Public FFA Midgame: Hold, Expand, or Ally With the Stronger Neighbor

Direct answer: hold until you have a second front, then expand

In a public FFA, the midgame is the phase where your first economy is built, your borders touch several neighbors, and the player who starts strongest usually looks unbeatable. The single decision that separates a player who survives that phase from one who is folded is not which border to attack first; it is whether you are allowed to attack at all right now. Default to holding: keep your core intact, convert ports and factories into Gold, and let your troop bar grow toward the headroom peak. Hold is not a retreat and not a fear response; it is the only posture that keeps a second front open. You expand only when a specific, visible condition appears: a neighbor commits troops away from the border you face, a frontier Nation or neutral pocket is cheaper to take than the strongest human border, or your own troop density now lets you push a second front without starving the first. You ally only when the neighbor you would otherwise be fighting is weaker than a third player, and the alliance is short, because the default alliance lasts five minutes and any lobby can set it from one to fifteen minutes or disable it entirely.

Run this as a three-question loop each time the board changes. First, who is the strongest human I currently face, and is that player actually moving troops at my border or sitting and growing? Second, what is the cheapest land I can take right now, the Nation or neutral pocket whose border gives back the most land for the least exposed push, and what is it worth compared to the strongest human? Third, do I have a second front, meaning a route to new land that does not depend on the same border, the same Port, or the same exposed City as my primary direction? If the answer to the third question is no, you hold and grow, because a single corridor is not a midgame strategy, it is a countdown to being folded. If the answer is yes and the second question has a real target, you expand with a bounded push and a named stop signal. If the strongest human is about to be answered by someone else, you ally and turn your attention to the pocket the alliance frees. The whole midgame is just repeating that loop with current numbers, not with the plan you made at spawn.

The reason the default is to hold is the growth math. Your troop growth is roughly (10 + troops^0.73 / 4) x (1 - troops / maxTroops) per tick, with about ten calculations per second, which means a mid-sized bar actually grows faster than a tiny one and slower than a bar near the cap. If you commit your whole bar to a border and lose most of it, you do not just lose those troops; you drop back to a part of the curve where regrowth is slower than it was. Holding near the efficient part of the curve is what funds both the defense and the next push. Expansion is a bet that the land you gain will grow your cap faster than the troops you spent regrow. That bet is usually worth it in the midgame, but only when the land is real, defensible, and cheaper than the alternative target, and only when you keep enough of the bar to answer the neighbor who is already strongest. The rest of this page is the evidence for each part of that default and the tests that tell you when to break it.

Hold: the growth curve is a budget, not a goal

Holding in the midgame means choosing not to spend a push your growth cannot cleanly fund. The troop cap and the growth formula are the budget you hold inside. The same formula the population guide uses applies here: growth per tick is (10 + troops^0.73 / 4) x (1 - troops / maxTroops), clamped so a single tick cannot push you above the maximum, with a normal tick at one hundred milliseconds and about ten growth steps per second. The first term rises with your current troops, and the final headroom term falls as the bar fills, so the curve has a broad peak rather than a rule that lower is always faster. In the reference example the mathematical peak sits close to forty-two percent of the cap, and that peak is the number to keep in mind: it is the share of your cap where you add the most troops per second.

The practical consequence is that an all-in push is expensive in a way the raw number does not show. Suppose your cap is one hundred thousand and your bar sits at forty-five thousand, near the peak. A push that loses half the bar leaves you at twenty-two500, which is still a reasonable number, but the shape of your front has changed: the land you were defending is now exposed, the neighbor who attacked has committed and may follow, and your next push has to leave the new, thinner bar and cross a longer approach. If you instead hold that forty-five thousand, you keep the front packed, you keep the efficient regrowth, and you keep the option to spend a bounded second push on the exact border where a neighbor just created an opening. Holding is therefore not passive; it is the act of staying at the top of your growth curve while you wait for a target that pays. The failure this prevents is the classic midgame fold: a player who spends the entire bar to take one Nation, then cannot answer the second human who was always stronger, and ends the game at zero.

