[Premier Diplomats] [1st Division Diplomats]
[2nd Division Diplomats] [3rd Division Diplomats] [Diplomat Climbers]
[Comments] [Explanations] [Past Years] [Hall of Fame]
Back to the Stats Home Page
Back to the Dip 2000 Home Page
|
October 31st 2008 Proposed System The new Eagles contribution will be used for any games calculated from January 2009 onwards. The plans are fairly well advanced – but I will still take any feedback and see if it can be used within the proposed structure. Thanks to everyone (public and private) who contributed to my initial thoughts on the Eagle points system. I have answered all the points raised, I hope. Thanks also for the positive words of encouragement. Many of you expressed a view that non- regular games could be as tough as regular games. I still disagree but I have taken your points on board and I will treat non-regular games and regular games the same way, so if 7 Premier players sign up for a non-regular game, it will have the same Eagle weighting as a regular game. However, overall, I still don’t see how games containing new players can be as difficult as those containing experienced player. I know some good players start in non-regular games but from a statistical point of view, I can’t measure without data. However, it is a statistical fact that regular players, overall, score better than non-regular players. While there may be individual discrepancies, the Eagle system is a good pointer not only to a player’s reliability but to communication, tactical ability and to strategic ability in diplomacy. I also agree that there is no guarantee of an easy game in a non-regular game, however I think it is more probable statistically. The new proposals will keep the things simple. No changes will be made retrospectively and the spirit of Karl’s original idea will remain. This means that any previous scores will not be affected and the principle of an Eagle score between 0 and 14 will remain. I believe the reward for a solo, in terms of Eagle points, is sufficient as it is. Those who are driven by other factors would not change their habits for more points for a solo and those who try to maximise their Eagles will see the current solo rewards sufficient. Those who play for draws will continue to do so. The aim of the Eagles is to attempt to measure results, not influence how they are achieved. I regard the Eagle points system as a good guide but I do not go out of my way to collect Eagles as such – I just try to get the best result in each game I play – I believe this is the view of the majority of people who play on this site. I know there are some who treat Eagles as the most important measure. That is their choice and I believe are a small minority. The idea is to have a more granular approach. It is not to trying to say a player at position 35 is significantly better than a player at position 36. They will in fact be almost equal. The discrepancy I am trying correct is to make a player at position 50 the same or similar to position 51 whereas there is a drop of 0.25 at the moment. Similarly - at positions100 and 101, 200 and 201, 300 and 301. The site continues to grow at a steady rate; we have had a net increase of over 50 since January. The Eagles will never be a perfect representation of any player’s skills and will remain a rough guide and a source of enjoyment for some (or many I hope). I come from a chess background and chess has an internationally used grading system called ELO. It is based on statistical results and probabilities and is a good to accurate representation of a chess players past performances. I have no intention of trying to produce an equivalent system. I need to dispel one point that somehow Eagles create some sort of self-perpetuating system. This is very misleading. The truth is that if you get to a high level after 10 games, you need to work hard to maintain that level. It would be ludicrous to take everyone else’s points into account except your own. The only way to self-perpetuate is to play consistently well as most of the top players do. There are simpler methods of measuring success and failure such as ten points for a solo and dividing any draws into ten to get the points. In this system there is no attempt to measure how difficult or easy the games are – but maybe this would even out over time. There are much more complicated options available such as trying to develop statistical inference data modelling packages or the complicated merits system on the stabbeurfou site. All will have their plusses and minuses but I do not intend to get involved with them. Other proposals for a medal based system (gold for solo, silver 2way draw and bronze for 3way – nothing for the rest) have their merits and are already in place an up to date on the site. [Medal Table] John Wright has produced an excellent game tracking page where the difficulty indicators for each game are the total Eagle Points (EPs) and the Total Experience for each game. As you would expect, there is a strong correlation between the value of a game in EAGLEs and the value of these indicators. [Game Stats] The EAGLE system will continue on the Dip2k site until such times as it no longer wanted. The spirit of the original will be maintained. I am proposing to make the following changes: 1. Move half of the game value into the player contribution (goes from 1.0
to 0.5). The calculation I will use is: Eagle contribution = 0.5 + (601-P)/400 (rounded to 2 decimals, note
0.005 rounds up) Where 0 < P <= 600 Eagle contribution = 0.5 P=Eagle table position when the calculation done. How will this affect the Eagles in a game? A complete comparison is available in the Eagle tables. A recent game, William, was completed with one Anarchy scored 7 under the current system and 5.88 under the proposed system. Example of a recent game – William
The lower eagles score reflects the fact some out of the 7 are not registered and are statistically weaker even though they may be good players. All feedback welcome. |
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||