Answers that move your ladder forward.

Six decisions, in this order.

1. Get ten names you can actually contact. Ten is the floor. Below that a ladder collapses into the same two players every week, and two more weeks of recruiting beats opening with six.
2. Split them into level bands if the spread is wide. A band holds players of roughly one standard. Most ladders run two or three. Bands are what stop a new member's first match from being a formality.
3. Pick a season length and publish the dates. Six, eight, ten or twelve weeks, twelve at most. A rung earned in March means nothing in November. If your group plays year round, run the ladder open-ended instead and let decay keep the order honest.
4. Set the two clocks and leave them alone. A challenge reaches up to five rungs, the player challenged has 48 hours to respond, and the two of them have thirty days to play.
5. Make the result require both players. The winner enters the score, the loser confirms, and a timer confirms it if the loser says nothing.
6. Decide what happens to the player who goes quiet, before it happens rather than after.

Step six is where most club ladders come apart. What organizers get covers the tools, and the ruleset covers the defaults.

Less than an organizer expects, if the deadlines are automatic. Challenges are sent by players. Scores are entered by players and confirmed by players. A missed response resolves itself as a walkover. What is left is the part software cannot do: recruiting, answering a question about a rule, and occasionally deciding a disputed score. Budget an hour in the first week and a few minutes a week after that.

Ten to open, twelve before it sustains itself. The number that matters is not the roster, it is how many people are willing to play a stranger on a Tuesday. A ladder of thirty where six people play is worse than a ladder of twelve where ten do. Recruit from a group that already exists: a club list, the park regulars, a group chat. What clubs get covers the multi-band case.

Yes, and most ladders in the world still are. It works until three things happen at once: somebody disputes a score, somebody goes on vacation, and somebody stops responding to challenges. Each of those is a phone call the organizer has to make, and the organizer is the person who quits first. Try the shape of your ladder in the simulator before you commit, then start a ladder when the group is real.

Cap it. Five rungs is the default here, and every published club ruleset we surveyed caps it somewhere. Uncapped ladders produce a 3.0 challenging a 4.5 in week one, which is a bad hour for both of them and the fastest way to lose a new member. Tighter caps (three rungs) make the ladder climb slowly and reward patience. Wider caps (eight or more) make it volatile. Rule 3.

Not on a well-run one. Upward only. Let challenges run downward and the strongest player in the band farms the bottom of the standings for safe wins, which is exactly the behavior a ladder exists to prevent. Play a friendly against whoever you like. It does not move the standings.

48 hours to respond, one to two weeks to play. That is where the surviving club rulesets landed independently, and the reason is practical: 24 hours punishes anyone who does not check email daily, and a week of silence kills the challenger's interest. The deadline only works if missing it has an automatic consequence attached. Rule 5.

Nothing, if the ladder is built correctly. The clock expires and the challenger takes the result as a walkover, without an organizer adjudicating and without anybody emailing an administrator for permission. If you are running this in a spreadsheet, write the rule down before the season starts and apply it the first time it happens. The first unenforced deadline teaches everyone that deadlines are optional.

A rematch cooldown. Fourteen days by default. Without one, two friends at rungs 6 and 7 can play each other every three days, swap positions on each result, and climb past the rest of the band on volume alone. The cooldown does not stop them playing, it stops those results counting twice in a fortnight.

Swap is the default and the right answer for most groups. The winner takes the loser's rung, everyone between shuffles down one, and any player can verify the standings by reading them. Points systems reward activity, which suits a ladder where availability varies wildly, but they require a table nobody reads and they let a busy 3.0 outrank an absent 4.0. Pick one before launch and publish it. Both systems compared.

Both players. The winner enters, the loser confirms, and silence auto-confirms on a timer so a sulk cannot freeze somebody's ranking. Accept only valid scorelines, which removes most typos before they reach the standings. Score verification is close to unsolved across this category: of five platforms surveyed, one required both-party verification and one published a penalty for falsifying a result.

Keep them separate, because they are different behaviors.

• Walkover. The opponent never answered, or the deadline passed. The result goes to the other player and no match was played.
• Forfeit. The match was accepted and then abandoned before it was played.
• Retirement. The match started and one player stopped. The completed part of the score is recorded with a retirement marker.

None of the three counts as a played match, which matters at the end of a season because playoff qualification is gated on matches played. The definitions.

Whatever the format allows, and nothing else. A best-of-three match cannot end 8-6. A pro set cannot end 6-4 6-2. A 7-6 set requires a tiebreak reaching at least seven points won by two, so 9-7 is valid and 7-6 as a bare game score is not. Publishing the accepted list is a five-minute job that prevents a season of ambiguous entries. Every accepted scoreline.

