Friday, December 2, 2011

Cribbage Pro Multiplayer Points System Revised

After a lot of analysis, quite a few good debates on skill versus luck, and a lot of feedback from many of the regular players of Cribbage Pro Online, we are happy to announce a completely new points system for multiplayer games that is now live effective immediately.  This new points system is a lot different compared to the old system, and as such probably needs a little explaining so that everyone can have confidence in what to expect when they play.

At the heart of the new system is points being given out based on how dramatically you win against your opponent.  Points are awarded based off of the cribbage board position, and more specifically based on the "block" or 5 point section that the losers peg is in relative to the winning player (which on a game played to completion is always the last block with the peg out hole of course).  We calculate the difference between that winning peg and the losers peg and determine how many "blocks" or sections on the cribbage board that peg is away from the winning peg.  Next time you play, pay attention to which score block you are in, as each 5 point score block section will change the points awarded or taken away, so every time you get a block ahead of your opponent, you stand to get more points awarded for winning.  If you are losing, then trying to get to that next block before the game is over could save you a lot of points as well.

Before we get into specifics, another big change is that the same number of points you can win is the same that is lost by the other player.  With that, pictured in the table and chart below is the breakdown of points awarded by that "block" mentioned above (there are 24 blocks plus a final hole for 121 on our game board).


You will see that special consideration is given to the "Double Skunk", "Skunk" and if you lose by only one point (designated as block "25").

Along with this new system for determining the points to be awarded when you win, or that are taken away when you lose, is a new "Maximum Points Possible" calculation.  This is a bit more complicated, but the picture you should probably have in your mind is that of a bucket of points that each unique opponent you play has to share points with you.  This maximum is based on each match-up you have, each opponent you play against, and is calculated based on two different systems.

The first maximum score limiting system is used when you have played 10 games or less against the same opponent.  For those first 10 games you play against someone, when you win we determine what I'll call a "game win delta" or a maximum number of games you have won more than the other player - your "play advantage" if you will.  So if you have won 7 games against them in the past, and they have only won 2 (a total of 9 games), your advantage is 5 games (losers games of 2 won subtracted from winners 7).  The breakdown of that maximum points possible in your match-up is as shown below.


So, if you are really beating up your opponent game after game, there is now a diminishing return on your wins.  What this really is doing is gauging your skill relative to that specific opponent and awarding you less if you are already the most likely winner and it will subsequently award you more if you were being trounced before but finally manage to get a win (as your delta or advantage will be negative).

That system of maximum points for 10 or less games will be the one that most people will encounter as they play because of the numerous people playing and possible opponents you will face.  However, for those that play more than 10 games against the same opponent, there needs to be consideration given beyond just a "games advantage", and this is where a percentage based system becomes more relevant and that is what is used.

Here is how the percentage system is used to calculate your maximum points possible.


As you can see, we calculate the win % of the person who lost the game (prior to adding in that loss) and use that to determine a maximum number of points won or lost.  If the loser wins 46% or more against the winner, then all the points are available, but from 45% on down that maximum will slowly drop all the way down to 5 points maximum if the winner wins nearly all the time (85% or more of the time to be exact).  So again, if you are always winning against that person, then it is expected that you will continue to win and there is a diminishing return on that win as you continue to prove yourself superior.

UPDATE: Also note that after the first 10 games you play against someone, all "skunks" will count as 2 games and any "double skunks" will count as 3 games when calculating this maximum points possible (not in your game stats or anywhere else, only for this purpose).

I'm sure there are those that will ask why we did all this, and why it is so complicated.  Well, we really didn't want to make it this complicated, but we were forced into it by simply examining the system we use today and coming to an understanding where it was simply not being fair to everyone.  Too many people were forced to play only certain levels of other players or risk being docked huge amounts of points for losing.  Similarly, someone who only plays their spouse may be able to accumulate points rapidly if they were a cribbage expert and the spouse not so much.

Is this new system perfect?  No, I highly doubt it is perfect, but I do believe it is a good step in the right direction.  What that means is that this will probably not be the last change you see to the points system, but hopefully it means that future changes will just be tweaks here and there.

If you have comments, thoughts or suggestions on this new system please let us know in the comments or email us at support@fullersystems.com


8 comments:

  1. Although I have not yet seen the change 'in action', I am loudly applauding the effort and the amount of consideration that has gone into this -- thank you.

    ReplyDelete
  2. Thanks "dph", I really appreciate your comments. I think you will like the changes and improvements too.

    ReplyDelete
  3. Josh, awesome work, buddy. The new scoring system is awesome. I will continue to purchase future apps you make.

    ReplyDelete
  4. Hi Dom, thanks for the comments and support.

    ReplyDelete
  5. The section about "The first maximum score limiting system" confuses me.

    You say that it "is used when you have played 10
    games or less against the same opponent".
    Yet in the example you talk about having won 10 games and having lost 5 games against the opponent.

    Is it a badly chosen example or is it in fact used also after the first 10 games in parallel with the second score limiting system?

    ReplyDelete
  6. You are right, that was a miss on my part.  We had originally set that "games played" limit at 20 instead of 10 and I did not update that example to match.  I have just updated it now with some extra clarification and fixed the example to add up correctly.

    Thanks,
    Josh

    ReplyDelete
  7. So, why doesn't the skunk/double-skunk rule apply to wins (best two of three, three of five, etc)?

    ReplyDelete
    Replies
    1. Hi "GillRigged", when you play a "best of" series, no games are won or lost individually but only all together. You will notice that if you play a "best of" match, only 1 record is set in your stats for that game. You are only risking your points and stats against the entire "best of" set of games, so no stats are tracked, no points won or lost, or anything else until the series is fully completed and one single entry recorded in your stats as the win or loss.

      As an example, say you were skunked in the first game of a "best of 3", then you skunk your opponent in the second game and finally lose the series by losing the final game - calculating any meaningful "skunk" of any kind from that is impossible (again, all games are not recorded, just the result of the overall series). So there really is no "skunk" concept for a game played in a "best of" series. The only way to record skunks would be to record the results of each game, and then that would defeat the purpose of having a "best of" in the first place.

      Delete