Holding also has a clear end, or it becomes a death by attrition. You keep holding only while at least one of three things is true: a real target is forming, a neighbor is spending troops elsewhere and their own border thins, or your cap is still rising fast enough that the time you buy is worth the land you will eventually take. If none of those is true, and the strongest human is simply growing while your own bar has stopped rising because your economy is maxed, holding is no longer holding, it is standing still. At that point the midgame has quietly become the lategame and the correct move is the kill push the overtime and endgame pages own. The line between holding and stalling is a slope: are your numbers improving or flat? If they are improving, hold is correct. If they are flat, you have run out of midgame and need to convert your position into a decision.

A useful way to see whether you are holding correctly is to track the ratio of your current troops to your cap. If that ratio is drifting down while you hold, you are spending more than you grow and you are leaking, so either stop spending or find cheaper land. If it is drifting up, you are growing faster than you spend, which is exactly the posture that makes the next push cheaper. If it is flat near the peak, you are at the most efficient point and you should be actively hunting for the second front rather than idling. The bar is a number, but the ratio to the cap is the decision signal. Players who lose the midgame are almost always looking at the absolute troop count, which can look reassuring while the ratio quietly collapses, and then the strongest neighbor attacks into a bar that is large in absolute terms but thin relative to what it was. Hold is the discipline of watching the ratio and spending only when the ratio can survive the push and the follow-up.

Expand: the cheapest land first, the strongest human last

Expansion in the midgame is not about taking the most impressive border; it is about taking the border that gives the most land for the least exposed push, and the strongest human is rarely that border. The reference attack-ratio framework exists precisely so you do not spend a full push on the wrong target. Routine land against a weaker or distracted neighbor is pushed at twenty-five to forty percent of your bar with a double-click when your growth funds the second click, because that range lands a push at least as large as the defender can field while leaving troops behind the line for the follow-up. The strongest human, by contrast, is the target you push at parity or with a full kill attempt and only when the land is worth more than the army you accept to lose. If you push the strongest human with a casual forty percent and they answer with their own growth, you have spent your cheapest push on the most expensive target and you have nothing left for the pocket that was actually free.

The ordering principle is to take the cheap land first and let the strong human deal with each other. In a public FFA the board is wide, and the neighbor who is strongest is often also the neighbor other humans are trying to answer. If two strong players are trading, the cheapest land for you is frequently a Nation or neutral pocket that neither of them is watching, and taking it while they are busy is the single most reliable midgame move. The community voices this directly: a player who has lost almost every FFA for a month describes winning at last by building troops while other players destroyed each other, and only then taking the board once the fighting was over. That is not a coincidence and not a lucky spawn; it is the consequence of refusing to spend your midgame on the strongest target and instead holding until the cheap land became free. The same thread’s advice to wait for about half population before pushing tribal territory and to send thirds to a land grab rather than going all-in is the same principle stated as a rule: do not spend the expensive push on a target that does not yet reward it.

A bounded push is the unit of expansion. A bounded push names the target, the force, the follow-up, and the stop signal before it is sent. It is not a fixed percentage of troops; it is a commitment that ends when the land is secured or the stop signal fires. The stop signal for an expansion into frontier land is one of three things: the pocket is taken and the next target is still open, in which case you reassess; the approach to the pocket is closed by a neighbor moving onto the neck, in which case you hold the land you already have and stop; or your bar drops below the level that can answer the strongest human, in which case the expansion itself was the mistake and you retreat the remaining troops. The point of the stop signal is that the push has an ending you can observe on the map, not a feeling that the map will sort itself out. Players who lose the midgame usually lose to a push with no stop signal: they spend until the bar is gone and then the board reveals that the land they took exposed a border they can no longer defend, and the strongest human was still standing the whole time.

The table below is a comparison instrument for choosing which land to expand into. It describes live-board shapes, not fixed map positions, and no row promises a specific Nation, Port, or target exists in your lobby. Record the actual target and the actual stop signal beside each row before you act.

Read the table as a choice of ordering, not a choice of percentage. The strongest human is always the last column’s row; the first two rows are where the midgame is actually won. If all three rows look equally attractive, take the one whose failure you can detect and reverse before your reserve is gone. The reserve, the troops you hold back specifically to answer the strongest human, is the number that makes the table work: every row spends against it, and the stop signal is the moment a row spends too much of it. Expansion without a reserve is just a slow way to hand the strongest neighbor the land they will take anyway.

