Before you make that meteoric rise to Admiralship there are a few things you need to know about the Star Chamber rating system.

In Star Chamber all players start out with 4 different ratings. Overall, Constructed, Sealed and Multiplayer. Each new player begins with an across the board rating of 1500. Between the rating of 1500 and 1600, Star Chamber players can only go up in rating. That is losing a game will not drop your rating. Once you reach a rating of 1600 you can never go below that rating again. Each rating is calculated independently based on the corresponding rating of an opponent.

The Overall Rating is not an average of the other ratings but goes up or down with each rated match whether it is Constructed, Sealed or Multiplayer. This is the reason why you can often see a large gap between the Overall Rating and the other ratings.

Example

A player in a Constructed tournament will have their Overall Rating calculated based on their opponents Overall Rating. They will also have their Constructed Rating calculated based upon their opponents Constructed Rating. Because players will often have different ratings for Overall and Constructed, especially at higher ranks, the values for calculating each new rating will be different. That means a players Overall Rating and their Constructed Rating can go up or down by different amounts.

Star Chamber ratings are based on the Elo system. Elo is the Hungarian Chess Brain who came up with the math for calculating ratings based on statistics. The Elo system was developed for chess player rankings but can be applied to any two player or multiplayer game. Here are the basics of the Elo system.

Player A has a Rating (RA) and Player B has a Rating (RB). To get Player A's Expected Win Percentage (EA) vs Player B you would use the following formula.

I don't even own a calculator that can process that formula but luckily someone did and put it all in a chart which allows you to get the Expected Win Percentage (P) based on the Difference (D) of the players ratings.

Example 1 - Kit vs AlienProbe in a Constructed Match

Kit's Constructed Rating is 1804 and AlienProbe's Constructed Rating is 1663.

To find Kit's Expected Win Percentage (P) vs AlienProbe, you take the Difference (D) in their ratings (1804 - 1663 =141). Find the D value of 141 in the chart which gives you Kits Expected Win Percentage (P) of .69 or 69% of the time vs AlienProbe.

To find Alienprobe's Expected Win Percentage (P), you take the Difference (D) in their ratings (1663-1804 = -141). Find the D value of -141 in the chart which gives you AlienProbe's Expected Win Percentage (P) of .31 or 31% of the time vs Kit.

Note that .69 + .31 = 1.0.

Now that we know how to find the Expected Win Percentage we can put it to good use.

The formula below shows how we can find a players New Rating (Rn). This is done by adding the Old Rating (Ro) to the difference of Actual Wins (W) and the Expected Wins (We) adjusted by the K factor.

The Expected Wins (We) is the same value as the Expected Wins Percentage (P) found in the chart above.

The K factor is a constant that varies from tourney to tourney. Most K factors are around 16 for top level tourneys and 32 for lower level tourneys.

Star Chamber uses a K factor of 16 for standard tourneys and 32 for the Galaxy Cup events.

Example 2

Kit defeats AlienProbe in constructed tourney match. Kit's Old Rating (Ro) is 1804. AlienProbe's Old Rating (Ro) is 1663.

We already have the Expected Win Ratios (We) from the example above. Kit's Expected Win Ratio (We) is .69. AlienProbe's Expected Win Ratio (We) is .31.

Because Kit won the match his Actual Win (W) is 1 and AlienProbe who lost has an Actual Win (W) of 0.

We will use the K factor of 16 because it is a standard tourney.

To figure Kit's new rating we put all of the factors in the formula Rn = Ro + K (W-We). Rounding will be necessary in these calculations.

Rn = 1804 + 16 (1 - .69) = 1809

Kit is now rated at 1809, a jump of 5 points.

Now we do the same for AlienProbe using his numbers.

Rn = 1663 + 16 (0 - .31) = 1658

AlienProbe's new rating is 1658, a loss of 5 points.

Example 3

Lets say that AlienProbe had defeated Kit in the match.

We would calculate Kit's new rating with 0 for the Actual Win (W).

Rn = 1804 + 16 (0 - .69) = 1793

Kit's new rating would be 1793. That's a drop of 11 points.

Using the formula with a win for AlienProbe we get a new rating of 1674 with a gain of 11 points.

This is a perfect example of how the system is based on statistics, not rewards. It goes by what is expected of a player not necessarily by whether they win or lose.

Players with high ratings playing opponents with lesser ratings have less to gain and more to lose. Conversely, players who are rated lower, have less to lose and more to gain by defeating higher rated opponents.

I have played tourneys where I have won 2 out of 3 matches and went down in rating. This happens when you play opponents of a lesser rating and lose.

Example 4

Kit who has a a Constructed Rating of 1804 plays against 3 opponents who all have a Constructed Rating of 1500. Kit wins the first 2 matches and loses the last one.

In standard chess tournaments they total all of the Wins Expected (We) and the Actual Wins (W) then plug the results into the formula. So in this tournament Kit's New Rating (Rn) would be calculated like this. Rn = Ro + 16 ( (W1 + W2 + W3) - (We1 + We2 + We3)).

In Star Chamber the ratings are calculated after each match with each new rating dependent upon the previous new rating that the players acquire.

Match 1 The Difference (D) in Kit's rating of 1804 and the first players rating of 1500 is 304. The Wins Expected (We) is .85.

Kit's New Rating (Rn) = 1804 + 16 (1 - .85) = 1806.

Match 2 The Difference (D) in Kit's rating of 1806 and the second players rating of 1500 is 306. The Wins Expected (We) is still .85.

Kit's New Rating (Rn) = 1806 + 16 (1 - .85) = 1808.

Match 3 The Difference (D) in Kit's rating of 1808 and the final players rating of 1500 is 308. The Wins Expected (We) is still .85.

Kit's New Rating (Rn) = 1808 = 16 ( 0 - .85) = 1794 due to the loss.

See how the one loss to a lesser rated opponent made his rating drop significantly despite the fact that he won most of his matches.

On the flip side of the coin, a 1500 rated player would go up in rating if he defeated only 1 of 3 players rated 1804.

Now that you know how to calculate your ratings you can embark upon the road to Admiralship. Along the way you will have many ups and down. But that's okay, it's expected. After all we are just statistics.

AlienProbe's formula for successful ratings: Win the games you are expected to win. Lose only to higher rated players and defeat those Fleet Captains whenever possible.

GL & HF - AlienProbe