OpenFront Island Defense: Hold a Port, Stop a Landing, or Leave

Direct answer: what should an island player do?

An island is safe only when the next landing has a visible answer. Start with one connected core, a Port with a job, and a Warship patrol tile that covers the likely approach. Keep a reserve on land rather than spending every troop on a distant capture. When a transport enters the 130-tile Warship targeting envelope, kill it before it reaches shore; when the enemy controls the water component or your Port cannot be repaired, stop adding naval investment and retreat to the smallest coherent land position. These rules apply to the released v0.33.12 family and are not a promise that every map has the same coastline.

Use three words to name the state after every scan: secure, screened, or leaving. Secure means your core has a connected land reserve and the Port is not the only path to income. Screened means a Warship can reach the transport lane, the Port is on the same water component, and a second route or ally can buy time. Leaving means the route is disconnected, the Port is under direct attack, or the gold needed for another ship would buy more survival as troops, a City, or a safer land border. Naming the state prevents a beautiful island from becoming a sunk-cost trap.

The rest of this guide turns that language into checks, numbers, and two explicit scenarios. The community questions that motivated it are practical: players report being surrounded by stronger humans, losing a game after choosing City or Port at the wrong moment, and trying to “just trade” while a rival quietly blocks the coast. The answer is not a fixed 20%, 35%, or 50% opening ratio. It is a sequence that ties spending to the next hostile approach.

Make the first decision reversible. Before clicking a building, zoom out far enough to see both shores, then zoom in and count the tiles along the likely transport path. Write down the nearest friendly City, the nearest neutral connector, and the Port that could repair a ship. This takes less than a minute and exposes plans that look rich only because the camera hides a second border. If the answer changes when the map is rotated, treat the position as screened rather than secure.

The reserve is a decision instrument, not a superstition. A 30% floor means roughly 126,000 troops in a 420,000-cap example, but the useful question is what that reserve can reach before the next landing. If a reserve must cross a narrow bridge, count the bridge as a separate risk. If an ally can cover the bridge, record that promise and a time when it expires. Recheck the floor after every capture, upgrade, or diplomatic change so an old percentage does not become a new blind spot.

Read the shoreline before spending gold

The first scan is geographic, not economic. Identify the land component that contains your spawn, the number of passable exits, and every water component touching your coast. A blue patch is not automatically a route. The v0.33.12 transport helper chooses a closest reachable shore; a target on a disconnected inland lake is rejected. That matters defensively: an opponent may click your inland-looking coast, but no boat appears if their Port and your shore do not share navigable water. Conversely, an open sea component can make a tiny island reachable from several directions even when the minimap looks remote.

Mark one landing tile and one fallback tile. The landing tile is the shore an attacker is most likely to reach first: a broad beach, a Port-adjacent tile, or a coast that shortens the route from a rival’s Port. The fallback tile is a land tile behind it that keeps your City, factories, and remaining coast connected if the beach changes owner. Do not put every building on the landing tile. A captured Port changes owner, while a Defense Post on captured land is destroyed; your nuclear structures transfer, but your conventional land-defense layer does not. A fallback therefore needs ordinary troops, not just a structure icon.

Distance has three different meanings. A Warship searches for targets within 130 tiles, but its patrol order is generated within a 100-tile patrol range around its patrol tile. Passive repair checks a 150-tile distance from one of your active Ports, and docking requires a five-tile approach on the same water component. Those circles overlap, but they are not interchangeable. A ship can see a Port and still be unable to dock if the water path is cut; it can be close enough to repair while outside the patrol square; and it can patrol the right square while still missing a landing that enters from another coast.