Four numbers on four different scales, none of which converts cleanly into the others. NTRP is a banded self-rated and computer-rated US system. UTR is a continuous scale driven by match results. WTN is the ITF's equivalent. A club level is whatever your club says it is. Any mapping between them is an approximation, and anybody presenting one as exact is selling something. How Deuce reads them.

Look at the actual spread of your group, not at a rating scale. If your twenty players run from a recreational 3.0 to a strong 4.5, two bands is right. If they are all within half a level, one band is right and splitting them just halves everybody's opponent pool. A band needs eight to ten players to function. Two thin bands are worse than one broad one.

Seed them from the most conservative signal you have, then let the first match correct it. The fix matters more than the guess: a new player's first challenge should reach their whole band, so one match puts them roughly where they belong. Placement accuracy is not a problem worth solving in advance, because a played match answers it better than any form.

Sandbagging is claiming a lower standard than you play at, to win a lower band. Signs: a placement well below the claimed external rating, a run of one-sided wins, and a player who declines to move up when asked. The fixes are structural. Read the conservative end of a claimed rating, give a new player a first challenge that reaches their whole band, and let the organizer move a player between bands with the move logged in public. How placement reads a claim.

The pair is the ranked entity, not the player. A pair holds one rung, challenges as a unit, and takes the result as a unit. Everything that applies to a singles player applies to the pair: one rung, one point total, one tiebreak record. That is the version that ships here. How the format works.

Fixed pairs are simpler and fairer, and they are what Deuce ranks. A rung means something because the same two people earned it. Rotating partners produces a more sociable ladder and a standings table that cannot be defended: if you played with the best player in the band twice, your rung says more about them than about you. Rotating partners is roadmap here, not shipped.

Decide before the season, not during it. The workable rule is that the pair holds its rung for a fixed window while it finds a replacement, the replacement is approved against the band's level, and the pair keeps its record. The alternative, dissolving the pair and releasing the rung, is cleaner to administer and much less popular with the player who is left.

Accept that it is harder and give it more time. A doubles ladder needs a longer play window than a singles ladder because four calendars have to intersect, and a court booking for four is usually longer. If your group is small, run singles and doubles side by side rather than choosing: the same people will play in both, and the singles ladder carries the weeks when doubles cannot be assembled.

Six to twelve weeks, and twelve is the ceiling. Shorter than six and half the field never gets three matches in. Longer than twelve and the standings stop feeling live, because a result from four months ago is still holding somebody's rung. Publish the start and end dates before anybody joins.

The other option is to run the ladder open-ended, with no end date. That works for a group that plays all year, but it puts the whole weight on inactivity decay: without a close, nothing else clears out a rung somebody stopped defending. Pick open-ended for continuity, a fixed season for a finish line and a playoff. Season rules.

Carry them over. A cold start every spring throws away the one thing the ladder built, which is a roughly accurate ordering of the group, and it forces everybody through a round of pointless mismatches to rebuild it. Carrying over does concentrate an advantage at the top, which is what the playoff and the level bands are for.

Ladders need reviving far more often than they need founding. What works:

1. Find out why it stopped. It is almost always one of three things: the organizer burned out, the deadlines were never enforced, or the group got too thin to produce matches.
2. Cut the roster to the people who actually played. Do not reopen with the old list. Reopen with the eight who showed up.
3. Start a new season with published dates, rather than resuming the old standings mid-air.
4. Automate whichever failure killed it, because willpower already lost once.

If it died on admin, start a ladder and hand the deadlines to software.

Let them in and place them at the bottom of their band, or seed them where their rating suggests and say so publicly. Both are defensible. What is not defensible is placing them by judgment, silently, one at a time, because the third such placement is the one somebody complains about. Publish the rule in week one and apply it the same way every time.

Publish the order in advance and apply it automatically, stopping at the first step that separates the players. The order used here:

1. Matches played, higher first.
2. Head-to-head between the tied players.
3. Sets won minus sets lost.
4. Games won minus games lost.
5. Most recent played match, earlier first.

Walkovers received do not count toward step 1, because they are not played matches. Publishing this before the season is the single cheapest thing an organizer can do to avoid an argument in week twelve. The full order.

Have one player dispute inside the confirmation window, then have the organizer decide from both submissions, then write the resolution to a log everybody can read. The log is the part people skip and the part that matters: a resolution nobody can see looks like a favor, whatever it was. Rule 18.

It is optional and it is usually worth it. A playoff gives the season an ending, and gating entry on matches played rather than final position means the top four cannot climb early and coast. If you run one, publish the qualification rule and the bracket size before the season starts, and freeze both once it does.

That is a band problem, not a rules problem. One player two levels above the rest of the field will win everything, stop being challenged, and get bored. Move them up a band, or split the ladder, or recruit two more players at their level. None of those is a punishment, and all three are better than letting them run out of opponents.

