1 <?xml version="1.0" encoding="UTF-8" ?>
3 html PUBLIC "-//W3C//DTD XHTML 1.1//EN"
4 "http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">
5 <html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" xmlns:t="http://template.sesse.net/">
7 <title>Foosball!</title>
8 <link rev="made" href="mailto:sgunderson@bigfoot.com" />
9 <meta name="MSSmartTagsPreventParsing" content="TRUE" />
14 <h2>Add a singles result</h2>
16 <form method="post" action="add-single-result.pl">
20 <td><input name="username1" value="" size="10" /></td>
24 <td><input name="username2" value="" size="10" /></td>
29 <input name="score1" value="" size="2" /> -
30 <input name="score2" value="" size="2" />
34 <td colspan="2"><input type="submit" /></td>
39 <h2>Add a doubles result</h2>
41 <form method="post" action="add-double-result.pl">
44 <th>Team 1 (usernames)</th>
46 <input name="team1_username1" value="" size="10" /> and
47 <input name="team1_username2" value="" size="10" />
51 <th>Team 2 (usernames)</th>
53 <input name="team2_username1" value="" size="10" /> and
54 <input name="team2_username2" value="" size="10" />
60 <input name="score1" value="" size="2" /> -
61 <input name="score2" value="" size="2" />
65 <td colspan="2"><input type="submit" /></td>
70 <h2>Singles score board</h2>
77 <th>Rating deviation</th>
78 <th>Conservative estimate</th>
79 <th>Change since yesterday</th>
82 <tbody t:id="singletop">
84 <td><t:username /></td>
87 <td><t:lowerbound /></td>
93 <h2>Doubles score board</h2>
100 <th>Rating deviation</th>
101 <th>Conservative estimate</th>
102 <th>Change since yesterday</th>
105 <tbody t:id="doubletop">
107 <td><t:username /></td>
108 <td><t:rating /></td>
110 <td><t:lowerbound /></td>
116 <h2>Last ten games</h2>
123 <th colspan="2">Opponents</th>
127 <tbody t:id="lastgames">
129 <td><t:gametime /></td>
131 <td><t:username1 /></td>
132 <td><t:username2 /></td>
133 <td><t:score1 /> – <t:score2 /></td>
138 <h2>About the ratings</h2>
140 <p>The rating system in use is called <em>FoosRank</em>, and is specially
141 designed for this system. It is a Bayesian rating system, inspired by
142 Glicko 1 but re-worked out from scratch with a statistically sound model
143 of non-binary results, and adjusted for teams (with some ideas from
144 Microsoft's TrueSkill system). For those not familiar with Bayesian
145 ratings, the most important parts are:</p>
148 <li>Your rating is a <em>statistical estimation</em> of your true skill.
149 It has a mean (the point estimate of your skill) and a deviation
150 (measuring the uncertainity of the estimate), called the RD. It is approximately
151 Gaussian (actually logistic).</li>
152 <li>When you win or lose a game, your rating will change accordingly,
153 based on your score and your opponent. <em>You do not get 'points'
154 for winning or losing, the estimate is merely getting more accurate.</em>
155 In the process, the RD gets lower as you play. However, the RD increases
156 with time, opening up for the fact that your true skill can change.
157 (Glicko 2 also supports a volatility measure, which better models change
158 in true skill, but it has not been implemented here.)</li>
159 <li>The score board is sorted by a conservative estimate of your rating
160 (mean - 3 * RD). This makes it non-attractive for people with artifically
161 high ratings (especially newcomers) to avoid playing to stay high up in
162 the score board.</li>
163 <li>The single and double rankings are separate. Even though you play as a
164 team and all four players' rankings and RDs influence the rating adjustment,
165 you are ranked as an individual, as we do not usually play with fixed