Use the following map read before a purchase:

Question	If yes	If no
Does the Port share water with the threatened lane?	Keep a Warship screen possible	Treat the Port as local economy only
Is there a reachable fallback shore?	Reserve troops for a short retreat	Move the core inland before upgrading
Can one patrol tile cover the likely approach?	One ship may deny the first wave	Budget a second angle or abandon the route
Does the next building protect a route or raise capacity?	Spend after checking the border	Save Gold and preserve options

This is also the point to check mode settings. A private lobby can disable Ports, transports, Warships, or alliances. Do not import an island plan into a Ports-disabled match. In a Team game, a teammate’s Port is not your Warship’s repair anchor; the tagged Warship code filters for the owner’s active Port on the same water component. In Water Nukes, re-read the shoreline after a strike because a new channel can create a route while a converted strip can remove the land connector you thought was safe.

Build a defensible island core

The first Port is an option, not a victory condition. It costs 125,000 Gold at the first shared Port/Factory count and takes 50 ticks when instant build is off. It enables trade ships, transports, and Warships, but it can also become the enemy’s shortest capture route. Build it when the coast has a named job: a profitable partner, a transport bridge, or a patrol lane. If the only reason is “water is nearby,” delay it and put the Gold into a City or reserve instead.

Cities add 250,000 troop cap per completed level and can be built inland. On an island, the first City should usually sit behind the landing shore, not on the Port tile. The extra cap is useful only if the position can keep troops alive long enough to fill it. A level-2 City that cannot be defended is a larger prize for the attacker, not a stronger island. Use the shared cost ladder deliberately: the next Port or Factory level is 250,000, then 500,000 and 1,000,000. Compare that purchase with a 50,000 first Defense Post, a second Warship whose first cost is 250,000, or a ground expansion that closes a border.

A Defense Post is a land chokepoint tool. Its completed zone reaches 30 tiles, multiplies defender mag by 5 and speed by 3, and posts do not stack with each other. Place it over the tile an enemy must cross to reach the Port or City, not over empty water and not in a peaceful interior. Construction takes 50 ticks. During that window, keep troops behind the tile; the post is a multiplier, not a garrison. If the tile is captured, the post is destroyed rather than transferred, so never let one post be the only reason your core survives.

For a compact island, a practical order is: secure neutral land until one border is stable; build one City in the fallback core; add a Port only when its route is reachable and useful; add one Defense Post on the credible land approach; then build a Warship if the water lane carries a real transport or trade threat. This differs from a mainland opening because every extra coast is another angle. On a wide continent, a second Port may diversify income. On a one-island spawn, a second Port can simply double the number of places an opponent can raid.

Keep a written stop line. A good default is “do not spend below 30% of troop cap while a hostile border is live.” The exact threshold changes with terrain, allies, and opponent strength, but the logic is stable: the army that arrives after a landing is worth more than a Port upgrade that cannot be protected. If the enemy has no route and no Warship, you can run a lower reserve to grow. If the enemy is already moving a transport, hold more and let the next safe tick decide.

Use Warships as a screen, not a chase command

The v0.33.12 Warship routine gives defenders a strong default: it searches nearby targets in this order, Transport Ship first, enemy Warship second, Trade Ship third. A transport hit does not consume the normal shell reload in the same way, and the code immediately patrols again. That is why one correctly placed Warship can deny a landing fleet, while several ships stacked on the wrong coast can all miss the approach. Set the patrol tile on the lane, then watch the incoming-unit notice and the actual water path.

A fresh Warship has 1,000 base health and fires 250-damage shells. Its build price is 250,000 for the first ship, 500,000 for the second, and 750,000 for the third before the 1,000,000 cap. Those values make a second screen a meaningful investment, but not an automatic buy. A single ship can cover one lane; a second ship is justified when the island has two reachable shores, when the first ship must retreat for repair, or when the attacker can escort transports with an enemy Warship.

Repair defines how long the screen can stay forward. Within 150 tiles of one of your active Ports, a ship heals 1 health per tick. Docked within 5 tiles, eligible Warships share a pool of 5 health per Port level per tick. A level-2 Port supplies 10 per tick total, not 10 per ship, and its dock capacity is tied to its level. Automatic retreat starts at or below 75% of the current veterancy-adjusted maximum, unless the side is under Doomsday. A manual patrol move grants a 50-tick grace period before the repair routine overrides it. Read these as time constraints: if the Port is 120 tiles away and a transport arrives in 20 ticks, “retreat now” may be the wrong answer even when the hull is below threshold.