The two players agree, and they split the cost unless the ladder says otherwise. A platform that assigns courts is claiming to know things about your city that it does not. Publish the local court options where players will see them and let the two of them sort out the hour. Nearby courts and ladders.

One tiebreak set is the safest choice. An eight-game pro set usually fits. Short sets to four and FAST4 both fit comfortably and suit a group that plays in the evening. A full best-of-three, the platform default, does not, unless the third set becomes a match tiebreak to ten, which the default allows when both players agree on court. Pick one format for the whole ladder, because two players negotiating a format at the net is a dispute waiting to happen. All six formats.

The standing convention, which is worth stating in your rules rather than assuming: you call your own side, you call it immediately, and you give the point when you are not sure. A ball you cannot confidently call out is in. If a caller keeps getting it wrong, play the point again rather than arguing about the last one. Nobody has ever won a ladder match by winning a line call.

Publish the penalty before it happens and apply it the same way to everybody. It takes two people to file a fake score when both sign off, so a falsified result is usually a lone player entering a match that did not happen. A published penalty that costs the rest of the season is a deterrent. A penalty decided case by case is a negotiation. The penalty.

A standing ranking of players. Challenge somebody a few rungs above you, play, and the winner ends up on the higher rung. No fixtures, no draw sheet. The full explanation.

A league hands you a schedule. A ladder hands you a list of opponents and lets you pick. Side by side.

No. You need to play a match and keep score. Ladders are split into level bands, so you are ranked among people of roughly your standard. If the placement is wrong, your first challenge reaches anyone in your band and fixes it in one match. How placement works.

Both, with one limit. A doubles ladder ranks fixed pairs, so you and your partner hold a rung together for the season and cannot swap in week five. Rotating partners is roadmap, not shipped. What organizers configure.

the maximum. Seasons exist so the standings mean something at the end. A ladder can also run open-ended, with no end date, no close and no playoff, and the ladder page tells you which one you are joining. Seasons and playoffs.

Qualification is gated on matches played, not position, so you cannot climb early and coast. Walkovers do not count. Why we count it that way.

The standings close, the playoff runs, and the standings carry into the next season. A ladder does not cold-start every spring and you keep the rung you earned. Season close and carryover.

Yes, and plenty will. Each ladder tracks your rung on its own, so a result in one moves nothing in another. Your match record is yours and travels with you.

Enter your city and Deuce shows every ladder in it, with the level band, format, season dates and how many players played in the last 30 days. If the answer is zero, the page says zero. Find a ladder, or browse the directory without an account.

Open the standings, pick a name inside your challenge range, send it. The default range is five rungs above you, because uncapped ladders produce mismatches nobody enjoys. Challenge rules in full.

No. Upward only, on every ladder. Let challenges run downward and a strong player farms the bottom of the standings for safe wins. Play a friendly against whoever you like; it will not move the standings.

Two by default, counting sent and received, so nobody keeps four opponents waiting while they work out their week.

They have 120 hours. If the clock runs out you get the result automatically, as a walkover. No organizer adjudicates and nobody emails an administrator for permission. One of the better-known rating platforms still makes you do exactly that.

No. A walkover is what you get when the other player never answered or the deadline passed. A forfeit is giving up after accepting. Both appear in your record, labeled differently, and neither counts as a played match. Both defined.

thirty days from acceptance, with one extension. Both players are equally responsible. If neither could get a court and neither stalled, the match is voided rather than pinned on one of you. Rule 7.

Wherever you two agree. Deuce owns no courts and assigns none. Nearby courts are listed on your city page to make that conversation shorter.

You split it, unless your ladder says otherwise. Deuce is not in that transaction.

As long as the format your ladder picked. The default is best of three tiebreak sets, and when court time runs out at a set apiece, both players can agree to a match tiebreak to ten in place of the third set. Six formats are accepted, including the pro set, short sets to four and FAST4. No-ad is an option inside a format, not a seventh one. All six.

Winner enters, loser confirms. Silence auto-confirms on a timer, so it cannot stall your ranking. Only valid scorelines are enterable. Scoring in detail.

Either player disputes inside the confirmation window. The organizer sees both submissions and decides, and the resolution goes to a public activity log, which exists so a resolution can never look like a favor. Rule 18.

It takes two people to file a fake one, since both sign off and only valid scorelines are enterable. The penalty is published in the rules rather than decided case by case, and it costs the rest of the season. The penalty.

Declare an absence: up to 21 days at a time, 42 across a season. You keep your rung and cannot be challenged. Without a valve like that, what players reach for is silence.

Yes, deliberately. After 21 days with no played match your rung slides one place a week until you play. Passive and automatic: nobody reports you, and it applies to the organizer too.