Ally: use the shortest treaty to buy the second front

Allying in the midgame is not a social decision; it is a timing decision, and the default treaty is short. The game’s default alliance lasts five minutes, and a lobby host can set it anywhere from one to fifteen minutes or disable alliances entirely. That means an alliance is a loan of attention, not a permanent border. You do not ally because you trust a neighbor; you ally because for the next few minutes that neighbor is not going to spend troops on your border, and that loaned attention is what lets you take the cheap land or build the second front that would be impossible while the neighbor is watching you. The mistake is to treat the alliance as a reason to relax the second front; the alliance ends on a timer, and when it ends the neighbor is the exact target you should have been keeping a reserve against.

The decision to ally is the third question from the direct answer: is the neighbor you would otherwise be fighting weaker than a third player? If the strongest human in the match is facing another strong human, allying with the one you would not want to fight is usually the right move, because it redirects the strongest attacker away from you and onto the player who was already going to be answered. If the strongest human is facing you, allying with them is rarely useful, because the alliance does not stop them from growing and it does cost you the troops you would have used to hold, and the alliance ends in five minutes anyway. If two humans are both strong and both facing you, the alliance you want is the one that makes them face each other, and the one you want to avoid is the one that makes them face you together. This is why the alliance is a second-front tool: it is the cheapest way to remove one border from your problem list so you can focus on the second front, and it is the most expensive way to remove the wrong border.

There is a real cost to allying that the short duration hides. While the alliance is active you stop spending troops on the ally’s border, which means the land between you and the ally stops being contested, which means when the alliance ends the border is exactly as exposed as the day you made it, except now the ally has been growing during the time you were not watching. The default five-minute alliance is long enough for a neighbor to gain a full growth cycle or two, and that gain is real troops that now face your unwatched border. The community reflects this: the same FFA player who won by letting others destroy each other also describes the frustration of a neighbor who pushes all game while you are trying to hold, and the alliance is how you stop being the unwatched border. The fix is to treat the alliance as the minimum duration that buys your specific move: if you need the neighbor to stand down for two minutes to take the pocket, do not let a five-minute alliance make you relax for three more minutes, and do not expect to renew it. Renewal is a new decision against a neighbor who has grown.

The stop signal for an alliance is the stop signal for the move it was buying. If the alliance was made to take a specific pocket, the alliance ends when the pocket is taken or when the pocket is no longer reachable. If the alliance was made to hold the strongest human at bay while you grow, the alliance ends when your bar has grown enough to hold them without the loan, or when they have grown past the point where the loan can hold them. Do not let the alliance outlive the move that justified it, because the cost of the loan continues to accrue even after the benefit is gone, and the neighbor who is now facing your unwatched border is the same neighbor who was facing someone else three minutes ago. A short, specific, one-shot alliance that buys one second front is the best use of the treaty in the midgame. A long, vague, evergreen alliance that you renew by habit is how players who “have an ally” still lose, because the ally is not a second front, the second front is the land you took while the ally was not watching, and the land is only worth what you can hold when the timer runs out.

Scenario A: the stronger neighbor is growing while you are flat

Assume a public FFA on a wide map with thirty to fifty humans, your core established, a Port or two and a Factory or two built, and a neighbor to the north who has been steadily outgrowing you for the last several minutes. You are not losing the border; you are just not winning it. Your troop bar is near the efficient part of the curve, the strongest human to the north has a bar visibly larger than yours, and a Nation and a neutral pocket sit between you and the rest of the board. These are scenario assumptions, not a fixed official rotation, and the current board in front of you decides ownership and legality. The question is whether to hold, expand into the Nation or pocket, or ally with the northern human.

Run the three questions. Who is the strongest human you face? The northern human, and they are sitting and growing, not pushing. What is the cheapest land you can take right now? The neutral pocket, which gives back land with a bounded push, versus the Nation, which answers, versus the northern human, which is the most expensive target and not worth a casual push. Do you have a second front? The pocket is one; the northern human’s border is the other, but it is the most expensive. The answer to the third question is yes but weak, because the pocket does not depend on the northern border and the Nation does. The correct move is to expand into the pocket with a bounded push and a named reserve, and to keep the reserve specifically sized to answer the northern human if they push while you are away.