The water component is a hard boundary. The retreat routine looks for a Port owned by the same player and connected to the Warship’s component. A nearby Port across impassable land or a disconnected lake does not count. If the current Port is destroyed, full, or cut off, the ship may switch to a significantly closer available Port; if none exists, the repair retreat is cancelled. In Doomsday, the ship stays on patrol instead of idling at a Port because repair is disabled. A defensive plan that ignores this state can strand its most valuable hull at the exact moment the island needs it.

Use the patrol as a screen with a named objective: “deny the next transport,” “cover the Port approach,” or “buy 30 ticks for the land reserve.” Do not send a veteran to chase a Trade Ship simply because it is visible. A captured trade vessel may be useful, but leaving the landing lane can convert a small economic gain into a lost City. The Warship veterancy guide covers the yellow bar; here the decision is whether the hull is in the right water at the right time.

Landing math: make the enemy pay for distance

Transport Ships cost no Gold, but they carry the sender’s troops. The tagged configuration caps a player at three active transports and sets a default boat attack amount to one fifth of the attacker’s troops. A transport moves one tile per tick along the water path. On arrival, it conquers the destination tile and resolves an attack unless the target is friendly. If the destination becomes water, the boat retreats toward a valid source and loses 25% of its troops on a cancelled retreat. These facts produce a simple defensive arithmetic: a longer route gives your Warship more ticks to find the boat, while a blocked destination can return most of the attack but still costs a quarter of the embarked troops.

Suppose a rival has 500,000 troops. A default transport carries about 100,000. If the route is 80 tiles and your Warship can begin firing at 130 tiles, the screen has roughly 50 movement ticks before shore contact, ignoring path bends and shell timing. A second transport does not double the defender’s response time; it creates a target-priority queue. Keep the patrol tile near the approach so the first kill does not require a long reposition. If the rival escorts with a Warship, decide whether the escort or the transport changes the board first. The routine prioritizes the transport, but a player can manually move a ship and expose the lane.

The best island defense is often a geometry change rather than a bigger army. Capture the neutral tile that shortens your fallback border, keep the Port behind a Defense Post-covered land route, and refuse a landing that opens a second enemy border. If the attacker can land on a broad beach and immediately touch your City, the Port’s income is not buying safety. If the attacker must cross a 30-tile Defense Post zone and your reserve can counterattack from two adjacent tiles, the same 100,000 troops may be insufficient even though the raw numbers look equal.

Do not overread the word “reachable.” The transport helper rejects disconnected shores, but it does not promise that the path is short, safe, or economically worthwhile. Water Nukes, captured Ports, and changing ownership can rebuild the path graph. After every terrain-changing strike or Port capture, check the new shore and water component. A route that was impossible one tick ago can become the shortest route after a channel opens; a route that looked safe can become a retreat path with a 25% loss after the target turns to water.

Scenario one: one Port, one Warship, two hostile shores

Assume a normal FFA island with 420,000 troops, one level-1 City, one level-1 Port, and a level-0 Warship at 930 health. The Port is 90 tiles from the west shore and 140 tiles from the east shore. A rival with 600,000 troops owns a Port on the same ocean component and begins a transport toward the east shore. A second rival is adjacent to the west fallback border but has not launched a boat. The assumptions are explicit: no Water Nukes, no Doomsday, normal build time, and both enemy players can attack.

The first temptation is to retreat the Warship because 930 health is below the 75% threshold of 1,000. That is mechanically understandable but strategically incomplete. The ship is still inside the 130-tile target range for the east approach, and the Port is close enough for passive repair after the landing is denied. If the ship starts a 140-tile retreat now, it abandons the only active lane and gives the transport a free arrival. The decision framework says screened: hold the east patrol tile, fire at the transport first, and keep the land reserve above 30% of cap.

