Monday, April 19, 2010

Leader Board System

Klokkhammer asked for some more details regarding the leader board. Here you are :)

I wanted a system that took into account the number of players in each game, the number og games that each player played and your finish place. There are several ways to do this but the main two are as follows:

1- The way we have done it thus far. If there are five players in a tourney, then 1st place gets 5 points and 5th place gets 1 point. Anyone that didn't play gets 0 points. This seems to cover all the bases pretty well. It rewards being a regular player and it rewards consistantly placing high. However, it does not reward playing to win. I have won 3 out of 4 tourneys, but was the first to bust out in the only one I didn't win. While MrSmith has only won 1 tourney but placed 2nd in the other 3. MrSmith has been more consistant and is thus at the top of the leader board in front of me.

2- Some systems give the tourney winner an extra point as a bonus for winning. If we used this system, then I would get 3 extra points for my three wins and MrSmith would get 1 extra point for his one win. Thus I would be leading him by 1 point. If we look at ROI then we see that only MrEMC2, MrSmith and myself have a positive ROI. I am in the lead with 127 %, MrSmith has 93% and MrEMC2 has 14%. (MrE has taken 3rd in all 4 tourneys however two have had payouts less than the buy in due to few players. So his roi should be higher). Everyone else has a negative ROI at this point. ROI might not be fair considering we play so few games.

I don't have any preferance as to what system we use. Come with something completely different if you have an idea. My main goal is just to get more people to play regularly. The lack of players has the largest impact on our statistics. Notice that the three leaders are the only three that have played every game... Furthermore, we are not even half way through the year. Anyone, even an unknown newcomer could still win the title. It also helps to pay Pstars their annual luck box fee. This comes in handy when you need to hit a 4 outer on the river like I did vs MrE last night :)

Here are the detailed results thus far. Im not posting everyone's ROI, but you can calculate it yourself if you really want to.

Any and all suggestions are welcome :)


  1. Thanks Roland,

    I think that it is more than sufficent to have an indication of the points and the number of games played, an example would be:

    Klokkhammer 8 (3)
    Wollmar 67 6 (2)

    This information would give a clear indication that Wollmar probable have done better than me. Without any reference to the number of games played it would be read the opposite way.

    Agree that the goal should be to have as many to play as often as possible. Simply publishing the points would be an incentive to play as much as possible.

    However, the longer the year goes it could also have the opposite effect. The distance would be too big. I expect it to increase more due to both the quality and the quantity of the games played, but it would be better if it was more connected to quality:)

    There is no way we can hide the fact that 3 of the players have done much better than the rest, and believe me - it's not my intention.

  2. Good stuff Roland! Suggestion; if we play 12+ games during the "season" maybe only the best 10 results should count? Then we could strike the 2+ worst results (or strike "no shows"). That way is both rewards regular players + allows a few "no shows" due to whatever reason.

  3. thanks for the feedback. Including games played as klokk suggests is easy. I'll do this for all future posts.

    MrSmith's idea is also easy to do at the end of the year but with so few games it can have some adverse effects. For example all the players who have made every game would essentially be losing points. We could drop the lowest game or two if we averaged the points, but that brings in some other mathematical issues. On the other hand, I do see you point.

    I can see if I can come up with a better model. Before I do that I have a few questions though.

    1. Do you dislike the current system?
    2. What are we really trying to measure? In my post I stated 3 variables: nr of players, games played and finish place. I could try to weight them differently or add in a new one if you have a suggestion.

    I have to take the kids to soccer practice now, but will give this some thought.

    Thanks for the input! Roland

  4. Thanks Roland.

    I like the current system (calculation of points). Just think its better to put a bit more emphasis on the quality by at least indicate the number of games played.

    MrSmiths proposal is interesting as it will maintain more uncertainty, but still secure that quality will prevail. I guess most of us are risking a couple of no shows, due to colliding commitments during the year. If this is the case, then MrSmiths proposal will even be a better indicator of strength than the one we have today. This because it does have a more realistic approach to the number of games played by all.

    Benko has two "no shows" all ready, and he will efficiently be out of the competition by the summer if this trend continues. On the other hand benkogambit needs to show up more consistently, and start winning - so do I :)

    I leave it to you to figure out the best approach :).