Send the bounded push at a second-third of your bar toward the pocket, because the pocket is a weaker or unowned target and the twenty-five to forty percent range is the right push. Name the stop signal: the pocket is taken and the next target is open, or the neck to the pocket is closed by another human, or your bar drops below the reserve that answers the northern human. If the northern human pushes while you are committed to the pocket, do not abandon the pocket to chase them with your full bar; hold the land you already have and let the reserve answer, because abandoning the pocket to chase a full-send is the failure with the worst math in the whole midgame. If the pocket is taken and the next target is still open, reassess with the new numbers: the pocket raised your cap and your bar, the northern human is still growing, and the cheapest target may now be the Nation that the pocket made reachable. The scenario ends correctly when you have a held pocket, a recorded reserve, a verified next target, and a plan for the northern human, and it ends wrong when you spent the whole bar on the pocket and the northern human is now the only human left on the board.

The alternative, allying with the northern human, is the move you make only if the pocket is not reachable or the reserve cannot hold. If you ally, the alliance is the shortest one that buys the pocket: you need the northern human to stand down for the two or three minutes it takes to take the pocket, not for the five-minute default. When the pocket is taken, the alliance has done its job and you should be re-reading the border, because the northern human has grown during the minutes they were not watching you. If the pocket is not reachable, the alliance is not a fix, it is a delay, and the real problem is that your second front does not exist, which is a hold problem, not an ally problem. The lesson of this scenario is that the stronger neighbor is a timer, not a wall: you do not have to beat them, you have to outlast them until the cheap land becomes free, and the second front is the thing that makes the timer work for you instead of against you.

Scenario B: you are the stronger neighbor and a human is probing your border

Assume the mirror of the previous scenario: you are the player with the larger bar, a human to the south has been making small probes at your border, and you have a Factory and a Port that make your income steady. You are the strongest human in the match right now, which is the position most players in the midgame fear and few players enjoy, because being the strongest human means every other human is watching your border and the land around you is the land everyone wants. The human to the south is not trying to beat you; they are probing to find the cheapest way to take a strip of your front while you are committed elsewhere. These are explicit hypothetical match state, not a fixed official setting, and the current board decides the actual density of your front.

The three questions from the stronger side. Who is the strongest human you face? You are, and the human to the south is the one probing. What is the cheapest land you can take? You do not need to take land; you need to defend the strip, so the cheapest “land” is the border itself, held with the troops you already have. Do you have a second front? Yes, the pocket and the Nation that the southern human is not touching, which means the southern human is not threatening your second front, only your border. The correct move is to hold the border with a packed front and to let the probing human pay for each strip they take, because a defender at full strength has the highest troops-per-tile on the border they are holding, and the reference attack-ratio framework says a defender’s losses are their average troops per tile, which is exactly the density a packed front maximizes.

Do not answer a small probe with a full-send, because the probe is cheap and the answer is expensive, and the probe is designed to make you spend the expensive push on the cheap target. The community’s fix for a full-send is the whittle, a five to fifteen percent click as the push crosses, and the same logic applies to a probe: a small, bounded answer that trades a little land for a proportional drop in the prober’s effective army is the correct answer, not a counter-push. If the southern human escalates from a probe to a real push, that is the moment the border becomes the most expensive target on the board and the push against them is no longer a hold problem but a ratio problem, where the answer is to push at parity or with a kill attempt against the now-exposed prober rather than to keep defending a border they have already decided to take. If the southern human keeps probing, the correct answer is to hold and to make the probes cost more than they pay, because a probe that cannot pay for itself is the southern human’s problem, not yours, and the timer of the midgame is on their side as long as your income is steady.

The table is the same table, read from the stronger side. The cheapest land row is now a defense row: hold the border, whittle the probe, and do not let the probe take the neck to your second front. The strongest human row is now you: the question is whether you can hold the border and still take the pocket, and the answer is the reserve, the troops you hold back specifically so the southern human cannot take the neck while you are committed to the pocket. The stop signal for the stronger human is the moment the probe becomes a push, the moment your reserve can no longer hold the neck, or the moment your second front is closed by the prober. The lesson of this scenario is that being the strongest human is a temporary position, not a permanent one, because the board is wide and the land around the strongest human is the land everyone wants, and the midgame is when the strongest human’s position is most fragile, not least. The probe is the signal that the other humans are reading your border and planning the push that will end your lead, and the packed front, the whittle, and the reserve are the three tools that make the probe cost more than it pays until the land around you is too expensive to take and the lead turns into a win.