Yes, from your settings. Your played matches stay in the record, your rung is released, and the players below move up. You can rejoin next season.

Not straight away. The rematch cooldown is 14 days by default, because without it two friends can trade wins and climb past everybody else.

If that sounds like something you would use, find a ladder near you.

Eleven questions about the group, the levels and the season you want, about five minutes. Then a person here builds it with you. Start a ladder.

Ten you can name. Below that we say so and help you recruit rather than launch something that dies in week three. Twelve is where a ladder stops being the same two people every week.

With the people you already have: a club list, your park regulars, the group chat that already plays. A ladder needs local density, not a national user base. What organizers get.

Not set yet. When it is, it goes on the pricing page in numbers, for everybody, with no quote-gated plan for clubs. Organizers are the ones who pay. Pricing.

Send us the sheet in whatever state it is in. A person does the parsing pass and you approve the result, not a self-serve upload. Your current order becomes the opening standings and nobody loses a rung they earned. For organizers.

The dials, yes: challenge range, deadlines, level bands, cooldown, decay rate, absence limits, season dates, playoff size, match format, and swap-versus-points movement. The full configurable list.

The parts that make a result trustworthy. Challenges go upward only, the range is capped, both players confirm scores, absence and decay both exist, and a new player's first challenge always reaches their whole band. Why those are fixed.

Yes, and several under one club split by band or format is the normal setup. One dashboard, and each gets its own public standings page. For clubs.

No, but most do. Playing organizers are subject to every rule they set, decay included, and every adjustment they make is written to the public activity log. That log is what makes a playing organizer credible.

Write to us. The ladder keeps running either way, because deadlines, walkovers and decay are handled by software. We can hand the organizer role to somebody else in the group. Email us.

Software handles them, not you. A missed deadline produces an automatic walkover, inactivity decays a rung, repeated walkovers flag a player inactive. You chase nobody in a group chat.

Next step is the application. Eleven questions, and a person replies inside one business day.

From what you tell us. Bring an NTRP, UTR, WTN, LTA rating, ITN, federation ranking or club level and we seed you from it, reading the conservative end of your range. The number is never public, only the band is. Your first match is where the real information comes from. How the systems map.

We accept all of them at signup as your declared level, and publish an approximate conversion table between them. It is Deuce's own estimate, not endorsed by the USTA, UTR or the ITF, and it does not round-trip perfectly. The declared level only places you in a band; after that, your rung is earned on court. The conversion table.

Placement reads the conservative end of what you claim, a new player's first challenge reaches their whole band so a wrong claim is corrected in one match, and organizers can move a player between bands with the move logged publicly. Rule 5.4.

Your rung is your position in one ladder under its rules. A rating (NTRP, UTR, WTN) is an outside system's estimate of your standard, read once at signup to place you in a band. Deuce computes no rating of its own: your rungs and your match record are the whole story. How the systems map.

Your name, rung, played results and ladder membership are public on the standings page, because that is what a standings page is. Your email, phone, exact location and availability are not. You can play under a shortened display name, and we sell nothing about you. The privacy policy.

No. Standings, ladder pages and city pages are public and always will be, a commitment on the about page, not a launch promo. Getting back to a ladder you play in? Sign in here.

No native app yet. The site works on a phone and does everything you need. Native apps are roadmap,

answer stays as it is.

It will say so. Ladders only work at local density, so they start one group at a time. Leave an email and hear from us once, when a ladder exists near you, or be the person who starts the first one. Start the first ladder in your city.

Every city page and the directory show live counts, including zero. An active ladder means one with a played match in the last 30 days, and Deuce is pre-launch, so in most places the honest number today is none. How we count.

Yes. A ladder can be created anywhere in the world, and no country is off the list. Deuce fills city by city rather than country by country, English-first everywhere. Court fees, booking and eligibility are whatever your local courts already do.

A ladder also does not have to be a club. The organizer sets the scope: one club, several clubs, a city, an area inside a city, a region or state, or a whole country. It is published on the ladder page so you know how far you may be traveling. Rule 1.10.

No. Some ladders are club-only and say so on the ladder page. Plenty are open to anyone in a city and play on public courts. Each ladder's eligibility rule is published before you join.

Yes, permanently. Browsing ladders, reading standings and looking up courts need no account and no payment. Playing sits behind whatever the organizer's season costs, and that number is not set yet.

Only in a ladder whose organizer turned it on, and the ladder page says so before you join. There are extra rules about who sees a junior's details and who can contact them. Rule 1.5 and the privacy policy.

Email a person: hello@deuceladder.com, with response times published on the contact page. No phone queue and no chatbot.

The glossary has all three alphabetically, with everything else the ruleset uses. Longer explanations live in the topic groups at the top of this page.