Topics

This page explains operations that span multiple screens and provides supplementary information for complex features.

This section describes how players can view information shared by the host. You can also try this on a single smartphone by viewing it in a web browser.

Log In on the Host Device

  1. Select the settings screen.
  2. Tap the Login button.
  3. Follow the instructions on the login screen.
[diagram]

Set the Event ID

  1. Tap the down arrow icon on the right to open the panel.
  2. Create and enter a unique event name.
    After registering, you can select it by tapping the registration list button.
  3. Tap the Register button.
  4. Start accepting player registrations from players.
[diagram]

Register on the Player Device

  1. Log in as you did on the host device.
  2. Switch to player mode.
  3. Enter the event ID provided by the host.
  4. Tap the Register button.
    After registering, you can select it by tapping the registration list button.
  5. Turn on “Receive notifications” to receive updates from the host automatically.
[diagram]

Information Sharing

  1. On the host device (blue toolbar), tap the Sync button to upload data to the cloud.
  2. On the player device (green toolbar), tap the Sync button to download and copy the host's data.
  3. When the host uploads data, a push notification is sent to registered players, which automatically triggers a data sync on their devices.
[diagram]

This section explains the ratings displayed on the Match screen.

  • The initial rating is 25.0 (8.33), where 25.0 is the average rating (µ) and 8.33 is the standard deviation (σ). 25.0 indicates strength, while 8.33 represents the amount of variability.
  • A rating is simply a relative value within an individual group that indicates how likely it is to win a match.
  • In general, a higher rating indicates a stronger player.
  • If we assume that the degree of variation follows a normal distribution, we can estimate the degree of accuracy as shown in the graph. This is an illustration of the initial values, so the variability is very large.
  • Initially, it fluctuates greatly, and with each game, μ rapidly approaches its true value. The amount of fluctuation is not constant. For example, beating an opponent with a relatively high rating will result in a larger increase.
[diagram]

The following practical example illustrates how ratings can differ.

  • Panda and Tiger both won one game and lost one game, but there is a difference in rating between Panda at 24.9 and Tiger at 25.1. Panda won the first round against a pair with a rating of 25.0+25.0=50.0 and lost the second round against a pair with a rating of 28.1+21.9=50.0. Tiger lost in the first round to a pair with 25.0+25.0=50 and won against a pair with 25.0+28.1=53.1.
  • The value of a loss is the same, but in the case of a win, the opponent's rating is higher for Tiger. This is one of the reasons why Tiger's rating is valued higher. However, in the case of doubles, it is not so simple because the partner's rating is also affected and the contribution to the win or loss is distributed. In this example, the second round was 25.0 + 28.1 = 53.1 vs. 28.1 + 21.9 = 50.0, but the lower-rated Frog and Tiger team won, so their ratings increased.
  • If you are interested in the details of the calculation method, try searching for the keyword “TrueSkill.”
[diagram]

By setting each player's participation status for the draw, you can control who is selected according to the situation. This is easier than manually changing statuses after the draw and makes it easier to level the playing field.

[diagram]Participation status (default)
  • Participate in the draw. Whether the player is selected to play or take a break depends on the draw result.
  • Use this status normally.
[diagram]Priority participation status
  • Participate in the draw and give this player the highest priority for being selected to play.
  • Set this option if you want to make sure to include a leader or other person in the game.
  • You can set this status by pressing and holding the Participate button.
[diagram]Break status
  • The player will always take a break.
  • Set this for players who want to take a one-game break due to fatigue, poor health, etc.
  • This is treated the same as participating in the draw and being left out, so the player will be prioritized in the draw upon returning.
[diagram]Continuous break status
  • This is the same as Break status, but it persists across multiple draws instead of being cleared automatically.
[diagram]Absence status
  • The player will always be absent.
  • Set this when the player is not available due to late arrival or early departure.
  • If the number of participants is increased in the middle of the event, the time before the increase will be treated as a late arrival and assigned this status.
  • Missed draws are not counted because the absence is considered personal. However, players who have not played recently are prioritized, so a player who has just returned will receive priority.
Differences Between Player Statuses
Status Continues Between Rounds Counted as a Missed Draw
[diagram] Continue Counted if the draw result is a break.
[diagram] Continue Counted, but this status usually results in a match rather than a break.
[diagram] Does not continue Counted as a missed draw (break).
[diagram] Continue Counted as a missed draw (break).
[diagram] Continue Does not count.

The following explains the algorithm used to determine combinations. Because this process is mechanical, please adjust the combinations as needed on the match table screen.