Failure modes and the counters that actually work

The midgame failure is almost never a single bad push; it is a sequence of pushes that each look correct in isolation and together spend the bar on targets that do not pay. The most common sequence is the greedy land grab: a player who spends the full bar on a Nation, takes the land, and then cannot answer the second human who was always stronger, and the land they took exposed a border they can no longer defend. The counter is the reserve, the troops held back specifically to answer the strongest human, and the bounded push, the commitment that ends when the stop signal fires rather than when the bar is gone. The community states this directly: do not rush the land grab, wait for about half population, send thirds to a land grab, never all-in, and economy is key. That is not a vague piece of advice; it is the reserve and the bounded push stated as a rule, and it is the single most repeated answer in the FFA community threads.

The second failure is the fixed-ratio habit, and it is the one that looks correct most often. A player who always pushes at thirty percent wins the midgame against a distracted neighbor and loses it against a growing one, because thirty percent is the right push for one situation and the wrong push for the other, and the habit cannot tell the difference. The counter is the three-question loop, which answers the same push question with three different answers depending on what the defender is doing, and it is the only answer that is right in more than one third of cases. The reference attack-ratio page makes this explicit: the percentage is a computed number, not a setting, and the three questions are the computation. The midgame is where the fixed-ratio habit costs the most, because the midgame has more situations than any other phase, and more situations mean more of the two-thirds the habit gets wrong.

The third failure is the premature second push, the push that looks like the correct follow-up but is actually a small, outnumbered push paying the cheap clamp against a defender who has already regrown most of what they lost. The double-click pattern only works while the first push’s density drop is still in effect, which is roughly the next growth cycle, and a second push two cycles late is a probe, not a push, and probes do not take land, they only read. The counter is the stop signal, the named ending that tells you when the second push has stopped being a push and become a probe, and the discipline to hold the line when the stop signal fires. Players who lose the midgame to the premature second push are the players who “should have won,” because the first push was correct and the second push was the mistake, and the mistake was not visible in the first push, it was only visible in the density the first push left behind.

The fourth failure is the stale hold, the hold that stops being a hold and becomes a stall because the numbers stopped improving while the posture stayed the same. The hold is correct while the ratio of your troops to your cap is drifting up, the peak is still being funded, and a real target is forming, and the hold becomes a stall when the ratio is flat, the peak is maxed, and the only target left is the strongest human. The counter is the slope test: are your numbers improving or flat? If they are improving, hold is correct and you are buying time. If they are flat, you have run out of midgame and the hold has quietly become the lategame, where the correct move is the kill push the endgame pages own, not the hold the midgame pages own. The distinction between holding and stalling is the only distinction that matters when you are the player everyone is watching, because the player who stalls while looking like they are holding is the player who loses the midgame without a single bad push, and the player who holds while their numbers improve is the player who wins it without a single lucky break.

Map and mode adjustments that change the midgame default

The midgame default of hold-then-expand-then-ally holds on every map and in every mode, but the numbers shift, and the shifts are where map and mode knowledge lives. On coastal maps the front you are defending is shorter than the front you are attacking, because the water removes a strip of border from both sides, and a shorter front means the defender’s density on it is higher than their average density across the whole territory. That means the whittle that is correct on open land is usually one band too high on the coast, and the bounded push that funds a second click on open land often only funds a single click on the coast, so the second click should be held for the growth cycle after. The midgame on a coastal map is therefore a midgame of holding the coast with a packed front and taking the second front by land rather than by sea, because the sea is the expensive second front on a coastal map and the land is the cheap one.

On small or compact maps the midgame is shorter, because the board has fewer neutral pockets and fewer unwatched targets, which means the cheap land runs out faster and the strongest human is reached sooner. The hold on a small map is a shorter hold, and the expand is a faster expand, and the ally is more important, because there are fewer targets to ally around and the alliance is the only tool that removes a border from the problem list on a board with no cheap land left. On large maps the opposite is true: the midgame is longer, the cheap land is abundant, and the strongest human can be ignored for many minutes while the cheap land is taken, so the hold is a longer hold and the expand is a slower expand. The reference map-size guide owns the per-size numbers, and this page defers to it for the exact tile counts and encounter order, but the directional rule is that the midgame is the length of the cheap land, and the cheap land is longer on big maps and shorter on small ones.