The west border changes the second half of the decision. Do not spend the next 250,000 Gold on a second Port while the transport is in flight. Place a Defense Post over the west land approach only if the 50-tick construction window completes before the rival can cross it; otherwise keep troops there and delay the post. If the Warship kills the transport, move it toward the Port’s 150-tile passive-heal circle without leaving the west reserve. If the enemy escort enters range, target the escort only after the transport is no longer the immediate landing threat.

The attacker’s best counter is a two-boat distraction: one transport enters east while another aims west, forcing the defender to split attention. The defender’s counter-counterplay is to keep the Warship on the lane with the shorter path and use the land army to absorb the slower route. If both boats arrive inside the same window, abandoning the Port is better than defending two exposed beaches. Capture the neutral connector behind the City, let one shore go, and remain secure on the fallback core. The win condition is not “save every tile”; it is “deny the landing that would make every remaining tile reachable.”

Scenario two: the Port is rich, but the water is lost

Assume a coastal island with 900,000 troops, two Ports, three Trade routes, and one level-1 Warship at 880 health. The first Port is level 2 and the second is level 1. An enemy captures the first Port’s shoreline, parks a Warship near the trade route, and sends a transport toward the only City that connects the island’s two halves. The surviving Port is on a different inland-water component. The island still has enough Gold to buy another Warship, but the next shared Port/Factory cost is 500,000.

The map now says leaving, even though the economy looks excellent. The Warship’s 880 health is below its 75% retreat threshold, but more importantly, neither Port is a valid repair anchor for the threatened component. The code’s same-component check means the surviving inland Port cannot heal or receive the ship. Building a new Warship at 250,000 does not solve the disconnected water; upgrading the rich Port to level 3 does not reconnect the route. Spend the Gold on a City behind the threatened shore, a Defense Post over the land connector, or a ground reserve that can retake the Port after the transport is denied.

The enemy’s likely mistake is to assume that captured Port income equals control of the whole island. Use that assumption against them. Let the first Port remain contested while you shorten the land border, then launch a ground attack when the enemy Warship is busy screening its own transport. If the transport destination turns to water through a Water Nuke, watch for the 25% retreat malus and the new reachable shore; do not chase a boat into a component you cannot repair in. Once the City and fallback shore are connected again, the state can move from leaving to screened.

The timing test is deliberately conservative. At 880 health, the ship is not merely “a little damaged”; it is below the normal automatic-return threshold, and every extra tick outside a valid repair circle is a tick that cannot be recovered. Mark the moment the enemy Warship leaves the lane. Only then consider a short patrol move or a replacement hull. If the enemy keeps the lane closed, the correct naval action may be zero: ground troops and a Defense Post create a position that survives even after the rich Port is gone.

This scenario explains why “build more Ports” is a poor universal answer. Maritime income is valuable when routes complete and ships can be replaced. It is negative value when each Port creates another capture target and the water graph is fragmented. Track completed trips, reachable repair, and the next hostile approach. If two of those three are negative, switch to land economy until the geometry changes.

Failure modes, counterplay, and map adjustments

The first failure is stacking every structure on the shoreline. The fix is a Port job, a City fallback, and a Defense Post-covered connector. The second is using the attack ratio as a substitute for a reserve. A 50% push can be efficient against neutral land and still lose an island when a human transport arrives during recovery. Name the reserve floor before clicking.

The third failure is assuming Warships protect a radius rather than a route. Their 130-tile target search, 100-tile patrol generation, and 20-tick shell attack rate reward a patrol tile aimed at traffic. A ship on the opposite coast may be “near” on the minimap but still late by dozens of ticks. The fourth is treating a Trade Ship as a harmless bonus target. Warships hunt Trade Ships only after transports and enemy Warships, and a player-controlled chase can abandon the one lane that matters.