  1. First, the players who will play are selected. Candidates are registered players who are not on Break or Absence status.
  2. Players with a low past win rate are favored, while those with a high past win rate are given lower priority. The main goal is to equalize everyone's chances of winning.
  3. To discourage long breaks, players with more consecutive breaks are given a slight advantage, while those with more consecutive appearances are given a slight disadvantage.
  4. Players with fewer appearances are given a slight advantage.
  5. Players are selected to play in priority order. If players have the same priority, they are selected randomly. When pairs are fixed, the average value is used for the draw.
  6. Next, pairing is done. For those who have a fixed pair, the pair is decided, and for those who do not have a fixed pair, the pair is randomly combined and multiple candidate patterns are generated internally. The number of past pairings is counted, and the combination with the fewest number of pairings is selected, so that people with different or fewer pairings are paired as often as possible.
  7. If “Balance the rating” is enabled, the game tries to pair opponents based on their ratings so that the expected win rate is not too uneven. The logic also tries to create pairs and matchups that have occurred less frequently in the past. Additional constraints from the Combination Settings screen are applied as well.
  • With Absence status, the participant is treated as having lost, but is hypothetically considered to have won or lost based on the average probability of all participating members except that participant. This ensures that absence is neither an advantage nor a disadvantage. Because the player has no consecutive appearances and fewer total appearances, they will often be selected soon after returning from an absence.
    The idea behind this software is that equal win probabilities are fair. If you instead consider an equal number of wins fair, set the break status to count as an unsuccessful result every time.
  • As an exception, when the number of playable members is N and the number of players who need a break is within the range of N ± 1/3 (rounded down), N/3 players (rounded up) are shuffled by swapping players between the playing and resting groups. This prevents the same person from alternating between playing and resting instead of taking consecutive breaks.
  • Gender-combination constraints such as “Mixed” or “Balanced,” participant statuses set to “must play,” players who rested or finished in the previous round in a “one court at a time” draw, and exclusive pairs take precedence over priority selection.

Data transfer procedure

  1. On the old device, select “Save to cloud” on the Backup screen.
  2. On the new device, log in with the same account and select “Load from cloud” on the Backup screen.
  3. The data on the Player, Match, and Register screens will be transferred to the new device.
  • If both devices are running Android 6.0 or later, Android's automatic backup feature is available. If automatic backup is enabled in the settings, the data will be restored on the new device. However, backups cannot be created at a precisely specified time. Under Android's backup behavior, a backup should be created when more than 24 hours have passed since the previous backup and the device is idle and connected to Wi-Fi.

To transfer history saved under “Local Save”

  1. On the old device's Backup screen, use “Output to external storage” for the desired history item.
  2. On the new device, use a file manager to share the exported file with this application.
  3. Confirm that the history item appears on the new device's Backup screen.

This section explains how to divide players into two groups for a doubles competition.

  1. Enable and configure group competition in Combination Settings.
    [diagram]
  2. Assign each player to one of the two competing groups. Each group must have enough players to fill its side of the court; two pairs require at least four players.
    [diagram]
  3. Tap the draw button to generate the match table.
    [diagram]

This section describes restrictions when using Fire OS (Amazon) or iOS (Apple) instead of standard Android.

  • This application can be installed on Fire OS from the Google Play Store. (Please check the installation procedure yourself.) However, the following functional differences apply.
    • On Fire OS, player devices cannot receive automatic update notifications from the host. (Manual synchronization using the toolbar icon is required.)
    • The browser-based Web Share view does not support automatic update notifications. (Manual refresh is required.)
    • If the host is on Fire OS, it cannot receive match results submitted from player devices.
    • The read-aloud feature may require installing Google's Speech Recognition & Synthesis app.
  • The app cannot be installed on iOS, but the browser-based Web Share feature is available. However, iOS only allows native apps to receive message notifications, resulting in the following functional difference.
    • On iOS, the Web Share view cannot receive automatic update notifications. (Manual refresh is required).
  • Even on some Android-like systems, incompatibilities or communication restrictions may occur.
    • The notification system relies on Firebase Cloud Messaging. Other services, such as Amazon Device Messaging (Fire OS), Apple Push Notification Service (iOS), and Huawei Mobile Services, are not compatible, so notifications will not work on those platforms.
    • Some features may be restricted on devices without Google Play Services or in regions where Google services are blocked.

You can merge match results from locally saved backup files, combining the results of multiple events into a single event. The following is an example of a merged result.

[diagram]