In a team or custom lobby the midgame default changes more than the map changes it, because a teammate can hold a second front for you, which means you do not need a second front of your own, and the reserve can be shared, which means the reserve is smaller per player and the bounded push is bigger. The team pages own the shared positioning and the team economy, and this page’s midgame default is the FFA default, which is the no-teammate version, where the second front is your own and the reserve is yours and the alliance is the only tool that removes a border. In a ranked or one-on-one lobby the midgame is the whole game, because there is no third human to answer the strongest, and the hold-expand-ally loop collapses to hold and expand, with the ally question removed, and the strongest human is always the only target, so the midgame is the lategame from the first minute. The directional rule for mode is that the midgame is the set of humans who can answer each other, and the more humans who can answer each other, the longer the midgame and the more the ally matters.

The patch chronology is the version boundary for every number on this page. The troop growth formula and the attack-ratio clamps are the v0.34.11 reference, verified against the tagged commit, and the alliance duration of five minutes by default with a one to fifteen minute host range is the v0.34.19 Config, verified against the tagged source. The reference attack-ratio and population-growth pages own the per-tile loss and the per-tick growth, and this page owns the midgame decision that those numbers feed. A historical video that shows a midgame strategy is a method of asking the question, not a current rule, because the board, the numbers, and the humans all change, and the midgame is a phase, not a state, and the phase is defined by the numbers in front of you, not by the timestamp in the video.

Evidence, version limits, and the next decision

The evidence chain for this page is deliberately separated between the source that fixes the numbers and the community that fixes the question. The official v0.34.19 Release and the tagged Config source establish the alliance duration of five minutes by default with a one to fifteen minute host range, and the temporary embargo of five minutes, and the minimum distance between players of thirty tiles. The v0.34.11 reference, verified against the tagged commit, establishes the troop growth formula of (10 + troops^0.73 / 4) x (1 - troops / maxTroops) per tick with about ten steps per second, and the attack-ratio clamps that make the twenty-five to forty percent range the routine push and the parity or kill attempt the push against the strongest human. The tagged combat source is the mechanics boundary for a real push or a real response, and the reference attack-ratio and population-growth pages are the pages that own the per-tile loss and the per-tick growth that this page assumes you have read.

The community evidence is recorded in the research source pack and explains why players ask the midgame question, not which precise push is legal today. The three analyzed Reddit discussions are a player who has lost almost every FFA for a month and is asking what they are doing wrong, a poll about how often the odds are stacked against a player, and a player who cannot think of a way to defeat the enemy players in a game that ends after three hours. The common thread is that the midgame is the phase where the player who starts strongest looks unbeatable and the player who is not strongest cannot find the second front, and the answers cluster on the same three tools: the reserve, the bounded push, and the short alliance. The three analyzed YouTube walkthroughs are a strategy video on a specific tactic, a grinding video on becoming the best FFA player, and a tutorial on the game’s fundamentals, and the verified captions from the grinding video are the closest thing to a worked midgame example, showing a player who lets the board sort itself and then takes the land that the fighting leaves free.

Before queueing, make a four-line note: the strongest human you face, the cheapest land you can take, the second front you have, and the stop signal that ends the current push. After each push, update all four. If the second front no longer exists, reduce your exposure until you can find one, because a single corridor is not a midgame strategy, it is a countdown to being folded. If the cheapest land has been taken, reassess with the new numbers, because the cheap land is a set, not a target, and the set changes with every push. If the strongest human is about to be answered by someone else, ally and turn your attention to the pocket the alliance frees, because the alliance is the cheapest way to remove the wrong border and the most expensive way to remove the right one. The next useful reading depends on which question you now own: the attack-ratio guide owns the percentage and the timing of the push, the population-growth guide owns the growth curve that funds the push, the threat-assessment guide owns the choice of which neighbor the push is computed against, the nation-alliance-decisions guide owns the Nation-side of the alliance decision, the land-combat guide owns the per-tile loss of a real push, the winning-overtime guide owns the endgame kill push that the midgame hands over to, the FFA opening strategy owns the tempo and the reserve from the first minute, and the economy-fundamentals strategy owns the Port and Factory income that funds the hold. The midgame answer is complete when the player can choose, disprove, and replace a push in the live match with the current numbers, not with the plan made at spawn, and the hold is the posture that makes that possible, because the hold is the only posture that keeps the second front open and the reserve intact and the ratio improving, and the expand and the ally are the two moves that the hold funds when the board finally pays for them.