment that applies to every mode and every map is the same: the 42 percent peak is the hinge, and the mode and map only change which side of the hinge you usually want to be on. If the mode and map make you the faster, more-pressuring side, hold at the cap and accept the falling-side refill. If they make you the slower, more-defensive side, keep a buffer at or just below the peak and take the rising-side refill. The mode and map never change the curve or the peak, they only change which side of the trade is the better one, which is the cleanest possible summary of the whole section and, in a sense, the whole guide.

How this combines with your other troop decisions

This guide is one node in a small web of troop decisions, and it is worth stating the edges clearly so you do not apply the buffer-versus-cap answer in isolation. The most direct edge is the attack-timing decision: the moment you decide to spend troops from whichever posture you are holding is exactly where the regen curve and the attack-timing curve meet, and the how to time your attacks to the troop bar guide is the other half of that edge. Holding at the cap and attacking from the top is a different timing problem than holding at the buffer and attacking from the rising side, and the two guides are only consistent when you read them together.

The second edge is the offensive regrow decision. When you are the attacker and you have just drained the opponent’s frontline, your own refill on the falling side of the curve is what determines whether you can keep pushing, and the attacker’s regrow advantage after a drain guide is the offensive mirror of this guide’s defensive logic. The two guides are the same curve from opposite sides of the map, and the buffer-versus-cap answer in this guide is the defensive half of the same trade that the regrow-advantage guide covers offensively. The third edge is the cap and Workers decision: where you place your Workers changes both the cap number and where your standing army can safely stand, and the population growth and cap formula guide is the stable reference for that, because the cap number and the Worker placement are the inputs to every decision in this guide.

The fourth edge is the mode-specific pressure, and the how Nations put you under pressure guide is the reference for the team-game and Nation-difficulty shifts that change which side of the trade is usually the better one. Read the four guides together and the full picture is: the cap and Workers set your pool, the mode and map set which side of the trade is better, the regen curve and the 42 percent peak set the hinge, and the buffer-versus-cap decision in this guide is the single most important call you make on top of all of that. The exploit patch is what made this particular call live again, because it removed the now-dead lever that had been quietly answering it for you, and the four edges above are the rest of the decision web you need to keep current so the answer stays right as the version and the meta move.