Opponents have reliable counterplay. They can send a transport as a distraction, escort it with a Warship, attack the Port so retreat becomes impossible, or capture the Defense Post tile and remove the multiplier. They can also use diplomacy to make you lower the reserve before the alliance expires. Respond with role separation: a fresh ship screens cheap transports, a veteran holds the decisive lane, a land reserve guards the City, and an ally watches the second shore. Never let one damaged ship be the only answer to two independent routes.

Map shape changes the thresholds. On Compact maps, a 20-tick detour can cross the entire meaningful sea lane, so pre-positioning beats a late retreat. On archipelagos, one Port per water component is a strategic anchor only if each island can defend its own fallback. On strait maps, a single patrol can deny a crossing, but losing the chokepoint may expose every Port at once. On large continents, an island Port may be a side economy; do not divert the main army to save it if the land core decides the match. In Team games, assign one player to the sea lane and tell the frontline exactly which shore is being screened. In Ports-disabled or transports-disabled modes, convert the same Gold into Cities, Defense Posts, and ground mobility.

Use a map-specific alarm instead of a universal build order. On a compact map, call an alarm when a transport is 30 tiles from the shore because the remaining travel time is short. On a large map, the same alarm can be 70 tiles if the ship must turn around a peninsula. In a team, name the player responsible for each alarm and the fallback tile they will hold. In a solo lobby, put that marker in your own notes. The point is to make the next action obvious while there is still time to perform it.

Mode restrictions are strategic information. If transports are disabled, an enemy cannot create a second border by sea, so a Port’s defensive value drops and a City may win the comparison. If alliances are disabled, remove diplomatic quiet from the reserve calculation. If Water Nukes are enabled, treat every crater as a possible graph change and rescan before spending. The same island can therefore deserve three different opening plans without any rule contradiction.

Next reading and an auditable close

After each naval contact, write one sentence that another player could verify: “I stay because the transport is inside 130 tiles,” “I retreat because 880 is below the adjusted 75% threshold and no same-component Port exists,” or “I leave the Port because the fallback City preserves one coherent border.” Then check the result on the next important tick: transport destroyed or landed, Warship health changed, Port component still connected, and reserve still above the chosen floor.

Use first-match fundamentals when the opening map read is still unfamiliar. Use team naval control when a team needs route ownership and escort assignments. Use land combat when the transport has landed and the fight is now a border calculation. Use Port versus Factory when the maritime plan competes with rail or City capacity. For the underlying values, check units, structures, and trade mechanics. On Four Islands, continue with the Four Islands team coordination guide to assign each island a job and time the handoff.

When the island is already secure but still lacks a route to winning land, continue with the tall versus wide guide to test a bounded mainland foothold without emptying the defended core.

All numeric boundaries here are the released v0.33.12 behavior: Warship base health 1,000, shell damage 250, target range 130, patrol range 100, passive repair within 150 tiles, five-tile docking, 75% retreat, Defense Post range 30 with mag x5 and speed x3, transport cap three, and a default boat attack of one fifth of the sender’s troops. If a later formal Release changes any of these, update the table, scenarios, and decision sentences together.

Keep the version boundary visible when you share a replay or team note. A sentence such as “hold the east lane for 50 ticks” is only useful if the reader also knows whether the lobby allowed instant construction, whether Doomsday was active, and which Port was on the same water component. Record those assumptions beside the result, not in a hidden comment. When a future Release changes a range or cost, update the number, the example that uses it, and the stop condition that follows from it.

The practical close is a four-line log: state, threatened tile, reserve floor, and next check. For example: “screened; east beach; 126,000 troops; check transport at 20 ticks.” If the next check says the route is disconnected, change the state to leaving and spend accordingly. This small log turns a long guide into a repeatable habit that another player can audit after the match.

To plan the opposite side of this fight, use the Transport Landings guide for coast selection, wave allocation, Warship screening, beachhead depth, and an explicit abort rule.

Sources: OpenFront v0.33.12 Release, WarshipExecution.ts, TransportShipExecution.ts, TransportShipUtils.ts, and Config.ts.